Difference: LatestNews (1 vs. 81)

Revision 812011-08-17 - ChrisWelti

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Windows 7 / Vista / 2008 TCP/IP tweaks

On Speedguide.net there is a great article about the TCP/IP Stack tuning possibilities in Windows 7/Vista/2008.

 

NDT Protocol Documented

A description of the protocol used by the NDT tool has recently been written up. Check this out if you want to implement a new client (or server), or if you simply want to understand the mechanisms.

Revision 802011-08-02 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

NDT Protocol Documented

A description of the protocol used by the NDT tool has recently been written up. Check this out if you want to implement a new client (or server), or if you simply want to understand the mechanisms.

-- SimonLeinen - 02 Aug 2011

 

What does the "+1" mean?

I would like to experiment with social web techniques, so I have added a (Google) "+1" button to the template for viewing PERT KB topics. It allows people to publicly share interesting Web pages. Please share what you think about this via the SocialWebExperiment topic!

Revision 792011-07-02 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

What does the "+1" mean

>
>

What does the "+1" mean?

  I would like to experiment with social web techniques, so I have added a (Google) "+1" button to the template for viewing PERT KB topics. It allows people to publicly share interesting Web pages. Please share what you think about this via the SocialWebExperiment topic!

Revision 782011-07-02 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

What does the "+1" mean

I would like to experiment with social web techniques, so I have added a (Google) "+1" button to the template for viewing PERT KB topics. It allows people to publicly share interesting Web pages. Please share what you think about this via the SocialWebExperiment topic!

-- SimonLeinen - 02 Jul 2011

 

News from Web10G

The Web10G project has released its first code in May 2011. If you are interested in this followup project of the incredibly useful web100, check out their site. I have added a short summary on the project to the WebTenG topic here.

Revision 772011-06-30 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

News from Web10G

The Web10G project has released its first code in May 2011. If you are interested in this followup project of the incredibly useful web100, check out their site. I have added a short summary on the project to the WebTenG topic here.

-- SimonLeinen - 30 Jun 2011

 

FCC Open Internet Apps Challenge - Vote for Measurement Tools by 15 July

The U.S. Federal Communications Commission (FCC) is currently holding a challenge for applications that can help assess the "openness" of Internet access offerings. There's a page on challenge.gov where people can vote for their favorite tool. A couple of candidate applications are familiar to our community, for example NDT and NPAD. But there are some new ones that look interesting. Some applications are specific for mobile data networks ("3G" etc.). There are eight submissions - be sure to check them out. Voting is possible until 15 July 2011.

Revision 762011-06-29 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

FCC Open Internet Apps Challenge - Vote for Measurement Tools by 15 July

The U.S. Federal Communications Commission (FCC) is currently holding a challenge for applications that can help assess the "openness" of Internet access offerings. There's a page on challenge.gov where people can vote for their favorite tool. A couple of candidate applications are familiar to our community, for example NDT and NPAD. But there are some new ones that look interesting. Some applications are specific for mobile data networks ("3G" etc.). There are eight submissions - be sure to check them out. Voting is possible until 15 July 2011.

-- SimonLeinen - 29 Jun 2011

 

Web10G

Catching up with the discussion@web100.org mailing list, I noticed an announcement by Matt Mathis (formerly PSC, now at Google) from last September (yes, I'm a bit behind with my mail reading :-): A follow-on project called Web10G has received funding from the NSF. I'm all excited about this because Web100 produced so many influential results. So congratulations to the proposers, good luck for the new project, and let's see what Web10G will bring!

Revision 752011-03-25 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Web10G

Catching up with the discussion@web100.org mailing list, I noticed an announcement by Matt Mathis (formerly PSC, now at Google) from last September (yes, I'm a bit behind with my mail reading :-): A follow-on project called Web10G has received funding from the NSF. I'm all excited about this because Web100 produced so many influential results. So congratulations to the proposers, good luck for the new project, and let's see what Web10G will bring!

-- SimonLeinen - 25 Mar 2011

 

eduPERT Training Workshop in Zurich, 18/19 November 2010

We held another successful training event in November, with 17 attendees from various NRENs including some countries that have joined GÉANT recently, or are in the process of doing so. The program was based on the one used for the workshops held in 2008 and 2007. But this time we added some case studies, where the instructors walked through some actual cases they had encountered in their PERT work, and at many points asked the audience for their interpretation of the symptoms seen so far, and for suggestions on what to do next. Thanks to the active participation of several attendees, this interactive part turned out to be both enjoyable and instructive, and was well received by the audience. Program and slides are available on the PERT Training Workshop page at TERENA.

Revision 742010-12-18 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

eduPERT Training Workshop in Zurich, 18/19 November 2010

>
>

eduPERT Training Workshop in Zurich, 18/19 November 2010

 
Changed:
<
<
We held another successful training event in November, with 17 attendees from various NRENs including some countries that have joined GÉANT recently, or are in the process of doing so. The program was based on the one used for the workshops held in 2008 and 2007. But this time we added some case studies, where the instructors walked through some actual cases they had encountered in their PERT work, and at many points asked the audience for their interpretation of the symptoms seen so far, and for suggestions on what to do next. Thanks to the active participation of several attendees, this interactive part turned out to be both enjoyable and instructive, and was well received by the audience. Program and slices are available on the PERT Training Workshop page at TERENA.
>
>
We held another successful training event in November, with 17 attendees from various NRENs including some countries that have joined GÉANT recently, or are in the process of doing so. The program was based on the one used for the workshops held in 2008 and 2007. But this time we added some case studies, where the instructors walked through some actual cases they had encountered in their PERT work, and at many points asked the audience for their interpretation of the symptoms seen so far, and for suggestions on what to do next. Thanks to the active participation of several attendees, this interactive part turned out to be both enjoyable and instructive, and was well received by the audience. Program and slides are available on the PERT Training Workshop page at TERENA.
  -- SimonLeinen - 18 Dec 2010

Revision 732010-12-18 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

eduPERT Training Workshop in Zurich, 18/19 November 2010

We held another successful training event in November, with 17 attendees from various NRENs including some countries that have joined GÉANT recently, or are in the process of doing so. The program was based on the one used for the workshops held in 2008 and 2007. But this time we added some case studies, where the instructors walked through some actual cases they had encountered in their PERT work, and at many points asked the audience for their interpretation of the symptoms seen so far, and for suggestions on what to do next. Thanks to the active participation of several attendees, this interactive part turned out to be both enjoyable and instructive, and was well received by the audience. Program and slices are available on the PERT Training Workshop page at TERENA.

-- SimonLeinen - 18 Dec 2010

 

Efforts to increase TCP's initial congestion window (updated)

At the 78th IETF meeting in Maastricht this July, one of the many topics discussed there was a proposal to further increase TCP's initial congestion window, possibly to something as big as ten maximum-size segments (10 MSS). This caused me to notice that the PERT KB was missing a topic on TCP's initial congestion window, so I created one. A presentation by Matt Mathis on the topic is scheduled for Friday's ICCRG meeting. The slides, as well as some other materials and minutes, can be found on the IETF Proceedings page. Edited to add: Around IETF 79 in November 2010, the discussion on the initial window picked up steam, and there are now three different proposals in the form of Internet-Drafts. I have updated the TcpInitialWindow topic to briefly describe these.

Line: 26 to 32
  -- ChrisWelti - 18 Jan 2010
Deleted:
<
<

FreeBSD 8.0 Released

Release 8.0 of the FreeBSD operating system was released on 24 November, 2009. Notable performance-related improvements in the networking area include:

  • a multi-threaded implementation of the netisr subsystem.
  • Linux-compatible TCP_CONGESTION socket option for selecting between pluggable congestion algorithms on a per-socket basis.

-- SimonLeinen - 30 Nov 2009

Pluggable TCP (and SCTP!) congestion control under discussion at OpenSolaris

On the networking discussion forum for OpenSolaris, a project proposal for a project on "Solaris Network Performance Projects" is currently under discussion:

http://www.opensolaris.org/jive/thread.jspa?messageID=435766

One of these projects is "pluggable congestion control". It foresees implementation of the HighSpeed, CUBIC, Westwood+, and Vegas congestion control algorithms, as well as ipadm subcommands and Linux-compatible socket options to get and set congestion control parameters. If successful, this will add OpenSolaris - and possibly at some point Solaris - to the group os OSes that support modern and configurable TCP congestion control (Linux, newer versions of Windows, newer versions of FreeBSD).

Updated 16 Dec 2009: An initial draft design document for this feature has been announced on the OpenSolaris networking-discuss forum.

-- SimonLeinen - 17 Nov 2009

http://fasterdata.es.net/

In a message to the NANOG mailing list, Kevin Oberman points to ESnet Network Performance Knowledge Base. It has TCP tuning tips, hints on choosing and configuring file transfer tools, and information about many other areas. There's obviously some overlap with the information here. One great advantage of the ESnet site is that the information is quite concise and presented in a very pragmatic way. The site focuses on one problem, namely helping users transfer large amounts of data over wide-area networks (see the FileTransfer topic here).

-- SimonLeinen - 09 Nov 2009

pert-discuss mailing list revival

Throughout the history of the PERT in the extended European research networking community, there has been a mailing list for discussions. This list is open for those who are interested in contributing to the PERT effort. With the transition from GN2 to the GN3 project, the home of the list (similar to other resources) has changed to a geant.net domain. In the process of this, archived messages since October 2004 have been made publicly available under http://mail.geant.net/public/pert-discuss/. The archive contains interesting discussions about specific and general performance issues. The list has been dormant for about a year, but we will try to revive it as the eduPERT activities in GN3 are gaining speed.

-- SimonLeinen - 29 Oct 2009

Mobile-optimized search page

This wiki uses a "custom search engine" provided by Google to implement its "Search" function. According to news items, Google now provide, for each such custom search engine, a search page that is optimized for "high-end" mobile devices with web browsers. So if you have an iPhone/iPod touch, Palm Pre, Android-powered device or similar, try the following link:

http://www.google.com/cse/home?cx=012443065831413718779:zokjgbpdcke

Optimizing Internet applications so that they work well with this kind of devices is an interesting general PERT problem, of course!

-- SimonLeinen - 28 Oct 2009

New domain names, new ToDo topic

The eduPERT activities in the GN3 project are spinning up. As a first step, many of the eduPERT-related domain names were moved from geant2.net to the geant.net namespace. This includes this PERT Knowledge Base, which now officially lives under http://kb.pert.geant.net/. Please update your pointers!

Then, I have added a ToDo topic, with some ideas on how the Knowledge Base could be improved. Please feel free to add to this list. But what would be really great is if people used the list as inspiration for contributions to the PERT Knowledge Base. Remember, this is supposed to be a community effort.

-- SimonLeinen - 21 Jul 2009

PERT/eduPERT talk at TNC 2009

I gave a talk about PERTs and the eduPERT at the Terena Networking Conference (TNC) 2009 in Málaga, Spain, last Thursday. Audio/video recording as well as presentation slides are available from the conference site. Here are the title and abstract:

eduPERT - helping European researchers bridge the "Wizard Gap"

PERTs (Performance Enhancement and Response Team) have been proposed to help network users achieve better end-to-end performance over today's fast research networks. Over the past few years, the PERT concept has been tested both within the GN2 project and within European NRENs with some success. This talk will explain "eduPERT", the current federated PERT model for the GÉANT community, talk about some interesting performance problems handled by PERTs, and about lessons learned and future work for advancing the PERT idea.

-- SimonLeinen - 14 Jun 2009

AWS (Amazon Web Services) Import/Export service

Amazon just announced a service (in beta) which allows you to send and retrieve your data from/to their cloud infrastructure through physical shipping of portable disk drives: http://aws.amazon.com/importexport/

The article has some interesting rules-of-thumb explaining when using this option is preferable to up/downloading your data through the network, depending on connection speed and data volume.

eduPERT sites (including this one) transitioned to geant.net

The GN2 (GÉANT2) project ended in April 2009. The eduPERT activity continues as part of the GN3 project. To reflect this change, various domain names were changed from geant2.net to geant.net (not geant3.net, so we won't have to do this again four years from now - clever, huh?). In particular,

edupert.geant2.net is now edupert.geant.net (main portal to eduPERT federation)
kb.pert.geant2.net is now kb.pert.geant.net (this site)
pert-registration@geant2.net is now pert-registration@geant.net (for registering information about PERTs)
pert-accreditation@geant2.net is now pert-accreditation@geant.net (to request accreditation of your PERT)

Please update pointers from other sites, your bookmarks, and address books. At least for the Web sites, the old addresses will still work and/or redirect to the new names, but it's best to start using the new version-independent names as soon as possible in order to avoid confusion.

One exception is the pert-discuss@geant2.net mailing list, which will have to be transitioned more carefully in the near future.

-- SimonLeinen - 10 Jun 2009

 

News Archive:

Added:
>
>
 

Revision 722010-12-09 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

Efforts to increase TCP's initial congestion window

>
>

Efforts to increase TCP's initial congestion window (updated)

 
Changed:
<
<
The 78th IETF meeting is taking place in Maastricht this week, and one of the many topics being discussed there is a proposal to further increase TCP's initial congestion window, possibly to something as big as ten maximum-size segments (10 MSS). This caused me to notice that the PERT KB was missing a topic on TCP's initial congestion window, so I created one. A presentation by Matt Mathis on the topic is scheduled for Friday's ICCRG meeting, so if you're in the office that day, take advantage of the convenient meeting location and tune in to the audio stream!
>
>
At the 78th IETF meeting in Maastricht this July, one of the many topics discussed there was a proposal to further increase TCP's initial congestion window, possibly to something as big as ten maximum-size segments (10 MSS). This caused me to notice that the PERT KB was missing a topic on TCP's initial congestion window, so I created one. A presentation by Matt Mathis on the topic is scheduled for Friday's ICCRG meeting. The slides, as well as some other materials and minutes, can be found on the IETF Proceedings page. Edited to add: Around IETF 79 in November 2010, the discussion on the initial window picked up steam, and there are now three different proposals in the form of Internet-Drafts. I have updated the TcpInitialWindow topic to briefly describe these.
 
Changed:
<
<
-- SimonLeinen - 28 Jul 2010
>
>
-- SimonLeinen - 28 Jul 2010 - 09 Dec 2010
 

PERT meeting at TNC 2010

Revision 712010-07-28 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

PERT meeting at TNC 2010

>
>

Efforts to increase TCP's initial congestion window

The 78th IETF meeting is taking place in Maastricht this week, and one of the many topics being discussed there is a proposal to further increase TCP's initial congestion window, possibly to something as big as ten maximum-size segments (10 MSS). This caused me to notice that the PERT KB was missing a topic on TCP's initial congestion window, so I created one. A presentation by Matt Mathis on the topic is scheduled for Friday's ICCRG meeting, so if you're in the office that day, take advantage of the convenient meeting location and tune in to the audio stream!

-- SimonLeinen - 28 Jul 2010

PERT meeting at TNC 2010

  We held a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the TNC 2010 conference in Vilnius. The attendees discussed technical and organizational issues facing PERTs in National Research and Education Networks (NRENs), large campus networks, and supporting network-intensive projects such as the multi-national research efforts in (Radio-) Astronomy, Particle Physics, or Grid/Cloud research. Slides are available on TNC2010 website here.

Revision 702010-07-01 - BartekGajda

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

PERT meeting at TNC 2010

Changed:
<
<
We held a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the TNC 2010 conference in Vilnius. The attendees discussed technical and organizational issues facing PERTs in National Research and Education Networks (NRENs), large campus networks, and supporting network-intensive projects such as the multi-national research efforts in (Radio-) Astronomy, Particle Physics, or Grid/Cloud research. Slides will be made available on TNC2010 website soon, and I'll make the pointer available here.
>
>
We held a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the TNC 2010 conference in Vilnius. The attendees discussed technical and organizational issues facing PERTs in National Research and Education Networks (NRENs), large campus networks, and supporting network-intensive projects such as the multi-national research efforts in (Radio-) Astronomy, Particle Physics, or Grid/Cloud research. Slides are available on TNC2010 website here.
  -- SimonLeinen - 15 Jun 2010

Revision 692010-06-24 - BartekGajda

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

PERT meeting at TNC 2010

Changed:
<
<
We held a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the TNC 2010 conference in Vilnius. The attendees discussed technical and organizational issues facing PERTs in National Research and Education Networks (NRENs), large campus networks, and supporting network-intensive projects such as the multi-national research efforts in (Radio-) Astronomy, Particle Physics, or Grid/Cloud research. Slides will be made available soon, and I'll make the pointer available here.
>
>
We held a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the TNC 2010 conference in Vilnius. The attendees discussed technical and organizational issues facing PERTs in National Research and Education Networks (NRENs), large campus networks, and supporting network-intensive projects such as the multi-national research efforts in (Radio-) Astronomy, Particle Physics, or Grid/Cloud research. Slides will be made available on TNC2010 website soon, and I'll make the pointer available here.
  -- SimonLeinen - 15 Jun 2010

Revision 682010-06-16 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

Upcoming PERT meeting at TNC 2010: Vilnius (Lithuania) 30 May

>
>

PERT meeting at TNC 2010

 
Changed:
<
<
Registration for the TERENA Networking Conference (TNC) 2010 is now open. When planing your trip, please consider the following:
>
>
We held a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the TNC 2010 conference in Vilnius. The attendees discussed technical and organizational issues facing PERTs in National Research and Education Networks (NRENs), large campus networks, and supporting network-intensive projects such as the multi-national research efforts in (Radio-) Astronomy, Particle Physics, or Grid/Cloud research. Slides will be made available soon, and I'll make the pointer available here.
 
Changed:
<
<
There will be a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the conference. So please try to be there in time to discuss network performance work.

We'll be interested in success/war stories from your performance work, as well as your opinions on future challenges and PERT practice. You will also hear about the eduPERT initiative in GN3, including ideas to raise the PERT's profile both globally and in your organisation.

If you have any suggestions on topics for the meeting, please post them to the pert-discuss mailing list or send them to Bartek Gajda and myself.

Hope to see you in Vilnius!

-- SimonLeinen - 07 Mar 2010 (link to PERT meeting corrected 31 Mar 2010)

>
>
-- SimonLeinen - 15 Jun 2010
 

Web100 Project Re-Opening Soon?

Revision 672010-03-31 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Line: 7 to 7
 Registration for the TERENA Networking Conference (TNC) 2010 is now open. When planing your trip, please consider the following:
Changed:
<
<
There will be a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the conference. So please try to be there in time to discuss network performance work.
>
>
There will be a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the conference. So please try to be there in time to discuss network performance work.
  We'll be interested in success/war stories from your performance work, as well as your opinions on future challenges and PERT practice. You will also hear about the eduPERT initiative in GN3, including ideas to raise the PERT's profile both globally and in your organisation.
Line: 16 to 16
  Hope to see you in Vilnius!
Changed:
<
<
-- SimonLeinen - 07 Mar 2010
>
>
-- SimonLeinen - 07 Mar 2010 (link to PERT meeting corrected 31 Mar 2010)
 

Web100 Project Re-Opening Soon?

Revision 662010-03-07 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Upcoming PERT meeting at TNC 2010: Vilnius (Lithuania) 30 May

Registration for the TERENA Networking Conference (TNC) 2010 is now open. When planing your trip, please consider the following:

There will be a PERT meeting on Sunday 30 May (1400-1730), just before the official start of the conference. So please try to be there in time to discuss network performance work.

We'll be interested in success/war stories from your performance work, as well as your opinions on future challenges and PERT practice. You will also hear about the eduPERT initiative in GN3, including ideas to raise the PERT's profile both globally and in your organisation.

If you have any suggestions on topics for the meeting, please post them to the pert-discuss mailing list or send them to Bartek Gajda and myself.

Hope to see you in Vilnius!

-- SimonLeinen - 07 Mar 2010

 

Web100 Project Re-Opening Soon?

In a mail to the ndt-users list, Matt Mathis has issued a call for input from people who use Web100, as part of NDT or otherwise. If you use this, please send Matt an e-mail. There's a possibility that the project will be re-opened "for additional development and some bug fixes".

Revision 652010-02-20 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Web100 Project Re-Opening Soon?

In a mail to the ndt-users list, Matt Mathis has issued a call for input from people who use Web100, as part of NDT or otherwise. If you use this, please send Matt an e-mail. There's a possibility that the project will be re-opened "for additional development and some bug fixes".

-- SimonLeinen - 20 Feb 2010

 

ICSI releases Netalyzr Internet connectivity tester

The International Computer Science Institute (ICSI) at Berkeley has released an awesome new java applet called Netalyzr which will let you do a whole bunch of connectivity tests. A very nice tool to get information about your Internet connectivity, to see if your provider blocks ports, your DNS works properly, and much more. Read about it on our page here or go directly to their homepage.

Revision 642010-01-18 - ChrisWelti

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

ICSI releases Netalyzr Internet connectivity tester

The International Computer Science Institute (ICSI) at Berkeley has released an awesome new java applet called Netalyzr which will let you do a whole bunch of connectivity tests. A very nice tool to get information about your Internet connectivity, to see if your provider blocks ports, your DNS works properly, and much more. Read about it on our page here or go directly to their homepage.

-- ChrisWelti - 18 Jan 2010

 

FreeBSD 8.0 Released

Release 8.0 of the FreeBSD operating system was released on 24 November, 2009. Notable performance-related improvements in the networking area include:

Revision 632009-12-16 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Line: 11 to 11
  -- SimonLeinen - 30 Nov 2009
Changed:
<
<

Pluggable TCP (and SCTP?) congestion control under discussion at OpenSolaris

>
>

Pluggable TCP (and SCTP!) congestion control under discussion at OpenSolaris

  On the networking discussion forum for OpenSolaris, a project proposal for a project on "Solaris Network Performance Projects" is currently under discussion:

http://www.opensolaris.org/jive/thread.jspa?messageID=435766

Changed:
<
<
One of these projects is "pluggable congestion control". It foresees implementation of the HighSpeed, CUBIC, Westwood+, and Vegas congestion control algorithms, as well as ipadm subcommands and socket options to get and set congestion control parameters. If successful, this will add OpenSolaris - and possibly at some point Solaris - to the group os OSes that support modern and configurable TCP congestion control (Linux, newer versions of Windows, newer versions of FreeBSD).
>
>
One of these projects is "pluggable congestion control". It foresees implementation of the HighSpeed, CUBIC, Westwood+, and Vegas congestion control algorithms, as well as ipadm subcommands and Linux-compatible socket options to get and set congestion control parameters. If successful, this will add OpenSolaris - and possibly at some point Solaris - to the group os OSes that support modern and configurable TCP congestion control (Linux, newer versions of Windows, newer versions of FreeBSD).

Updated 16 Dec 2009: An initial draft design document for this feature has been announced on the OpenSolaris networking-discuss forum.

  -- SimonLeinen - 17 Nov 2009

Revision 622009-11-30 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

FreeBSD 8.0 Released

Release 8.0 of the FreeBSD operating system was released on 24 November, 2009. Notable performance-related improvements in the networking area include:

  • a multi-threaded implementation of the netisr subsystem.
  • Linux-compatible TCP_CONGESTION socket option for selecting between pluggable congestion algorithms on a per-socket basis.

-- SimonLeinen - 30 Nov 2009

 

Pluggable TCP (and SCTP?) congestion control under discussion at OpenSolaris

On the networking discussion forum for OpenSolaris, a project proposal for a project on "Solaris Network Performance Projects" is currently under discussion:

Line: 73 to 82
  -- SimonLeinen - 10 Jun 2009
Deleted:
<
<

Recordings and other Materials from eduPERT Training Event now available

In late November 2008, the GN2 project held an eduPERT training workshop in Zurich, which was attended by 17 participants. Audio/video recordings, slide presentations, and workbooks with exercises can be found under the PertTraining2008 topic.

-- SimonLeinen - 19 Dec 2008

TERENA NREN Compendium 2008 released, shows growing interest in PERTs

The 2008 edition of the TERENA compendium was just released. TERENA surveys all European NRENs (National Research and Education Networks) yearly about their infrastructure, service offerings, and other activities. The "summary of key findings" section of the printed edition has the following to say about PERTs: "At the moment, 16 out of the 30 EU/EFTA countries have Performance Enhancement and Response Teams (PERTs). Four are planning to establish one within the next year and a further six plan to do this within the next three years. Many NRENs from other countries also either have a PERT or are planning to establish one." Progress with PERT establishment is also mentioned in a companion release, What the 2008 TERENA Compendium reveals. The summarized responses can be found on pages 63 and 64 of the printed edition. Under the individual NRENs' responses, PERT information can be found under the respective "Network Connectivity and Services" section.

-- SimonLeinen - 24 Oct 2008

PERT Training Event announced

We will be doing another PERT training event in Zurich on 27/28 November. See PertTraining2008 for more information, including a registration link.

-- SimonLeinen - 21 Sep 2008

TWiki software update

Today I upgraded the Debian twiki package on this host, in order to activate a security fix. This entry is mainly to test whether the wiki can still be written to.

-- SimonLeinen - 21 Sep 2008

eduPERT

As of 1 September 2008, the GEANT2 community's Federated PERT is known under the name eduPERT. It has a new Web site on http://edupert.geant2.net/.

-- SimonLeinen - 21 Sep 2008

More GridFTP information: new portal, tutorials, success stories

I found tutorial slides and handouts on the Web, under a new GridFTP portal. This should be quite useful to get people started with these tools. Pointers have been added to the GridFtpProtocol topic.

-- SimonLeinen - 28 Jul 2008

DTrace IP Provider integrated into OpenSolaris

(message from 7 Jul 2008, edited to add note about Nevada b93)

DTrace is a mechanism for system-wide instrumentation for performance analysis and debugging that was pioneered in Sun's Solaris operating system. It has since been added to other systems such as BSD (forgot which variant(s)) and Apple's Mac OS X. In the OpenSolaris community, there is a project to add network -specific probes to DTrace. I just learned that the DTrace IP Provider, which is outlined in this "onepager", was integrated into OpenSolaris Nevada four weeks ago. It was released yesterday as part of Nevada b93.

It is interesting to compare the DTrace approach with the web100 system. I think that there is a large overlap in the things that you can do with both, but the interfaces are quite different. Of course, web100 is very specific to TCP, while DTrace is a general tool for observing system behavior.

-- SimonLeinen - 14 Jul 2008

Linux 2.6.26

Another new Linux kernel version has been released today. Notable new features related to network performance are support for some 802.11n high-speed WLAN cards, and a performance improvement in CUBIC TCP. See the CUBIC topic for the patch, and for some measurements of CUBIC compared with other TCP variants.

-- SimonLeinen - 14 Jul 2008

Dilbert on Network Performance

-- SimonLeinen - 14 Jul 2008

New GridFTP supports SSH for its control channel

The new version 4.2 of the Globus Toolkit was released on July 2. One of its new features is that its GridFTP component can now use SSH for its control channel. This could make it much more attractive for users who just want to haul large amounts of data over LFNs, but who don't necessarily want to run the full Globus infrastructure. Anybody wants to give this a try?

-- SimonLeinen - 04 Jul 2008

ACM Queue: Kode Vicious articles about Latency and Livelock

ACM Queue is a relatively new publication of the Association for Computing Machinery (ACM) oriented to IT practitioners. It includes a section where Kode Vicious, the slightly psychotic alter ego of George V. Neville-Neil, responds to (real?) questions from readers. The column from the March/April 2008 issue is now available online, with two articles relevant for (network) performance: One talks about the often neglected impact of latency on performance even where bandwidth is plentiful, and the other explains a problem called "livelock", which can be seen when servers get persistently overloaded. At SWITCH, I believe we have seen a nice example of this with the RRD MA server that we host for PerfSONAR. Unfortunately, the article about latency has an annoying decimal-point error, which leads to some hand-waving about router-induced delays that is inappropriate - I posted a correction as a reader comment.

RFC 5236 (Reorder Density)

There is a new RFC with an alternative ("improved") metric for packet reordering. It doesn't seem to be an official product of the IPPM WG, and its status is informational (not standards-track), and it doesn't obsolete RFC 4737. I added the reference to the PacketReordering and IPPerformanceMetrics topics.

LSO vs. MDT

There is an interesting discussion on the OpenSolaris networking forum that is still ongoing. The original topic was whether OpenSolaris supported LSO for UDP (apparently it doesn't). The discussion quickly evolved into a debate of the relative merits of LSO - where the adapter knows how to segment transport data units and construct all headers for them - and MDT, which is like a software-only variant of LSO, where the host (operating system) still has to do the segmentation and header construction, but can send all the segments in a single (bus) transaction. I added a few sentences on this trade-off to the LargeSendOffloadLSO topic.

-- SimonLeinen - 04 Jun 2008

bwctl 1.3 release candidate

The first release candidate for a new version 1.3 of bwctl was released yesterday. Changes include more robust scheduling, improved error handling and logging, parallel streams (-P), CSV output, and support for several throughput measurement tools: thrulay, nuttcp, and iperf. Here is the original announcement.

-- SimonLeinen - 22 Apr 2008

FreeBSD 7.0 networking enhancements

Release 7.0 of the FreeBSD operating system was released on 27 February. This versions claims important improvements in the area of network performance and scalability. It joins Linux and Windows (Vista and newer) in implementing TCP buffer auto-tuning, and also adds large-send (LSO) and large-receive offload (LRO) for some adapters. See this ONLamp article for discussions of some of the enhancements.

-- SimonLeinen - 02 Apr 2008

Wireshark news: 1.0 released; First Sharkfest 31 March - 2 April 2008

Wireshark 1.0 was announced on 31 March 2008. For changes from previous versions, see the Wireshark 1.0.0 Release Notes.

The release was made just in time for the first annual SHARKFEST event, which took place on 31 March - 2 April at Foothill College in Los Altos Hills, CA, USA, and which featured a keynote speech by Vint Cerf.

-- SimonLeinen - 02 Apr 2008 (updated)

Vista Service Pack 1 (SP1) Released

The first Service Pack to Microsoft's Windows Vista operating system was published on Microsoft's download site on 18 March. The Service Pack should be available through Windows Update. Reportedly, the kernel of Vista SP1 is based on Windows 2008, while the kernel in the original Vista was based on Windows 2003.

I have skimmed the "Notable Changes" document for performance-relevant changes in networking code. There are several significant additions such as more encryption and random-number generation mechanisms for IPSec, 802.11n (high-speed wireless LAN) support, and support for full TCP offload (TOE) called TCP Chimney.

-- SimonLeinen - 26 Mar 2008

Discussion on High-Performance Networking and SSH in OpenSolaris community

Sun's Solaris operating system is changing to an open-source/community-based development model, so most of the design discussions about the evolution of that system can now be found on the Web and participated in. I'm following the networking community's forum, and found an interesting thread on Improving HPC network performance, specifically SSH/SCP/SFTP perf starting 14 February 2008. The discussion touches topics of TCPBufferAutoTuning, (HPN-) SecureShell (SSH), and other improvements that would help Bulk File Transfers.

Unfortunately, the discussion stopped short of somebody volunteering to actually start work on these improvements in the form of an OpenSolaris project...

-- SimonLeinen - 07 Mar 2008

Amazon Web Services (AWS) and high-performance networking

It is well known that Amazon uses a large computing and network infrastructure to power its bookstores and other e-business activities, Over the past few years, they have also diversified by commercially offering access to this infrastructure to third parties. These services are called "Amazon Web Services (AWS)", and they include a storage service ("S3" for Simple Storage Service), compute services based on virtual machine environments that can be installed on large numbers of nodes ("EC2" for "Elastic Compute Cloud") and others.

According to an item in yesterday's newsletter, they recently added support for Window Scaling and SACK to the TCP configuration of their servers, to improve performance for their many remote users.

-- SimonLeinen - 05 Mar 2008

 

News Archive:

Added:
>
>
 

Revision 612009-11-17 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Pluggable TCP (and SCTP?) congestion control under discussion at OpenSolaris

On the networking discussion forum for OpenSolaris, a project proposal for a project on "Solaris Network Performance Projects" is currently under discussion:

http://www.opensolaris.org/jive/thread.jspa?messageID=435766

One of these projects is "pluggable congestion control". It foresees implementation of the HighSpeed, CUBIC, Westwood+, and Vegas congestion control algorithms, as well as ipadm subcommands and socket options to get and set congestion control parameters. If successful, this will add OpenSolaris - and possibly at some point Solaris - to the group os OSes that support modern and configurable TCP congestion control (Linux, newer versions of Windows, newer versions of FreeBSD).

-- SimonLeinen - 17 Nov 2009

 

http://fasterdata.es.net/

In a message to the NANOG mailing list, Kevin Oberman points to ESnet Network Performance Knowledge Base. It has TCP tuning tips, hints on choosing and configuring file transfer tools, and information about many other areas. There's obviously some overlap with the information here. One great advantage of the ESnet site is that the information is quite concise and presented in a very pragmatic way. The site focuses on one problem, namely helping users transfer large amounts of data over wide-area networks (see the FileTransfer topic here).

Revision 602009-11-09 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

http://fasterdata.es.net/

In a message to the NANOG mailing list, Kevin Oberman points to ESnet Network Performance Knowledge Base. It has TCP tuning tips, hints on choosing and configuring file transfer tools, and information about many other areas. There's obviously some overlap with the information here. One great advantage of the ESnet site is that the information is quite concise and presented in a very pragmatic way. The site focuses on one problem, namely helping users transfer large amounts of data over wide-area networks (see the FileTransfer topic here).

-- SimonLeinen - 09 Nov 2009

 

pert-discuss mailing list revival

Throughout the history of the PERT in the extended European research networking community, there has been a mailing list for discussions. This list is open for those who are interested in contributing to the PERT effort. With the transition from GN2 to the GN3 project, the home of the list (similar to other resources) has changed to a geant.net domain. In the process of this, archived messages since October 2004 have been made publicly available under http://mail.geant.net/public/pert-discuss/. The archive contains interesting discussions about specific and general performance issues. The list has been dormant for about a year, but we will try to revive it as the eduPERT activities in GN3 are gaining speed.

Revision 592009-10-29 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

pert-discuss mailing list revival

Throughout the history of the PERT in the extended European research networking community, there has been a mailing list for discussions. This list is open for those who are interested in contributing to the PERT effort. With the transition from GN2 to the GN3 project, the home of the list (similar to other resources) has changed to a geant.net domain. In the process of this, archived messages since October 2004 have been made publicly available under http://mail.geant.net/public/pert-discuss/. The archive contains interesting discussions about specific and general performance issues. The list has been dormant for about a year, but we will try to revive it as the eduPERT activities in GN3 are gaining speed.

-- SimonLeinen - 29 Oct 2009

 

Mobile-optimized search page

This wiki uses a "custom search engine" provided by Google to implement its "Search" function. According to news items, Google now provide, for each such custom search engine, a search page that is optimized for "high-end" mobile devices with web browsers. So if you have an iPhone/iPod touch, Palm Pre, Android-powered device or similar, try the following link:

Revision 582009-10-28 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Mobile-optimized search page

This wiki uses a "custom search engine" provided by Google to implement its "Search" function. According to news items, Google now provide, for each such custom search engine, a search page that is optimized for "high-end" mobile devices with web browsers. So if you have an iPhone/iPod touch, Palm Pre, Android-powered device or similar, try the following link:

http://www.google.com/cse/home?cx=012443065831413718779:zokjgbpdcke

Optimizing Internet applications so that they work well with this kind of devices is an interesting general PERT problem, of course!

-- SimonLeinen - 28 Oct 2009

 

New domain names, new ToDo topic

The eduPERT activities in the GN3 project are spinning up. As a first step, many of the eduPERT-related domain names were moved from geant2.net to the geant.net namespace. This includes this PERT Knowledge Base, which now officially lives under http://kb.pert.geant.net/. Please update your pointers!

Revision 572009-07-21 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

New domain names, new ToDo topic

The eduPERT activities in the GN3 project are spinning up. As a first step, many of the eduPERT-related domain names were moved from geant2.net to the geant.net namespace. This includes this PERT Knowledge Base, which now officially lives under http://kb.pert.geant.net/. Please update your pointers!

Then, I have added a ToDo topic, with some ideas on how the Knowledge Base could be improved. Please feel free to add to this list. But what would be really great is if people used the list as inspiration for contributions to the PERT Knowledge Base. Remember, this is supposed to be a community effort.

-- SimonLeinen - 21 Jul 2009

 

PERT/eduPERT talk at TNC 2009

I gave a talk about PERTs and the eduPERT at the Terena Networking Conference (TNC) 2009 in Málaga, Spain, last Thursday. Audio/video recording as well as presentation slides are available from the conference site. Here are the title and abstract:

Revision 562009-06-14 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PERT/eduPERT talk at TNC 2009

I gave a talk about PERTs and the eduPERT at the Terena Networking Conference (TNC) 2009 in Málaga, Spain, last Thursday. Audio/video recording as well as presentation slides are available from the conference site. Here are the title and abstract:

eduPERT - helping European researchers bridge the "Wizard Gap"

PERTs (Performance Enhancement and Response Team) have been proposed to help network users achieve better end-to-end performance over today's fast research networks. Over the past few years, the PERT concept has been tested both within the GN2 project and within European NRENs with some success. This talk will explain "eduPERT", the current federated PERT model for the GÉANT community, talk about some interesting performance problems handled by PERTs, and about lessons learned and future work for advancing the PERT idea.

-- SimonLeinen - 14 Jun 2009

 

AWS (Amazon Web Services) Import/Export service

Amazon just announced a service (in beta) which allows you to send and retrieve your data from/to their cloud infrastructure through physical shipping of portable disk drives: http://aws.amazon.com/importexport/

Revision 552009-06-10 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

AWS (Amazon Web Services) Import/Export service

Amazon just announced a service (in beta) which allows you to send and retrieve your data from/to their cloud infrastructure through physical shipping of portable disk drives: http://aws.amazon.com/importexport/

The article has some interesting rules-of-thumb explaining when using this option is preferable to up/downloading your data through the network, depending on connection speed and data volume.

 

eduPERT sites (including this one) transitioned to geant.net

The GN2 (GÉANT2) project ended in April 2009. The eduPERT activity continues as part of the GN3 project. To reflect this change, various domain names were changed from geant2.net to geant.net (not geant3.net, so we won't have to do this again four years from now - clever, huh?). In particular,

Revision 542009-06-10 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

eduPERT sites (including this one) transitioned to geant.net

The GN2 (GÉANT2) project ended in April 2009. The eduPERT activity continues as part of the GN3 project. To reflect this change, various domain names were changed from geant2.net to geant.net (not geant3.net, so we won't have to do this again four years from now - clever, huh?). In particular,

edupert.geant2.net is now edupert.geant.net (main portal to eduPERT federation)
kb.pert.geant2.net is now kb.pert.geant.net (this site)
pert-registration@geant2.net is now pert-registration@geant.net (for registering information about PERTs)
pert-accreditation@geant2.net is now pert-accreditation@geant.net (to request accreditation of your PERT)

Please update pointers from other sites, your bookmarks, and address books. At least for the Web sites, the old addresses will still work and/or redirect to the new names, but it's best to start using the new version-independent names as soon as possible in order to avoid confusion.

One exception is the pert-discuss@geant2.net mailing list, which will have to be transitioned more carefully in the near future.

-- SimonLeinen - 10 Jun 2009

 

Recordings and other Materials from eduPERT Training Event now available

In late November 2008, the GN2 project held an eduPERT training workshop in Zurich, which was attended by 17 participants. Audio/video recordings, slide presentations, and workbooks with exercises can be found under the PertTraining2008 topic.

Revision 532008-12-19 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Recordings and other Materials from eduPERT Training Event now available

In late November 2008, the GN2 project held an eduPERT training workshop in Zurich, which was attended by 17 participants. Audio/video recordings, slide presentations, and workbooks with exercises can be found under the PertTraining2008 topic.

-- SimonLeinen - 19 Dec 2008

 

TERENA NREN Compendium 2008 released, shows growing interest in PERTs

The 2008 edition of the TERENA compendium was just released. TERENA surveys all European NRENs (National Research and Education Networks) yearly about their infrastructure, service offerings, and other activities. The "summary of key findings" section of the printed edition has the following to say about PERTs: "At the moment, 16 out of the 30 EU/EFTA countries have Performance

Revision 522008-11-02 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Line: 48 to 48
 

Linux 2.6.26

Changed:
<
<
Another new Linux kernel version has been released today. Notable new features related to network performance are support for some 802.11n high-speed WLAN cards, and a performance improvement in [[CubicTcp][CUBIC TCP]. See the CUBIC topic for the patch, and for some measurements of CUBIC compared with other TCP variants.
>
>
Another new Linux kernel version has been released today. Notable new features related to network performance are support for some 802.11n high-speed WLAN cards, and a performance improvement in CUBIC TCP. See the CUBIC topic for the patch, and for some measurements of CUBIC compared with other TCP variants.
  -- SimonLeinen - 14 Jul 2008

Revision 512008-10-24 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

TERENA NREN Compendium 2008 released, shows growing interest in PERTs

The 2008 edition of the TERENA compendium was just released. TERENA surveys all European NRENs (National Research and Education Networks) yearly about their infrastructure, service offerings, and other activities. The "summary of key findings" section of the printed edition has the following to say about PERTs: "At the moment, 16 out of the 30 EU/EFTA countries have Performance Enhancement and Response Teams (PERTs). Four are planning to establish one within the next year and a further six plan to do this within the next three years. Many NRENs from other countries also either have a PERT or are planning to establish one." Progress with PERT establishment is also mentioned in a companion release, What the 2008 TERENA Compendium reveals. The summarized responses can be found on pages 63 and 64 of the printed edition. Under the individual NRENs' responses, PERT information can be found under the respective "Network Connectivity and Services" section.

-- SimonLeinen - 24 Oct 2008

 

PERT Training Event announced

We will be doing another PERT training event in Zurich on 27/28 November. See PertTraining2008 for more information, including a registration link.

Revision 502008-09-21 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PERT Training Event announced

We will be doing another PERT training event in Zurich on 27/28 November. See PertTraining2008 for more information, including a registration link.

-- SimonLeinen - 21 Sep 2008

TWiki software update

Today I upgraded the Debian twiki package on this host, in order to activate a security fix. This entry is mainly to test whether the wiki can still be written to.

-- SimonLeinen - 21 Sep 2008

eduPERT

As of 1 September 2008, the GEANT2 community's Federated PERT is known under the name eduPERT. It has a new Web site on http://edupert.geant2.net/.

-- SimonLeinen - 21 Sep 2008

 

More GridFTP information: new portal, tutorials, success stories

I found tutorial slides and handouts on the Web, under a new GridFTP portal. This should be quite useful to get people started with these tools. Pointers have been added to the GridFtpProtocol topic.

Revision 492008-07-28 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

More GridFTP information: new portal, tutorials, success stories

I found tutorial slides and handouts on the Web, under a new GridFTP portal. This should be quite useful to get people started with these tools. Pointers have been added to the GridFtpProtocol topic.

-- SimonLeinen - 28 Jul 2008

 

DTrace IP Provider integrated into OpenSolaris

(message from 7 Jul 2008, edited to add note about Nevada b93)

Revision 482008-07-14 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

Dilbert on Network Performance

>
>

DTrace IP Provider integrated into OpenSolaris

 
Changed:
<
<
>
>
(message from 7 Jul 2008, edited to add note about Nevada b93)

DTrace is a mechanism for system-wide instrumentation for performance analysis and debugging that was pioneered in Sun's Solaris operating system. It has since been added to other systems such as BSD (forgot which variant(s)) and Apple's Mac OS X. In the OpenSolaris community, there is a project to add network -specific probes to DTrace. I just learned that the DTrace IP Provider, which is outlined in this "onepager", was integrated into OpenSolaris Nevada four weeks ago. It was released yesterday as part of Nevada b93.

It is interesting to compare the DTrace approach with the web100 system. I think that there is a large overlap in the things that you can do with both, but the interfaces are quite different. Of course, web100 is very specific to TCP, while DTrace is a general tool for observing system behavior.

  -- SimonLeinen - 14 Jul 2008
Changed:
<
<

DTrace IP Provider integrated into OpenSolaris

>
>

Linux 2.6.26

 
Changed:
<
<
DTrace is a mechanism for system-wide instrumentation for performance analysis and debugging that was pioneered in Sun's Solaris operating system. It has since been added to other systems such as BSD (forgot which variant(s)) and Apple's Mac OS X. In the OpenSolaris community, there is a project to add network -specific probes to DTrace. I just learned that the DTrace IP Provider, which is outlined in this "onepager", was integrated into OpenSolaris Nevada four weeks ago.
>
>
Another new Linux kernel version has been released today. Notable new features related to network performance are support for some 802.11n high-speed WLAN cards, and a performance improvement in [[CubicTcp][CUBIC TCP]. See the CUBIC topic for the patch, and for some measurements of CUBIC compared with other TCP variants.
 
Changed:
<
<
It is interesting to compare the DTrace approach with the web100 system. I think that there is a large overlap in the things that you can do with both, but the interfaces are quite different. Of course, web100 is very specific to TCP, while DTrace is a general tool for observing system behavior.
>
>
-- SimonLeinen - 14 Jul 2008

Dilbert on Network Performance

 
Changed:
<
<
-- SimonLeinen - 07 Jul 2008
>
>

-- SimonLeinen - 14 Jul 2008

 

New GridFTP supports SSH for its control channel

Revision 472008-07-14 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Dilbert on Network Performance

-- SimonLeinen - 14 Jul 2008

 

DTrace IP Provider integrated into OpenSolaris

DTrace is a mechanism for system-wide instrumentation for performance analysis and debugging that was pioneered in Sun's Solaris operating system. It has since been added to other systems such as BSD (forgot which variant(s)) and Apple's Mac OS X. In the OpenSolaris community, there is a project to add network -specific probes to DTrace. I just learned that the DTrace IP Provider, which is outlined in this "onepager", was integrated into OpenSolaris Nevada four weeks ago.

Revision 462008-07-07 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

DTrace IP Provider integrated into OpenSolaris

DTrace is a mechanism for system-wide instrumentation for performance analysis and debugging that was pioneered in Sun's Solaris operating system. It has since been added to other systems such as BSD (forgot which variant(s)) and Apple's Mac OS X. In the OpenSolaris community, there is a project to add network -specific probes to DTrace. I just learned that the DTrace IP Provider, which is outlined in this "onepager", was integrated into OpenSolaris Nevada four weeks ago.

It is interesting to compare the DTrace approach with the web100 system. I think that there is a large overlap in the things that you can do with both, but the interfaces are quite different. Of course, web100 is very specific to TCP, while DTrace is a general tool for observing system behavior.

-- SimonLeinen - 07 Jul 2008

 

New GridFTP supports SSH for its control channel

The new version 4.2 of the Globus Toolkit was released on July 2. One of its new features is that its GridFTP component can now use SSH for its control channel. This could make it much more attractive for users who just want to haul large amounts of data over LFNs, but who don't necessarily want to run the full Globus infrastructure. Anybody wants to give this a try?

Revision 452008-07-04 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

New GridFTP supports SSH for its control channel

The new version 4.2 of the Globus Toolkit was released on July 2. One of its new features is that its GridFTP component can now use SSH for its control channel. This could make it much more attractive for users who just want to haul large amounts of data over LFNs, but who don't necessarily want to run the full Globus infrastructure. Anybody wants to give this a try?

-- SimonLeinen - 04 Jul 2008

 

ACM Queue: Kode Vicious articles about Latency and Livelock

ACM Queue is a relatively new publication of the Association for Computing Machinery (ACM) oriented to IT practitioners. It includes a section where Kode Vicious, the slightly psychotic alter ego of George V. Neville-Neil, responds to (real?) questions from readers. The column from the March/April 2008 issue is now available online, with two articles relevant for (network) performance: One talks about the often neglected impact of latency on performance even where bandwidth is plentiful, and the other explains a problem called "livelock", which can be seen when servers get persistently overloaded. At SWITCH, I believe we have seen a nice example of this with the RRD MA server that we host for PerfSONAR. Unfortunately, the article about latency has an annoying decimal-point error, which leads to some hand-waving about router-induced delays that is inappropriate - I posted a correction as a reader comment.

Revision 442008-06-04 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

ACM Queue: Kode Vicious articles about Latency and Livelock

ACM Queue is a relatively new publication of the Association for Computing Machinery (ACM) oriented to IT practitioners. It includes a section where Kode Vicious, the slightly psychotic alter ego of George V. Neville-Neil, responds to (real?) questions from readers. The column from the March/April 2008 issue is now available online, with two articles relevant for (network) performance: One talks about the often neglected impact of latency on performance even where bandwidth is plentiful, and the other explains a problem called "livelock", which can be seen when servers get persistently overloaded. At SWITCH, I believe we have seen a nice example of this with the RRD MA server that we host for PerfSONAR. Unfortunately, the article about latency has an annoying decimal-point error, which leads to some hand-waving about router-induced delays that is inappropriate - I posted a correction as a reader comment.

 

RFC 5236 (Reorder Density)

There is a new RFC with an alternative ("improved") metric for packet reordering. It doesn't seem to be an official product of the IPPM WG, and its status is informational (not standards-track), and it doesn't obsolete RFC 4737. I added the reference to the PacketReordering and IPPerformanceMetrics topics.

Revision 432008-06-04 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

RFC 5236 (Reorder Density)

There is a new RFC with an alternative ("improved") metric for packet reordering. It doesn't seem to be an official product of the IPPM WG, and its status is informational (not standards-track), and it doesn't obsolete RFC 4737. I added the reference to the PacketReordering and IPPerformanceMetrics topics.

LSO vs. MDT

There is an interesting discussion on the OpenSolaris networking forum that is still ongoing. The original topic was whether OpenSolaris supported LSO for UDP (apparently it doesn't). The discussion quickly evolved into a debate of the relative merits of LSO - where the adapter knows how to segment transport data units and construct all headers for them - and MDT, which is like a software-only variant of LSO, where the host (operating system) still has to do the segmentation and header construction, but can send all the segments in a single (bus) transaction. I added a few sentences on this trade-off to the LargeSendOffloadLSO topic.

-- SimonLeinen - 04 Jun 2008

 

bwctl 1.3 release candidate

The first release candidate for a new version 1.3 of bwctl was released yesterday. Changes include more robust scheduling, improved error handling and logging, parallel streams (-P), CSV output, and support for several throughput measurement tools: thrulay, nuttcp, and iperf. Here is the original announcement.

Revision 422008-04-22 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

bwctl 1.3 release candidate

The first release candidate for a new version 1.3 of bwctl was released yesterday. Changes include more robust scheduling, improved error handling and logging, parallel streams (-P), CSV output, and support for several throughput measurement tools: thrulay, nuttcp, and iperf. Here is the original announcement.

-- SimonLeinen - 22 Apr 2008

 

FreeBSD 7.0 networking enhancements

Release 7.0 of the FreeBSD operating system was released on 27 February. This versions claims important improvements in the area of network performance and scalability. It joins Linux and Windows (Vista and newer) in implementing TCP buffer auto-tuning, and also adds large-send (LSO) and large-receive offload (LRO) for some adapters. See this ONLamp article for discussions of some of the enhancements.

Revision 412008-04-02 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

Vista Service Pack 1 (SP1) Released

>
>

FreeBSD 7.0 networking enhancements

 
Changed:
<
<
The first Service Pack to Microsoft's Windows Vista operating system was published on Microsoft's download site on 18 March. The Service Pack should be available through Windows Update. Reportedly, the kernel of Vista SP1 is based on Windows 2008, while the kernel in the original Vista was based on Windows 2003.
>
>
Release 7.0 of the FreeBSD operating system was released on 27 February. This versions claims important improvements in the area of network performance and scalability. It joins Linux and Windows (Vista and newer) in implementing TCP buffer auto-tuning, and also adds large-send (LSO) and large-receive offload (LRO) for some adapters. See this ONLamp article for discussions of some of the enhancements.
 
Changed:
<
<
I have skimmed the "Notable Changes" document for performance-relevant changes in networking code. There are several significant additions such as more encryption and random-number generation mechanisms for IPSec, 802.11n (high-speed wireless LAN) support, and support for full TCP offload (TOE) called TCP Chimney.
>
>
-- SimonLeinen - 02 Apr 2008
 
Changed:
<
<
-- SimonLeinen - 26 Mar 2008
>
>

Wireshark news: 1.0 released; First Sharkfest 31 March - 2 April 2008

Wireshark 1.0 was announced on 31 March 2008. For changes from previous versions, see the Wireshark 1.0.0 Release Notes.

 
Changed:
<
<

Wireshark news: 1.0 release imminent; First Sharkfest 31 March - 2 April 2008

>
>
The release was made just in time for the first annual SHARKFEST event, which took place on 31 March - 2 April at Foothill College in Los Altos Hills, CA, USA, and which featured a keynote speech by Vint Cerf.
 
Changed:
<
<
On 18 March, The Wireshark-announcements mailing list had an announcement of Wireshark1.0.0pre1, the first prerelease to the 1.0 (or 1.0.0) release. The announcement also has a pointer to the Wireshark 1.0.0 Release Notes.
>
>
-- SimonLeinen - 02 Apr 2008 (updated)
 
Changed:
<
<
We can assume that the Wireshark team wants to have that 1.0.0 release ready in time for the first annual SHARKFEST event, which will take place on 31 March - 2 April at Foothill College in Los Altos Hills, CA, USA, and which will feature a keynote speech by Vint Cerf.
>
>

Vista Service Pack 1 (SP1) Released

 
Changed:
<
<
-- SimonLeinen - 19 Mar 2008
>
>
The first Service Pack to Microsoft's Windows Vista operating system was published on Microsoft's download site on 18 March. The Service Pack should be available through Windows Update. Reportedly, the kernel of Vista SP1 is based on Windows 2008, while the kernel in the original Vista was based on Windows 2003.

I have skimmed the "Notable Changes" document for performance-relevant changes in networking code. There are several significant additions such as more encryption and random-number generation mechanisms for IPSec, 802.11n (high-speed wireless LAN) support, and support for full TCP offload (TOE) called TCP Chimney.

-- SimonLeinen - 26 Mar 2008

 

Discussion on High-Performance Networking and SSH in OpenSolaris community

Revision 402008-03-26 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Vista Service Pack 1 (SP1) Released

The first Service Pack to Microsoft's Windows Vista operating system was published on Microsoft's download site on 18 March. The Service Pack should be available through Windows Update. Reportedly, the kernel of Vista SP1 is based on Windows 2008, while the kernel in the original Vista was based on Windows 2003.

I have skimmed the "Notable Changes" document for performance-relevant changes in networking code. There are several significant additions such as more encryption and random-number generation mechanisms for IPSec, 802.11n (high-speed wireless LAN) support, and support for full TCP offload (TOE) called TCP Chimney.

-- SimonLeinen - 26 Mar 2008

 

Wireshark news: 1.0 release imminent; First Sharkfest 31 March - 2 April 2008

On 18 March, The Wireshark-announcements mailing list had an announcement of Wireshark1.0.0pre1, the first prerelease to the 1.0 (or 1.0.0) release. The announcement also has a pointer to the Wireshark 1.0.0 Release Notes.

Revision 392008-03-19 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Wireshark news: 1.0 release imminent; First Sharkfest 31 March - 2 April 2008

On 18 March, The Wireshark-announcements mailing list had an announcement of Wireshark1.0.0pre1, the first prerelease to the 1.0 (or 1.0.0) release. The announcement also has a pointer to the Wireshark 1.0.0 Release Notes.

We can assume that the Wireshark team wants to have that 1.0.0 release ready in time for the first annual SHARKFEST event, which will take place on 31 March - 2 April at Foothill College in Los Altos Hills, CA, USA, and which will feature a keynote speech by Vint Cerf.

-- SimonLeinen - 19 Mar 2008

 

Discussion on High-Performance Networking and SSH in OpenSolaris community

Sun's Solaris operating system is changing to an open-source/community-based development model, so most of the design discussions about the evolution of that system can now be found on the Web and participated in. I'm following the networking community's forum, and found an interesting thread on Improving HPC network performance, specifically SSH/SCP/SFTP perf starting 14 February 2008. The discussion touches topics of TCPBufferAutoTuning, (HPN-) SecureShell (SSH), and other improvements that would help Bulk File Transfers.

Revision 382008-03-07 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Discussion on High-Performance Networking and SSH in OpenSolaris community

Sun's Solaris operating system is changing to an open-source/community-based development model, so most of the design discussions about the evolution of that system can now be found on the Web and participated in. I'm following the networking community's forum, and found an interesting thread on Improving HPC network performance, specifically SSH/SCP/SFTP perf starting 14 February 2008. The discussion touches topics of TCPBufferAutoTuning, (HPN-) SecureShell (SSH), and other improvements that would help Bulk File Transfers.

Unfortunately, the discussion stopped short of somebody volunteering to actually start work on these improvements in the form of an OpenSolaris project...

-- SimonLeinen - 07 Mar 2008

 

Amazon Web Services (AWS) and high-performance networking

It is well known that Amazon uses a large computing and network infrastructure to power its bookstores and other e-business activities, Over the past few years, they have also diversified by commercially offering access to this infrastructure to third parties. These services are called "Amazon Web Services (AWS)", and they include a storage service ("S3" for Simple Storage Service), compute services based on virtual machine environments that can be installed on large numbers of nodes ("EC2" for "Elastic Compute Cloud") and others.

Revision 372008-03-05 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Amazon Web Services (AWS) and high-performance networking

Changed:
<
<
It is well known that Amazon uses a large computing and network infrastructure to power its bookstores and other e-business activities, Over the past few years, they have also diversified by commercially offering access to this infrastructure to third parties. These services are called "Amazon Web Services", and they include a storage service ("S3" for Simple Storage Service), compute services based on virtual machine environments that can be installed on large numbers of nodes ("EC2" for "Elastic Compute Cloud") and others.
>
>
It is well known that Amazon uses a large computing and network infrastructure to power its bookstores and other e-business activities, Over the past few years, they have also diversified by commercially offering access to this infrastructure to third parties. These services are called "Amazon Web Services (AWS)", and they include a storage service ("S3" for Simple Storage Service), compute services based on virtual machine environments that can be installed on large numbers of nodes ("EC2" for "Elastic Compute Cloud") and others.
  According to an item in yesterday's newsletter, they recently added support for Window Scaling and SACK to the TCP configuration of their servers, to improve performance for their many remote users.

Revision 362008-03-05 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

SC'07 Bandwidth Challenge results

>
>

Amazon Web Services (AWS) and high-performance networking

 
Changed:
<
<
According to this press release from Indiana University,
>
>
It is well known that Amazon uses a large computing and network infrastructure to power its bookstores and other e-business activities, Over the past few years, they have also diversified by commercially offering access to this infrastructure to third parties. These services are called "Amazon Web Services", and they include a storage service ("S3" for Simple Storage Service), compute services based on virtual machine environments that can be installed on large numbers of nodes ("EC2" for "Elastic Compute Cloud") and others.
 
Changed:
<
<
"A team led by Indiana University, with partners from the Technische Universitaet Dresden, Rochester Institute of Technology, Oak Ridge National Laboratory and the Pittsburgh Supercomputing Center, was awarded first place in an international competition for leading-edge, high-bandwidth computing applications. [...] Using the IU Data Capacitor, a system designed to store and manipulate massive data sets, the IU team achieved a peak transfer rate of 18.21 Gigabits/second out of a possible maximum of 20 Gigabits/second. This performance was nearly twice the peak rate of the nearest competitor. The IU team achieved an overall sustained rate of 16.2 Gigabits/second (roughly equivalent to sending 170 CDs of data per minute) using a transatlantic network path that included the Internet2, GÉANT, and DFN research networks."
>
>
According to an item in yesterday's newsletter, they recently added support for Window Scaling and SACK to the TCP configuration of their servers, to improve performance for their many remote users.
 
Changed:
<
<
(Hmm, CDs? I thought that these days, the canonical unit for expressing incredibly large amounts of data was the 160 Gigabyte iPod ... oh well smile

Congratulations to the team!

-- SimonLeinen - 22 Nov 2007

Wireshark 0.99.7 to be released soon

According to this announcement, Wireshark 0.99.7 should be released by the end of November 2007. It fixes numerous security issues, moves the capture code out of the GUI so that Wireshark can be run without root privileges, and of course includes many new or improved protocol dissectors. For more information about the changes, see the Wireshark 0.99.7 Release Notes living document.

-- SimonLeinen - 22 Nov 2007

PERT Training Workshop: Recorded Audio/Video streams available

The PERT Training workshop from 19-21 September 2007 was recorded by SWITCH's end-user team. See PertTrainingWorkshop for pointers to the slides and audio/video streams.

-- SimonLeinen - 03 Oct 2007

PAM 2008 Call for Papers

The Passive & Active Measurement Conference (PAM 2008) will take place in Cleveland, Ohio on 29/30 April 2008. Contributions must be registered by 14 October, and submitted by 21 October 2007.

Successful PERT Training workshop

As announced earlier, SWITCH hosted a training event from 19-21 September. It was attended by about thirty engineers from National Research Networks, Campus IT/Networking divisions and other organizations, in three separate "streams". We will try to make the meeting materials and audio/video recordings available soon.

-- SimonLeinen - 24 Sep 2007

Another server move

There hasn't been a lot of activity here over the summer months, for which I apologize. We have been busy preparing the PERT Training Event (see below) - please check it out, it will be fun.

The Knowledge Base (TWiki server) was moved to a different (newer and faster) machine today, because the old one will be retired when we have to move to different racks at CERN, probably in September, i.e. soon. Hopefully everything continues to work as usual, only faster (it's a nice machine).

Prompted by an interesting post on the DCCP mailing list, I finally started to write some topics on interactive (real-time) applications. Please feel free to add to them or enter suggestions on what you would like improved.

-- SimonLeinen - 24 Aug 2007

PERT Training September 19-21: Registration open

The GN2 project is organizing a training event in September. The event will be hosted by SWITCH in Zurich. Preliminary information and a registration page can be found here: http://www.terena.org/activities/training/pert/

-- SimonLeinen - 10 Jul 2007

Internet² Spring 2007 Member Meeting

I haven't been to the Internet² meeting in late April, but started to browse through some of the presentations/Webcasts. Joe St. Sauver's talk about Capacity Planning and System and Network Security is certainly relevant to network performance. It mentions a new network tuning tool for Mac OS X, which will be documented here shortly.

Knowledge Base administrativia

Emergency server move

The server that has been hosting the PERT KB wiki since its inception has finally started to break down a week ago. We had to move the wiki to another server, which should be transparent to users (the new server is actually a faster machine). We also started the purchase process for getting a new server for the wiki and other tools that had traditionally been running on the same host (Looking Glass, NDT etc.). Looks like this will be a nice eight-core Intel box.

Making the PERT KB more popular and more useful

The PERT KB sees pretty stable usage, mostly from search engine referrals. Over the past few months, the KB was mentioned twice on networking fora, once on the NANOG mailing list and once on the OpenSolaris networking forum. Both these postings left clear traces in the access statistics, with significantly higher-than-usual accesses over a couple of days. So, if you find a nice occasion to mention the PERT KB - or a particular topic - please don't hesitate to do so. Please use URLs of the "canonical" form http://kb.pert.switch.ch/PERTKB/TopicName when doing so.

It would also be great if you added links to the PERT KB - or specific topics - to Web pages that you or your organization maintains that are related to network performance. This will improve our ranking in search engines. And there is room for improvement; when you type "PERT Team" into Google, the first hit is the Psychiatric Emergency Response Team at the San Diego Sheriff's office. We don't even make it on the first page of search results!

In the same vein, if you have any ideas on how to improve the PERT KB - better organization, missing topics, anything really - you can either contact me via e-mail, or simply edit your suggestions into the Wiki at any suitable place. Someone will notice the changes and act upon them!

-- SimonLeinen - 15 May 2007

Materials from ICCRG and High-speed TCP meetings published

There were two other meetings around the PFLDnet meeting (see below), and the slides from these meetings were just put up on the Web:

  • The IRTF ICCRG (Internet Congestion Control Research Group) meeting - slides, minutes
  • A workshop on high-speed TCP modifications organized by Microsoft Research - slides

-- SimonLeinen - 07 Mar 2007

PFLDnet 2007 fallout

Before Chris went to PFLDnet, I had him promise to add interesting stuff to the knowledge base when he's back... which he has now started to do. Check out the TcpHighSpeedVariants section, which has new sections about YeahTcp and TcpFusion, as well as an overdue one on CompoundTCP, which is in Microsoft Vista. Hopefully more topics will come soon, but don't let that stop you from proofreading those that exist...

-- SimonLeinen - 26 Feb 2007

PFLDnet 2007, Web100 for Linux 2.6.20, Sun 10GE card

After a week of vacation, I found lots of exciting news.

  • My colleague Chris is back from the PFLDnet workshop and brought the proceedings. From a first quick browse, this seemed to include lots of relevant stuff. I hope Chris and others will help me add pointers to the PERT KB...
  • The Web100 patch has been updated for 2.6.20 already (presumably on John Heffner's flight back from PFLDnet :-). I built a new kernel and installed it on a test machine that Chris Rapier from PSC wants to use for work on his HPC SSH patch
  • Sun announced a new dual-port 10GE adapter with PCI Express 1.1 (x8) host interface. It seems to have interesting on-board demultiplexing logic including support for many parallel DMA channels. Probably very useful for a busy server with many (not necessary so fast...) CPUs. Maybe also useful for "Land-Speed Record" stuff because it also supports "bonding" of the two channels to a 20 Gb/s logical channel.

-- SimonLeinen - 20 Feb 2007

Linux kernel 2.6.20 is out

Past Sunday, the new Linux kernel release came out. There are a few improvements in the networking area, notably

  • support for NetXen GE and 10GE adapters
  • TCP congestion control choice for mere mortals, restricted by system-wide configuration
  • SACK fixes
  • Hardware TSO for IPv6 on Intel e1000 adapters
  • Support for more Chelsio 10GE cards

You probably don't want to read the entire 47290-line ChangeLog, but there's a nice concise list of changes on the kernelnewbies.org wiki.

-- SimonLeinen - 06 Feb 2007

>
>
-- SimonLeinen - 05 Mar 2008
 

News Archive:

Added:
>
>
 

Revision 352007-11-22 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

PERT Training Workshop: Recorded Audio/Video streams available

>
>

SC'07 Bandwidth Challenge results

According to this press release from Indiana University,

"A team led by Indiana University, with partners from the Technische Universitaet Dresden, Rochester Institute of Technology, Oak Ridge National Laboratory and the Pittsburgh Supercomputing Center, was awarded first place in an international competition for leading-edge, high-bandwidth computing applications. [...] Using the IU Data Capacitor, a system designed to store and manipulate massive data sets, the IU team achieved a peak transfer rate of 18.21 Gigabits/second out of a possible maximum of 20 Gigabits/second. This performance was nearly twice the peak rate of the nearest competitor. The IU team achieved an overall sustained rate of 16.2 Gigabits/second (roughly equivalent to sending 170 CDs of data per minute) using a transatlantic network path that included the Internet2, GÉANT, and DFN research networks."

(Hmm, CDs? I thought that these days, the canonical unit for expressing incredibly large amounts of data was the 160 Gigabyte iPod ... oh well smile

Congratulations to the team!

 
Changed:
<
<
The PERT Training workshop from 19-21 September 2007 was recorded by SWITCH's end-user team. Here is a list of available sessions, with pointers to Adobe "Flash" format streams. We can make these streams available in a few other formats (notably QuicktimeTM) if requested. Send mail to simon dot leinen at switch dot ch.
>
>
-- SimonLeinen - 22 Nov 2007

Wireshark 0.99.7 to be released soon

According to this announcement, Wireshark 0.99.7 should be released by the end of November 2007. It fixes numerous security issues, moves the capture code out of the GUI so that Wireshark can be run without root privileges, and of course includes many new or improved protocol dissectors. For more information about the changes, see the Wireshark 0.99.7 Release Notes living document.

-- SimonLeinen - 22 Nov 2007

PERT Training Workshop: Recorded Audio/Video streams available

 
Changed:
<
<
>
>
The PERT Training workshop from 19-21 September 2007 was recorded by SWITCH's end-user team. See PertTrainingWorkshop for pointers to the slides and audio/video streams.
  -- SimonLeinen - 03 Oct 2007

Revision 342007-10-03 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PERT Training Workshop: Recorded Audio/Video streams available

The PERT Training workshop from 19-21 September 2007 was recorded by SWITCH's end-user team. Here is a list of available sessions, with pointers to Adobe "Flash" format streams. We can make these streams available in a few other formats (notably QuicktimeTM) if requested. Send mail to simon dot leinen at switch dot ch.

-- SimonLeinen - 03 Oct 2007

 

PAM 2008 Call for Papers

The Passive & Active Measurement Conference (PAM 2008) will take place in Cleveland, Ohio on 29/30 April 2008. Contributions must be registered by 14 October, and submitted by 21 October 2007.

Revision 332007-09-24 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PAM 2008 Call for Papers

The Passive & Active Measurement Conference (PAM 2008) will take place in Cleveland, Ohio on 29/30 April 2008. Contributions must be registered by 14 October, and submitted by 21 October 2007.

Successful PERT Training workshop

As announced earlier, SWITCH hosted a training event from 19-21 September. It was attended by about thirty engineers from National Research Networks, Campus IT/Networking divisions and other organizations, in three separate "streams". We will try to make the meeting materials and audio/video recordings available soon.

-- SimonLeinen - 24 Sep 2007

 

Another server move

There hasn't been a lot of activity here over the summer months, for which I apologize. We have been busy preparing the PERT Training Event (see below) - please check it out, it will be fun.

Revision 322007-08-24 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Another server move

There hasn't been a lot of activity here over the summer months, for which I apologize. We have been busy preparing the PERT Training Event (see below) - please check it out, it will be fun.

The Knowledge Base (TWiki server) was moved to a different (newer and faster) machine today, because the old one will be retired when we have to move to different racks at CERN, probably in September, i.e. soon. Hopefully everything continues to work as usual, only faster (it's a nice machine).

Prompted by an interesting post on the DCCP mailing list, I finally started to write some topics on interactive (real-time) applications. Please feel free to add to them or enter suggestions on what you would like improved.

-- SimonLeinen - 24 Aug 2007

 

PERT Training September 19-21: Registration open

The GN2 project is organizing a training event in September. The event will be hosted by SWITCH in Zurich. Preliminary information and a registration page can be found here: http://www.terena.org/activities/training/pert/

Revision 312007-07-10 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PERT Training September 19-21: Registration open

The GN2 project is organizing a training event in September. The event will be hosted by SWITCH in Zurich. Preliminary information and a registration page can be found here: http://www.terena.org/activities/training/pert/

-- SimonLeinen - 10 Jul 2007

 

Internet² Spring 2007 Member Meeting

I haven't been to the Internet² meeting in late April, but started to browse through some of the presentations/Webcasts. Joe St. Sauver's talk about Capacity Planning and System and Network Security is certainly relevant to network performance. It mentions a new network tuning tool for Mac OS X, which will be documented here shortly.

Revision 302007-05-15 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

Materials from ICCRG and High-speed TCP meetings published

>
>

Internet² Spring 2007 Member Meeting

I haven't been to the Internet² meeting in late April, but started to browse through some of the presentations/Webcasts. Joe St. Sauver's talk about Capacity Planning and System and Network Security is certainly relevant to network performance. It mentions a new network tuning tool for Mac OS X, which will be documented here shortly.

Knowledge Base administrativia

Emergency server move

The server that has been hosting the PERT KB wiki since its inception has finally started to break down a week ago. We had to move the wiki to another server, which should be transparent to users (the new server is actually a faster machine). We also started the purchase process for getting a new server for the wiki and other tools that had traditionally been running on the same host (Looking Glass, NDT etc.). Looks like this will be a nice eight-core Intel box.

Making the PERT KB more popular and more useful

The PERT KB sees pretty stable usage, mostly from search engine referrals. Over the past few months, the KB was mentioned twice on networking fora, once on the NANOG mailing list and once on the OpenSolaris networking forum. Both these postings left clear traces in the access statistics, with significantly higher-than-usual accesses over a couple of days. So, if you find a nice occasion to mention the PERT KB - or a particular topic - please don't hesitate to do so. Please use URLs of the "canonical" form http://kb.pert.switch.ch/PERTKB/TopicName when doing so.

It would also be great if you added links to the PERT KB - or specific topics - to Web pages that you or your organization maintains that are related to network performance. This will improve our ranking in search engines. And there is room for improvement; when you type "PERT Team" into Google, the first hit is the Psychiatric Emergency Response Team at the San Diego Sheriff's office. We don't even make it on the first page of search results!

In the same vein, if you have any ideas on how to improve the PERT KB - better organization, missing topics, anything really - you can either contact me via e-mail, or simply edit your suggestions into the Wiki at any suitable place. Someone will notice the changes and act upon them!

-- SimonLeinen - 15 May 2007

Materials from ICCRG and High-speed TCP meetings published

  There were two other meetings around the PFLDnet meeting (see below), and the slides from these meetings were just put up on the Web:

Revision 292007-03-07 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Materials from ICCRG and High-speed TCP meetings published

There were two other meetings around the PFLDnet meeting (see below), and the slides from these meetings were just put up on the Web:

  • The IRTF ICCRG (Internet Congestion Control Research Group) meeting - slides, minutes
  • A workshop on high-speed TCP modifications organized by Microsoft Research - slides

-- SimonLeinen - 07 Mar 2007

 

PFLDnet 2007 fallout

Before Chris went to PFLDnet, I had him promise to add interesting stuff to the knowledge base when he's back... which he has now started to do. Check out the TcpHighSpeedVariants section, which has new sections about YeahTcp and TcpFusion, as well as an overdue one on CompoundTCP, which is in Microsoft Vista. Hopefully more topics will come soon, but don't let that stop you from proofreading those that exist...

Revision 282007-02-26 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PFLDnet 2007 fallout

Before Chris went to PFLDnet, I had him promise to add interesting stuff to the knowledge base when he's back... which he has now started to do. Check out the TcpHighSpeedVariants section, which has new sections about YeahTcp and TcpFusion, as well as an overdue one on CompoundTCP, which is in Microsoft Vista. Hopefully more topics will come soon, but don't let that stop you from proofreading those that exist...

-- SimonLeinen - 26 Feb 2007

 

PFLDnet 2007, Web100 for Linux 2.6.20, Sun 10GE card

After a week of vacation, I found lots of exciting news.

Line: 26 to 32
  -- SimonLeinen - 06 Feb 2007
Changed:
<
<

New TWiki software

I'm in the process of upgrading the TWiki software to the new 4.0.5 release. Since this is a major update, things are in a somewhat broken stage right now. I'm trying to make the wiki at least usable again. The major purpose of this entry, aside from warning users, is to check whether topics can be updated at all. OK, so this works. The new TWiki software has a "WYSIWYG" (what you see is what you get) editing feature. While I'm sure that this will appeal to many people, I'm slightly worried that its use could break consistency of layout compared to the old ASCII-based markup. For fixing typos etc. the WYSIWYG feature is probably great, but maybe not so much for adding new topics and paragraphs.

-- SimonLeinen - 17 Dec 2006

Linux post-2.6.19 changes

As always, after a new Linux kernel release (see the last entry), there is a flurry of changes that have been queued while the previous release was "frozen". I already noticed a few changes that are relevant for network performance. The drivers for the Intel PRO/1000 family of (Gigabit) Ethernet adapters were improved, including new dynamic interrupt throttling modes for Interrupt Coalescence (or Interrupt Moderation). The e1000 driver will also support IPv6 TSO. All adapters from Chelsio should now be supported by the stock kernel, although the TOE functions will still require Chelsio's proprietary driver. A new family of Gigabit and 10Gb Ethernet adapters from NetXen is now supported.

Also, the TCP Vegas implementation was slightly modified to better cope with delayed ACKs.

-- SimonLeinen - 3 Dec 2006

Linux 2.6.19 Release

Linux 2.6.19 was released today. It includes the fixes to H-TCP and CUBIC mentioned below. I found another interesting change related to network performance, and TSO in particular: John Heffner had sent a patch to the netdev mailing list under the subject of Bound TSO defer. He had observed that TSO can make traffic more bursty, in particular over slow links. The patch should reduce the burstiness.

Other features in 2.6.19 include new filesystems such as ext4, GFS2, and eCryptfs.

In addition, normal (non-root) users can now access some information using ethtool.

-- SimonLeinen - 2 Dec 2006

Fixes for H-TCP and CUBIC to Linux kernel tree

Two fixes were applied to Linus' kernel sources yesterday:

  1. For a possible integer overflow with H-TCP at rates over 500 Mb/s
  2. For a scaling-related math error in CUBIC

-- SimonLeinen - 27 Oct 2006

Linux default TCP congestion control algorithm changed from BIC to CUBIC

Shortly after the release of the 2.6.18 kernel, there was a huge flurry of changes being integrated for the next release (2.6.19). Two of these changes concern TCP congestion control:

  1. The default congestion control algorithm will be changed from BIC to CUBIC
  2. The default congestion control algorithm will be more deterministic (independent of module load order), and selectable during kernel configuration. The current configuration choices include Bic, Cubic, Htcp, Vegas, Westwood, and Reno.

-- SimonLeinen - 27 Sep 2006

ABC (TCP Appropriate Byte Counting) default changed in Linux 2.6.18 kernel

According to a change note in Linus' kernel source management system, the default for TCP ABC ( RFC3465 Appropriate Byte Counting) has been changed from on to off. This change was integrated in the 2.6.18 release. The kernel patch includes an update to the documentation for the kernel option, which concisely explains what ABC is about.

The reason that was given for changing the default to off is that ABC would "unfairly penalize[...] applications that do small writes". Well, I'm always wary when I hear about "fairness" in connection with TCP, so I cannot judge the merits of the (three-character smile code change. But the documentation change definitely looks like an improvement!

OpenSolaris DTrace Network Provider

DTrace (Dynamic Tracing) is an operating system facility that can be used to "instrument" software systems for measurement and diagnosis without modifying their code. It includes the "D" programming language and provides dynamic measurement points from different "providers", which can be in kernel components, run-time libraries, or separate run-time systems such as the Java VM.

DTrace was initially implemented in Sun's Solaris 10 system, but has been (at least in part) ported to FreeBSD, and its integration announced into a future version of MacOS X. (Linux has KProbe, which is also a dynamic instrumentation tool, but apparently limited to the kernel.)

Most DTrace usage so far has focused on traditional application and OS issues such as virtual memory, file I/O, CPU, and lock contention issues. But in principle it would be very attractive for network performance debugging, in particular because it can facilitate measurements over several "layers" fairly seamlessly.

The proposal for a new Network DTrace provider has been posted to the DTrace community discussion site on OpenSolaris.org with a call for feedback. Please consider to read the proposal and post your comments from a network performance worker's point of view!

-- SimonLeinen - 19 Sep 2006

NDT 3.3.12 Release

Last Wednesday new version of the NetworkDiagnosticTool (NDT) can be downloaded. This version is the product of a "Google Summer of Code" project awarded to Jakub Slawinski. The announcement can be found in the ndt-announce mailing list archive. Notable new features include IPv6 support, and a distribution of the server side (including Web100-enhanced Linux 2.6.17 kernel) as a bootable Live-CD based on the popular Knoppix system.

I have tested the software on a separate server at SWITCH last week, and today I decided to upgrade our "production" NDT server on ndt.switch.ch to the new version. The applet user interface definitely looks nicer than with the old version!

IPv6 support does work, although I haven't managed to use it from an applet inside a browser, although I am 100% certain that both my browser - I usually use Mozilla 1.7 on Linux and Solaris - and my Java runtime (1.5 something) have IPv6 support. The web100clt command-line client uses IPv6 by default, and I included an example in the NetworkDiagnosticTool topic. When I download the applet code ( Tcpbw100.jar) and run it locally as a Java application with the right options ( java -Djava.net.preferIPv6Addresses=true -jar Tcpbw100.jar ndt.switch.ch), it will do an IPv6 measurement. But inside the browser I'll always get IPv4 measurements. I still need to find out why. At any rate, it would be nice if the applet noticed when both IPv4 and IPv6 are available, and give the user the choice.

-- SimonLeinen - 28 Aug 2006

DS3.3.3 (PERT Performance Guides) is out

GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!)

Calls for Papers: PFLDnet 2007, PAM 2007

The Call for Papers for PFLDnet 2007 just came in. The conference will be on 7-8 February, 2007 in Marina del Rey, CA (US). I added a note to the PFLDnet section in the TcpHighSpeedVariants topic. Submission deadline is 13 October, 2006.

Another conference of interest is the Passive and Active Measurement Conference (PAM). PAM2007 will take place in Louvain-la-Neuve (BE) on 5-6 April, 2007. The Call for Papers is also out for this one. Registration deadline is 5 October, 2006, with the full papers due one week later (same day as PFLDnet).

-- SimonLeinen - 22 Aug 2006

IP Journal article about Gigabit TCP; TCP Westwood+; Joint Techs

Gigabit TCP article in the IP Journal

I was at the IETF meeting in Montréal last week, where the latest issue of the IP Journal was handed out. It has an article on Gigabit TCP by Geoff Huston, with significant review from Larry Dunn. So it must be good - I haven't yet had the time to read the article, but it has interesting graphs about the congestion control behavior of different TCP variants. Seems to be very helpful for those of us who want to be able to understand the many different methods that have been proposed in the past. See the TcpHighSpeedVariants topic for a pointer to the article.

TCP Westwood+

On the end2end-interest mailing list, there was an announcement of a patch for the Linux kernel to implement the TCP Westwood+ variant developed at the Politecnico de Bari. I have added a new topic WestwoodPlusTCP under TcpHighSpeedVariants.

Internet2 Joint Techs Meeting: 16-19 July 2006

The Joint Techs Meeting is going on these days in Madison, Wisconsin. It is being streamed in several formats. In particular, there's a HD video version consisting of a 20 Mb/s MPEG-2 stream over IPv4 multicast. Try to receive it - it should be an excellent test of your LAN infrastructure and host processing power. Channel information is on the abovementioned streaming page: http://winmedia.internet2.edu/jtmadison2006

Incidentally as I am typing this, someone is presenting measurement tools that are very relevant to our work (e.g. Thrulay). Unfortunately I cannot receive the HD stream because I am at home. The low-bandwidth stream has decent quality though.

-- SimonLeinen - 18 Jul 2006

Recent Linux kernel performance improvements

GSO (Generic Segmentation Offload)

Herbert Xu explains GSO (Generic Segmentation Offload) in a post to the netdev@kernel.vger.org mailing list. Before GSO, Linux had separate support for LargeSendOffloadLSO for TCP (called "TSO" or TCP Segmentation Offload) and for UDP (called "UFO" or UDP Fragmentation Offload), but only for devices that have hardware "offloading" support for these features.

The idea behind GSO seems to be that many of the performance benefits of LSO (TSO/UFO/...) can be obtained in a hardware-independent way, by passing large "superpackets" around for as long as possible, and deferring segmentation to the last possible moment - for devices without hardware segmentation/fragmentation support, this would be when data is actually handled to the device driver; for devices with hardware support, it could even be done in hardware.

The GSO code is being added to the Linux kernel tree between 2.6.17 and 2.6.18.

TSO/ ExplicitCongestionNotification conflicts being resolved?

Apparently, the current Linux kernel implementation disables TSO when ECN is used. The move to GSO could provide a good opportunity to lift this restriction.

TCP Segmentation Offload over IPv6 for tg3

In addition, a patch for TSO support for TCP over IPv6 for the tg3 driver has been committed today, despite David Miller's warnings about the difficulties of implementing TSO for IPv6.

-- SimonLeinen - 04 Jul 2006

Changes for SC Bandwidth Challenge

The SC06 conference sent their June 2006 newsletter last night, including a note about their Bandwidth Challenge, which they seem to be attempting to make more realistic every year:

"SC06 Bandwidth Challenge: End-to-End Achievement

For six years, the Bandwidth Challenge has been an exciting and engrossing activity at SC conferences. This year the Bandwidth Challenge will focus on a different important facet of networking: End-to-End achievement. Can you fully utilize one 10 Gig path, end-to-end, disk-to-disk, from SC06 in Tampa back to your home institution, using the actual production network back home? Can you realize, demonstrate and publish all of the configuration, troubleshooting, tuning and policies, not only to show off at SC06, but to leave a legacy at your home institution whereby your scientists can achieve the same results after you? This is a decidedly different slant from the previous Bandwidth Challenge competitions, but one that is well worth embracing and will prove both challenging and inspiring!"

More information: http://sc06.supercomp.org/conference/hpc_bandwidth

This Web page mentions that for the next show, there should be a lower number of dedicated "lambdas" into the conference venue, the idea being that users share the existing research networking infrastructure to get there.

-- SimonLeinen - 21 Jun 2006

New York Times mentioning propagation delay and its impact on performance

The Technology section of the New York Times ran an article about huge new data centers being built by Google and other companies, Hiding in Plain Sight, Google Seeks More Power. The article mentioned that Google in particular is distributing their servers globally because of speed-of-light based delay:

"Google has found that for search engines, every millisecond longer it takes to give users their results leads to lower satisfaction. So the speed of light ends up being a constraint, and the company wants to put significant processing power close to all of its users."

-- SimonLeinen - 15 Jun 2006

Slashdot topic: "ISPs Offer Faster Speeds, Why Don't We Get Them?"

This is an interesting topic in general, and although this is mostly about commercial issues such as "truth in advertising", some of the discussion might be relevant for PERT/end-to-end performance work. Here's the catchphrase from the topic introduction:

"my grandmother signed up for the 3Mbps DSL plan through Verizon, however a speed test said she was only getting 750Kbps"

My current personal pet peeve is that we should stop using "bandwidth" as the indicator for performance (or even "connectivity", see e.g. slide 11 of David West's talk about DANTE's intercontinental activities at TNC'06).

-- SimonLeinen - 2 Jun 2006

Xtrace

While randomly surfing the Internet, I stumbled over the new RAD Lab at U.C. Berkeley. The Lab is holding their summer retreat as I write this, and one of the talks today is about Xtrace: a Cross-layer network trace tool. This seems to study ways of providing performance (and other) instrumentation over multiple nodes and multiple protocol layers. In a somewhat researchy stage, but certainly relevant to PERT/end-to-end performance work!

(The talk starts in about six hours, so hurry if you want to participate. I don't know where it is, but it seems to be a 4-5 hour bus ride from Berkeley. Also the meeting is called a "retreat", so it is probably not public. smile

-- SimonLeinen - 1 Jun 2006

Discussions at TNC 2006

I'm back from the TERENA Networking Conference (TNC 2006) in Catania, Italy. Had some interesting discussions there about NetworkBufferSizing. Many backbone operators are afraid of buying routers with small buffers. I have been, too, but now I have convinced myself that with the money we save by buying less-buffered routers, we should be able to provision enough bandwidth so that queueing will never be a problem...

nepim (Network Pipemeter)

Found the announcement for a new tool called nepim or "Network Pipemeter" in the comp.protocols.tcp-ip USENET newsgroup. This seems to be similar to Iperf - can someone look at it?

-- SimonLeinen - 20 May 2006

Back from the IETF meeting. The pmtud minutes had some information about Linux plans for implementing the new PathMTU proposal, which I noted in that topic.

-- SimonLeinen - 27 Mar 2006

Linux 2.6.16 came out today. Its BIC TCP implementation was changed to use the cubic window growth function suggested in CUBIC. Added a note to TcpHighSpeedVariants.

Linux 2.6.16 also adds a new random-packet-corruption feature to netem. Added a note to the NetEm topic, and cleaned up the text a bit.

-- SimonLeinen - 20 Mar 2006

The deadline for the second release of the "performance guides" deliverable (DS3.3.2v2) is drawing close - it should be ready for internal review by the end of March. In preparation, I made some modifications to this Knowledge Base, in particular surrounding the traceroute family of tools, where there was some duplication as well as some missing information. There are still gaps that I'd like others to work on, for example some demos of Windows tools like PingPlotter or PathPing, as well as more introductory text on when and how to use traceroute, its limitations, etc.

I'd also get rid of the GeneralTools topic, which I consider a catch-all category that can only confuse readers. By moving traceroute out of it, it's already much smaller. The remaining text on "ping" could be moved to a separate topic, maybe under "active measurement" (but I'm not thrilled about that category either :-), and the other small sections on Linux-specific end-system monitoring and configuration tools could be moved to LinuxOSSpecific.

-- SimonLeinen - 28 Feb 2006

ACM Queue has started to publish " QueueCasts" on their Web server. These are interviews in the form of "podcasts", i.e. MP3 files that you can put on your portable music player and listen to while on the bus etc. I found the interview with Jarod Jensen, " Large Scale Systems: Best Practices", particularily interesting from a network performance point of view. Jarod talks about modern system instrumentation such as Sun's DTrace, the general problem of "finger-pointing", and many other performance issues in the context of large distributed systems. On the ACM Queue site, there is also a text version of an interview with Jarod Jensen by Kirk McCusick, " A Conversation with Jarod Jenson".

-- SimonLeinen - 17 Feb 2006

Continued my investigation of Van Jacobson's networking rearchitecture ideas (see yesterday's entry). Added the pointers to the VanJacobson topic, and sent them to Jamal Hadi Salim, who is favorably mentioned in Van's slides.

-- SimonLeinen - 04 Feb 2006

Today there was quite a bit of traffic on the IRTF End-to-end mailing list on comparative evaluation of TcpHighSpeedVariants, including a somewhat heated discussion between Doug Leith and Injong Rhee, who were both posting from PFLDnet 2006 in Tokyo. This prompted me to look at Injong et al.'s A step toward realistic evaluation of high-speed TCP protocols again (already referenced from TcpHighSpeedVariants), and print it out. While a was at the site, I surfed up to the BIC TCP Web page and noticed a broken reference to the abovementioned comparison paper, which I pointed out to Injong (who was probably on his way back from Tokyo).

Surfing on from there I somehow stumbled over Dave S. Miller's (DaveM) blog entry about a talk at LCA 2006 by Van Jacobson on rearchitecting device drivers and buffer management for networking. Very intriguing. Googled through the Blogosphere until I found the slides of this presentation.

-- SimonLeinen - 03 Feb 2006

TobyRodwell suggested to add a "Latest News" section to the PERT Knowledge Base, so let's try this.

>
>

News Archive:

 
Deleted:
<
<
-- SimonLeinen - 31 Jan 2006
 \ No newline at end of file
Added:
>
>

Revision 272007-02-20 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

PFLDnet 2007, Web100 for Linux 2.6.20, Sun 10GE card

After a week of vacation, I found lots of exciting news.

  • My colleague Chris is back from the PFLDnet workshop and brought the proceedings. From a first quick browse, this seemed to include lots of relevant stuff. I hope Chris and others will help me add pointers to the PERT KB...
  • The Web100 patch has been updated for 2.6.20 already (presumably on John Heffner's flight back from PFLDnet :-). I built a new kernel and installed it on a test machine that Chris Rapier from PSC wants to use for work on his HPC SSH patch
  • Sun announced a new dual-port 10GE adapter with PCI Express 1.1 (x8) host interface. It seems to have interesting on-board demultiplexing logic including support for many parallel DMA channels. Probably very useful for a busy server with many (not necessary so fast...) CPUs. Maybe also useful for "Land-Speed Record" stuff because it also supports "bonding" of the two channels to a 20 Gb/s logical channel.

-- SimonLeinen - 20 Feb 2007

 

Linux kernel 2.6.20 is out

Past Sunday, the new Linux kernel release came out. There are a few improvements in the networking area, notably

Revision 262007-02-06 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Linux kernel 2.6.20 is out

Past Sunday, the new Linux kernel release came out. There are a few improvements in the networking area, notably

  • support for NetXen GE and 10GE adapters
  • TCP congestion control choice for mere mortals, restricted by system-wide configuration
  • SACK fixes
  • Hardware TSO for IPv6 on Intel e1000 adapters
  • Support for more Chelsio 10GE cards

You probably don't want to read the entire 47290-line ChangeLog, but there's a nice concise list of changes on the kernelnewbies.org wiki.

-- SimonLeinen - 06 Feb 2007

 

New TWiki software

I'm in the process of upgrading the TWiki software to the new 4.0.5 release. Since this is a major update, things are in a somewhat broken stage right now. I'm trying to make the wiki at least usable again. The major purpose of this entry, aside from warning users, is to check whether topics can be updated at all. OK, so this works. The new TWiki software has a "WYSIWYG" (what you see is what you get) editing feature. While I'm sure that this will appeal to many people, I'm slightly worried that its use could break consistency of layout compared to the old ASCII-based markup. For fixing typos etc. the WYSIWYG feature is probably great, but maybe not so much for adding new topics and paragraphs.

Revision 252006-12-17 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

New TWiki software

I'm in the process of upgrading the TWiki software to the new 4.0.5 release. Since this is a major update, things are in a somewhat broken stage right now. I'm trying to make the wiki at least usable again. The major purpose of this entry, aside from warning users, is to check whether topics can be updated at all. OK, so this works. The new TWiki software has a "WYSIWYG" (what you see is what you get) editing feature. While I'm sure that this will appeal to many people, I'm slightly worried that its use could break consistency of layout compared to the old ASCII-based markup. For fixing typos etc. the WYSIWYG feature is probably great, but maybe not so much for adding new topics and paragraphs.

-- SimonLeinen - 17 Dec 2006

 

Linux post-2.6.19 changes

As always, after a new Linux kernel release (see the last entry), there is a flurry of changes that have been queued while the previous release was "frozen". I already noticed a few changes that are relevant for network performance. The drivers for the Intel PRO/1000 family of (Gigabit) Ethernet adapters were improved, including new dynamic interrupt throttling modes for Interrupt Coalescence (or Interrupt Moderation). The e1000 driver will also support IPv6 TSO. All adapters from Chelsio should now be supported by the stock kernel, although the TOE functions will still require Chelsio's proprietary driver. A new family of Gigabit and 10Gb Ethernet adapters from NetXen is now supported.

Line: 83 to 89
 

Gigabit TCP article in the IP Journal

Changed:
<
<
I was at the IETF meeting in Montréal last week, where the latest issue of the IP Journal was handed out. It has an article on Gigabit TCP by Geoff Huston, with significant review from Larry Dunn. So it must be good - I haven't yet had the time to read the article, but it has interesting graphs about the congestion control behavior of different TCP variants. Seems to be very helpful for those of us who want to be able to understand the many different methods that have been proposed in the past. See the TcpHighSpeedVariants topic for a pointer to the article.
>
>
I was at the IETF meeting in Montréal last week, where the latest issue of the IP Journal was handed out. It has an article on Gigabit TCP by Geoff Huston, with significant review from Larry Dunn. So it must be good - I haven't yet had the time to read the article, but it has interesting graphs about the congestion control behavior of different TCP variants. Seems to be very helpful for those of us who want to be able to understand the many different methods that have been proposed in the past. See the TcpHighSpeedVariants topic for a pointer to the article.
 

TCP Westwood+

Line: 107 to 113
  The GSO code is being added to the Linux kernel tree between 2.6.17 and 2.6.18.
Changed:
<
<

TSO/ExplicitCongestionNotification conflicts being resolved?

>
>

TSO/ ExplicitCongestionNotification conflicts being resolved?

  Apparently, the current Linux kernel implementation disables TSO when ECN is used. The move to GSO could provide a good opportunity to lift this restriction.
Line: 145 to 151
  "my grandmother signed up for the 3Mbps DSL plan through Verizon, however a speed test said she was only getting 750Kbps"
Changed:
<
<
My current personal pet peeve is that we should stop using "bandwidth" as the indicator for performance (or even "connectivity", see e.g. slide 11 of David West's talk about DANTE's intercontinental activities at TNC'06).
>
>
My current personal pet peeve is that we should stop using "bandwidth" as the indicator for performance (or even "connectivity", see e.g. slide 11 of David West's talk about DANTE's intercontinental activities at TNC'06).
  -- SimonLeinen - 2 Jun 2006
Line: 183 to 189
  -- SimonLeinen - 28 Feb 2006
Changed:
<
<
ACM Queue has started to publish "QueueCasts" on their Web server. These are interviews in the form of "podcasts", i.e. MP3 files that you can put on your portable music player and listen to while on the bus etc. I found the interview with Jarod Jensen, "Large Scale Systems: Best Practices", particularily interesting from a network performance point of view. Jarod talks about modern system instrumentation such as Sun's DTrace, the general problem of "finger-pointing", and many other performance issues in the context of large distributed systems. On the ACM Queue site, there is also a text version of an interview with Jarod Jensen by Kirk McCusick, "A Conversation with Jarod Jenson".
>
>
ACM Queue has started to publish " QueueCasts" on their Web server. These are interviews in the form of "podcasts", i.e. MP3 files that you can put on your portable music player and listen to while on the bus etc. I found the interview with Jarod Jensen, " Large Scale Systems: Best Practices", particularily interesting from a network performance point of view. Jarod talks about modern system instrumentation such as Sun's DTrace, the general problem of "finger-pointing", and many other performance issues in the context of large distributed systems. On the ACM Queue site, there is also a text version of an interview with Jarod Jensen by Kirk McCusick, " A Conversation with Jarod Jenson".
  -- SimonLeinen - 17 Feb 2006

Revision 242006-12-10 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Line: 8 to 8
  Also, the TCP Vegas implementation was slightly modified to better cope with delayed ACKs.
Added:
>
>
-- SimonLeinen - 3 Dec 2006
 

Linux 2.6.19 Release

Linux 2.6.19 was released today. It includes the fixes to H-TCP and CUBIC mentioned below. I found another interesting change related to network performance, and TSO in particular: John Heffner had sent a patch to the netdev mailing list under the subject of Bound TSO defer. He had observed that TSO can make traffic more bursty, in particular over slow links. The patch should reduce the burstiness.

Line: 16 to 18
  In addition, normal (non-root) users can now access some information using ethtool.
Added:
>
>
-- SimonLeinen - 2 Dec 2006
 

Fixes for H-TCP and CUBIC to Linux kernel tree

Two fixes were applied to Linus' kernel sources yesterday:

Revision 232006-12-03 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Linux post-2.6.19 changes

As always, after a new Linux kernel release (see the last entry), there is a flurry of changes that have been queued while the previous release was "frozen". I already noticed a few changes that are relevant for network performance. The drivers for the Intel PRO/1000 family of (Gigabit) Ethernet adapters were improved, including new dynamic interrupt throttling modes for Interrupt Coalescence (or Interrupt Moderation). The e1000 driver will also support IPv6 TSO. All adapters from Chelsio should now be supported by the stock kernel, although the TOE functions will still require Chelsio's proprietary driver. A new family of Gigabit and 10Gb Ethernet adapters from NetXen is now supported.

Also, the TCP Vegas implementation was slightly modified to better cope with delayed ACKs.

 

Linux 2.6.19 Release

Linux 2.6.19 was released today. It includes the fixes to H-TCP and CUBIC mentioned below. I found another interesting change related to network performance, and TSO in particular: John Heffner had sent a patch to the netdev mailing list under the subject of Bound TSO defer. He had observed that TSO can make traffic more bursty, in particular over slow links. The patch should reduce the burstiness.

Revision 222006-11-30 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Linux 2.6.19 Release

Linux 2.6.19 was released today. It includes the fixes to H-TCP and CUBIC mentioned below. I found another interesting change related to network performance, and TSO in particular: John Heffner had sent a patch to the netdev mailing list under the subject of Bound TSO defer. He had observed that TSO can make traffic more bursty, in particular over slow links. The patch should reduce the burstiness.

Other features in 2.6.19 include new filesystems such as ext4, GFS2, and eCryptfs.

In addition, normal (non-root) users can now access some information using ethtool.

 

Fixes for H-TCP and CUBIC to Linux kernel tree

Two fixes were applied to Linus' kernel sources yesterday:

Revision 212006-10-27 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

Linux default TCP congestion control algorithm changed from BIC to CUBIC

>
>

Fixes for H-TCP and CUBIC to Linux kernel tree

Two fixes were applied to Linus' kernel sources yesterday:

  1. For a possible integer overflow with H-TCP at rates over 500 Mb/s
  2. For a scaling-related math error in CUBIC

-- SimonLeinen - 27 Oct 2006

Linux default TCP congestion control algorithm changed from BIC to CUBIC

  Shortly after the release of the 2.6.18 kernel, there was a huge flurry of changes being integrated for the next release (2.6.19). Two of these changes concern TCP congestion control:

Revision 202006-09-27 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

ABC (TCP Appropriate Byte Counting) default changed in Linux sources

>
>

Linux default TCP congestion control algorithm changed from BIC to CUBIC

 
Changed:
<
<
According to a change note in Linus' kernel source management system, the default for TCP ABC (RFC3465 Appropriate Byte Counting) has been changed from on to off. This change will probably be integrated into the 2.6.18 release. The kernel patch includes an update to the documentation for the kernel option, which concisely explains what ABC is about.
>
>
Shortly after the release of the 2.6.18 kernel, there was a huge flurry of changes being integrated for the next release (2.6.19). Two of these changes concern TCP congestion control:

  1. The default congestion control algorithm will be changed from BIC to CUBIC
  2. The default congestion control algorithm will be more deterministic (independent of module load order), and selectable during kernel configuration. The current configuration choices include Bic, Cubic, Htcp, Vegas, Westwood, and Reno.

-- SimonLeinen - 27 Sep 2006

ABC (TCP Appropriate Byte Counting) default changed in Linux 2.6.18 kernel

According to a change note in Linus' kernel source management system, the default for TCP ABC (RFC3465 Appropriate Byte Counting) has been changed from on to off. This change was integrated in the 2.6.18 release. The kernel patch includes an update to the documentation for the kernel option, which concisely explains what ABC is about.

  The reason that was given for changing the default to off is that ABC would "unfairly penalize[...] applications that do small writes". Well, I'm always wary when I hear about "fairness" in connection with TCP, so I cannot judge the merits of the (three-character smile code change. But the documentation change definitely looks like an improvement!

Revision 192006-09-19 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

ABC (TCP Appropriate Byte Counting) default changed in Linux sources

According to a change note in Linus' kernel source management system, the default for TCP ABC (RFC3465 Appropriate Byte Counting) has been changed from on to off. This change will probably be integrated into the 2.6.18 release. The kernel patch includes an update to the documentation for the kernel option, which concisely explains what ABC is about.

The reason that was given for changing the default to off is that ABC would "unfairly penalize[...] applications that do small writes". Well, I'm always wary when I hear about "fairness" in connection with TCP, so I cannot judge the merits of the (three-character smile code change. But the documentation change definitely looks like an improvement!

OpenSolaris DTrace Network Provider

DTrace (Dynamic Tracing) is an operating system facility that can be used to "instrument" software systems for measurement and diagnosis without modifying their code. It includes the "D" programming language and provides dynamic measurement points from different "providers", which can be in kernel components, run-time libraries, or separate run-time systems such as the Java VM.

DTrace was initially implemented in Sun's Solaris 10 system, but has been (at least in part) ported to FreeBSD, and its integration announced into a future version of MacOS X. (Linux has KProbe, which is also a dynamic instrumentation tool, but apparently limited to the kernel.)

Most DTrace usage so far has focused on traditional application and OS issues such as virtual memory, file I/O, CPU, and lock contention issues. But in principle it would be very attractive for network performance debugging, in particular because it can facilitate measurements over several "layers" fairly seamlessly.

The proposal for a new Network DTrace provider has been posted to the DTrace community discussion site on OpenSolaris.org with a call for feedback. Please consider to read the proposal and post your comments from a network performance worker's point of view!

-- SimonLeinen - 19 Sep 2006

 

NDT 3.3.12 Release

Last Wednesday new version of the NetworkDiagnosticTool (NDT) can be downloaded. This version is the product of a "Google Summer of Code" project awarded to Jakub Slawinski. The announcement can be found in the ndt-announce mailing list archive. Notable new features include IPv6 support, and a distribution of the server side (including Web100-enhanced Linux 2.6.17 kernel) as a bootable Live-CD based on the popular Knoppix system.

Revision 182006-08-28 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

NDT 3.3.12 Release

Last Wednesday new version of the NetworkDiagnosticTool (NDT) can be downloaded. This version is the product of a "Google Summer of Code" project awarded to Jakub Slawinski. The announcement can be found in the ndt-announce mailing list archive. Notable new features include IPv6 support, and a distribution of the server side (including Web100-enhanced Linux 2.6.17 kernel) as a bootable Live-CD based on the popular Knoppix system.

I have tested the software on a separate server at SWITCH last week, and today I decided to upgrade our "production" NDT server on ndt.switch.ch to the new version. The applet user interface definitely looks nicer than with the old version!

IPv6 support does work, although I haven't managed to use it from an applet inside a browser, although I am 100% certain that both my browser - I usually use Mozilla 1.7 on Linux and Solaris - and my Java runtime (1.5 something) have IPv6 support. The web100clt command-line client uses IPv6 by default, and I included an example in the NetworkDiagnosticTool topic. When I download the applet code (Tcpbw100.jar) and run it locally as a Java application with the right options (java -Djava.net.preferIPv6Addresses=true -jar Tcpbw100.jar ndt.switch.ch), it will do an IPv6 measurement. But inside the browser I'll always get IPv4 measurements. I still need to find out why. At any rate, it would be nice if the applet noticed when both IPv4 and IPv6 are available, and give the user the choice.

-- SimonLeinen - 28 Aug 2006

 

DS3.3.3 (PERT Performance Guides) is out

GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!)

Revision 172006-08-22 - TobyRodwell

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

DS3.3.3 (PERT Performance Guides) is out

Changed:
<
<
GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!
>
>
GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!)
 

Calls for Papers: PFLDnet 2007, PAM 2007

Revision 162006-08-22 - TobyRodwell

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

DS3.3.3 (PERT Performance Guides) is out

Changed:
<
<
GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!
>
>
GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!
 

Calls for Papers: PFLDnet 2007, PAM 2007

Revision 152006-08-22 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

DS3.3.3 (PERT Performance Guides) is out

GN2 project deliverable DS3.3.3 was published today as GN2-06-135v2: PERT Performance Guides. This is sort of a snapshot of this PERT Knowledge Base. Please read it, and contribute to the wiki to make the next release (even smile better!

 

Calls for Papers: PFLDnet 2007, PAM 2007

The Call for Papers for PFLDnet 2007 just came in. The conference will be on 7-8 February, 2007 in Marina del Rey, CA (US). I added a note to the PFLDnet section in the TcpHighSpeedVariants topic. Submission deadline is 13 October, 2006.

Revision 142006-08-22 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Calls for Papers: PFLDnet 2007, PAM 2007

The Call for Papers for PFLDnet 2007 just came in. The conference will be on 7-8 February, 2007 in Marina del Rey, CA (US). I added a note to the PFLDnet section in the TcpHighSpeedVariants topic. Submission deadline is 13 October, 2006.

Another conference of interest is the Passive and Active Measurement Conference (PAM). PAM2007 will take place in Louvain-la-Neuve (BE) on 5-6 April, 2007. The Call for Papers is also out for this one. Registration deadline is 5 October, 2006, with the full papers due one week later (same day as PFLDnet).

-- SimonLeinen - 22 Aug 2006

 

IP Journal article about Gigabit TCP; TCP Westwood+; Joint Techs

Gigabit TCP article in the IP Journal

Revision 132006-07-18 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

IP Journal article about Gigabit TCP; TCP Westwood+; Joint Techs

Gigabit TCP article in the IP Journal

I was at the IETF meeting in Montréal last week, where the latest issue of the IP Journal was handed out. It has an article on Gigabit TCP by Geoff Huston, with significant review from Larry Dunn. So it must be good - I haven't yet had the time to read the article, but it has interesting graphs about the congestion control behavior of different TCP variants. Seems to be very helpful for those of us who want to be able to understand the many different methods that have been proposed in the past. See the TcpHighSpeedVariants topic for a pointer to the article.

TCP Westwood+

On the end2end-interest mailing list, there was an announcement of a patch for the Linux kernel to implement the TCP Westwood+ variant developed at the Politecnico de Bari. I have added a new topic WestwoodPlusTCP under TcpHighSpeedVariants.

Internet2 Joint Techs Meeting: 16-19 July 2006

The Joint Techs Meeting is going on these days in Madison, Wisconsin. It is being streamed in several formats. In particular, there's a HD video version consisting of a 20 Mb/s MPEG-2 stream over IPv4 multicast. Try to receive it - it should be an excellent test of your LAN infrastructure and host processing power. Channel information is on the abovementioned streaming page: http://winmedia.internet2.edu/jtmadison2006

Incidentally as I am typing this, someone is presenting measurement tools that are very relevant to our work (e.g. Thrulay). Unfortunately I cannot receive the HD stream because I am at home. The low-bandwidth stream has decent quality though.

-- SimonLeinen - 18 Jul 2006

 

Recent Linux kernel performance improvements

GSO (Generic Segmentation Offload)

Revision 122006-07-04 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

Recent Linux kernel performance improvements

GSO (Generic Segmentation Offload)

Herbert Xu explains GSO (Generic Segmentation Offload) in a post to the netdev@kernel.vger.org mailing list. Before GSO, Linux had separate support for LargeSendOffloadLSO for TCP (called "TSO" or TCP Segmentation Offload) and for UDP (called "UFO" or UDP Fragmentation Offload), but only for devices that have hardware "offloading" support for these features.

The idea behind GSO seems to be that many of the performance benefits of LSO (TSO/UFO/...) can be obtained in a hardware-independent way, by passing large "superpackets" around for as long as possible, and deferring segmentation to the last possible moment - for devices without hardware segmentation/fragmentation support, this would be when data is actually handled to the device driver; for devices with hardware support, it could even be done in hardware.

The GSO code is being added to the Linux kernel tree between 2.6.17 and 2.6.18.

TSO/ExplicitCongestionNotification conflicts being resolved?

Apparently, the current Linux kernel implementation disables TSO when ECN is used. The move to GSO could provide a good opportunity to lift this restriction.

TCP Segmentation Offload over IPv6 for tg3

In addition, a patch for TSO support for TCP over IPv6 for the tg3 driver has been committed today, despite David Miller's warnings about the difficulties of implementing TSO for IPv6.

-- SimonLeinen - 04 Jul 2006

 

Changes for SC Bandwidth Challenge

The SC06 conference sent their June 2006 newsletter last night, including a note about their Bandwidth Challenge, which they seem to be attempting to make more realistic every year:

Revision 112006-06-21 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Changed:
<
<

SimonLeinen - 15 Jun 2006

>
>

Changes for SC Bandwidth Challenge

The SC06 conference sent their June 2006 newsletter last night, including a note about their Bandwidth Challenge, which they seem to be attempting to make more realistic every year:

"SC06 Bandwidth Challenge: End-to-End Achievement

For six years, the Bandwidth Challenge has been an exciting and engrossing activity at SC conferences. This year the Bandwidth Challenge will focus on a different important facet of networking: End-to-End achievement. Can you fully utilize one 10 Gig path, end-to-end, disk-to-disk, from SC06 in Tampa back to your home institution, using the actual production network back home? Can you realize, demonstrate and publish all of the configuration, troubleshooting, tuning and policies, not only to show off at SC06, but to leave a legacy at your home institution whereby your scientists can achieve the same results after you? This is a decidedly different slant from the previous Bandwidth Challenge competitions, but one that is well worth embracing and will prove both challenging and inspiring!"

More information: http://sc06.supercomp.org/conference/hpc_bandwidth

This Web page mentions that for the next show, there should be a lower number of dedicated "lambdas" into the conference venue, the idea being that users share the existing research networking infrastructure to get there.

-- SimonLeinen - 21 Jun 2006

 

New York Times mentioning propagation delay and its impact on performance

Line: 10 to 22
  "Google has found that for search engines, every millisecond longer it takes to give users their results leads to lower satisfaction. So the speed of light ends up being a constraint, and the company wants to put significant processing power close to all of its users."
Changed:
<
<

SimonLeinen - 2 Jun 2006

>
>
-- SimonLeinen - 15 Jun 2006
 

Slashdot topic: "ISPs Offer Faster Speeds, Why Don't We Get Them?"

Line: 20 to 32
  My current personal pet peeve is that we should stop using "bandwidth" as the indicator for performance (or even "connectivity", see e.g. slide 11 of David West's talk about DANTE's intercontinental activities at TNC'06).
Changed:
<
<

SimonLeinen - 1 Jun 2006

>
>
-- SimonLeinen - 2 Jun 2006
 

Xtrace

Line: 28 to 40
  (The talk starts in about six hours, so hurry if you want to participate. I don't know where it is, but it seems to be a 4-5 hour bus ride from Berkeley. Also the meeting is called a "retreat", so it is probably not public. smile
Changed:
<
<

SimonLeinen - 20 May 2006

>
>
-- SimonLeinen - 1 Jun 2006
 

Discussions at TNC 2006

Line: 38 to 50
  Found the announcement for a new tool called nepim or "Network Pipemeter" in the comp.protocols.tcp-ip USENET newsgroup. This seems to be similar to Iperf - can someone look at it?
Changed:
<
<

SimonLeinen - 27 Mar 2006

>
>
-- SimonLeinen - 20 May 2006
  Back from the IETF meeting. The pmtud minutes had some information about Linux plans for implementing the new PathMTU proposal, which I noted in that topic.
Changed:
<
<

SimonLeinen - 20 Mar 2006

>
>
-- SimonLeinen - 27 Mar 2006
  Linux 2.6.16 came out today. Its BIC TCP implementation was changed to use the cubic window growth function suggested in CUBIC. Added a note to TcpHighSpeedVariants.

Linux 2.6.16 also adds a new random-packet-corruption feature to netem. Added a note to the NetEm topic, and cleaned up the text a bit.

Changed:
<
<

SimonLeinen - 28 Feb 2006

>
>
-- SimonLeinen - 20 Mar 2006
  The deadline for the second release of the "performance guides" deliverable (DS3.3.2v2) is drawing close - it should be ready for internal review by the end of March. In preparation, I made some modifications to this Knowledge Base, in particular surrounding the traceroute family of tools, where there was some duplication as well as some missing information. There are still gaps that I'd like others to work on, for example some demos of Windows tools like PingPlotter or PathPing, as well as more introductory text on when and how to use traceroute, its limitations, etc.

I'd also get rid of the GeneralTools topic, which I consider a catch-all category that can only confuse readers. By moving traceroute out of it, it's already much smaller. The remaining text on "ping" could be moved to a separate topic, maybe under "active measurement" (but I'm not thrilled about that category either :-), and the other small sections on Linux-specific end-system monitoring and configuration tools could be moved to LinuxOSSpecific.

Changed:
<
<

SimonLeinen - 17 Feb 2006

>
>
-- SimonLeinen - 28 Feb 2006
  ACM Queue has started to publish "QueueCasts" on their Web server. These are interviews in the form of "podcasts", i.e. MP3 files that you can put on your portable music player and listen to while on the bus etc. I found the interview with Jarod Jensen, "Large Scale Systems: Best Practices", particularily interesting from a network performance point of view. Jarod talks about modern system instrumentation such as Sun's DTrace, the general problem of "finger-pointing", and many other performance issues in the context of large distributed systems. On the ACM Queue site, there is also a text version of an interview with Jarod Jensen by Kirk McCusick, "A Conversation with Jarod Jenson".
Changed:
<
<

SimonLeinen - 04 Feb 2006

>
>
-- SimonLeinen - 17 Feb 2006
  Continued my investigation of Van Jacobson's networking rearchitecture ideas (see yesterday's entry). Added the pointers to the VanJacobson topic, and sent them to Jamal Hadi Salim, who is favorably mentioned in Van's slides.
Changed:
<
<

SimonLeinen - 03 Feb 2006

>
>
-- SimonLeinen - 04 Feb 2006
  Today there was quite a bit of traffic on the IRTF End-to-end mailing list on comparative evaluation of TcpHighSpeedVariants, including a somewhat heated discussion between Doug Leith and Injong Rhee, who were both posting from PFLDnet 2006 in Tokyo. This prompted me to look at Injong et al.'s A step toward realistic evaluation of high-speed TCP protocols again (already referenced from TcpHighSpeedVariants), and print it out. While a was at the site, I surfed up to the BIC TCP Web page and noticed a broken reference to the abovementioned comparison paper, which I pointed out to Injong (who was probably on his way back from Tokyo).

Surfing on from there I somehow stumbled over Dave S. Miller's (DaveM) blog entry about a talk at LCA 2006 by Van Jacobson on rearchitecting device drivers and buffer management for networking. Very intriguing. Googled through the Blogosphere until I found the slides of this presentation.

Changed:
<
<

SimonLeinen - 31 Jan 2006

>
>
-- SimonLeinen - 03 Feb 2006
  TobyRodwell suggested to add a "Latest News" section to the PERT Knowledge Base, so let's try this.
Added:
>
>
-- SimonLeinen - 31 Jan 2006

Revision 102006-06-15 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 15 Jun 2006

New York Times mentioning propagation delay and its impact on performance

The Technology section of the New York Times ran an article about huge new data centers being built by Google and other companies, Hiding in Plain Sight, Google Seeks More Power. The article mentioned that Google in particular is distributing their servers globally because of speed-of-light based delay:

"Google has found that for search engines, every millisecond longer it takes to give users their results leads to lower satisfaction. So the speed of light ends up being a constraint, and the company wants to put significant processing power close to all of its users."

 

SimonLeinen - 2 Jun 2006

Slashdot topic: "ISPs Offer Faster Speeds, Why Don't We Get Them?"

Revision 92006-06-02 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 2 Jun 2006

Slashdot topic: "ISPs Offer Faster Speeds, Why Don't We Get Them?"

This is an interesting topic in general, and although this is mostly about commercial issues such as "truth in advertising", some of the discussion might be relevant for PERT/end-to-end performance work. Here's the catchphrase from the topic introduction:

"my grandmother signed up for the 3Mbps DSL plan through Verizon, however a speed test said she was only getting 750Kbps"

My current personal pet peeve is that we should stop using "bandwidth" as the indicator for performance (or even "connectivity", see e.g. slide 11 of David West's talk about DANTE's intercontinental activities at TNC'06).

 

SimonLeinen - 1 Jun 2006

Xtrace

Revision 82006-06-01 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 1 Jun 2006

Xtrace

While randomly surfing the Internet, I stumbled over the new RAD Lab at U.C. Berkeley. The Lab is holding their summer retreat as I write this, and one of the talks today is about Xtrace: a Cross-layer network trace tool. This seems to study ways of providing performance (and other) instrumentation over multiple nodes and multiple protocol layers. In a somewhat researchy stage, but certainly relevant to PERT/end-to-end performance work!

(The talk starts in about six hours, so hurry if you want to participate. I don't know where it is, but it seems to be a 4-5 hour bus ride from Berkeley. Also the meeting is called a "retreat", so it is probably not public. smile

 

SimonLeinen - 20 May 2006

Discussions at TNC 2006

Revision 72006-05-21 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 20 May 2006

Discussions at TNC 2006

I'm back from the TERENA Networking Conference (TNC 2006) in Catania, Italy. Had some interesting discussions there about NetworkBufferSizing. Many backbone operators are afraid of buying routers with small buffers. I have been, too, but now I have convinced myself that with the money we save by buying less-buffered routers, we should be able to provision enough bandwidth so that queueing will never be a problem...

nepim (Network Pipemeter)

Found the announcement for a new tool called nepim or "Network Pipemeter" in the comp.protocols.tcp-ip USENET newsgroup. This seems to be similar to Iperf - can someone look at it?

 

SimonLeinen - 27 Mar 2006

Back from the IETF meeting. The pmtud minutes had some information about Linux plans for implementing the new PathMTU proposal, which I noted in that topic.

Revision 62006-03-28 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

SimonLeinen - 27 Mar 2006

Changed:
<
<
Back from the IETF meeting. The pmtud minutes had some information about Linux plans for implementing the new PathMTU proposal, which I noticed in that topic.
>
>
Back from the IETF meeting. The pmtud minutes had some information about Linux plans for implementing the new PathMTU proposal, which I noted in that topic.
 

SimonLeinen - 20 Mar 2006

Revision 52006-03-27 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 27 Mar 2006

Back from the IETF meeting. The pmtud minutes had some information about Linux plans for implementing the new PathMTU proposal, which I noticed in that topic.

 

SimonLeinen - 20 Mar 2006

Linux 2.6.16 came out today. Its BIC TCP implementation was changed to use the cubic window growth function suggested in CUBIC. Added a note to TcpHighSpeedVariants.

Revision 42006-03-21 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 20 Mar 2006

Linux 2.6.16 came out today. Its BIC TCP implementation was changed to use the cubic window growth function suggested in CUBIC. Added a note to TcpHighSpeedVariants.

Linux 2.6.16 also adds a new random-packet-corruption feature to netem. Added a note to the NetEm topic, and cleaned up the text a bit.

 

SimonLeinen - 28 Feb 2006

The deadline for the second release of the "performance guides" deliverable (DS3.3.2v2) is drawing close - it should be ready for internal review by the end of March. In preparation, I made some modifications to this Knowledge Base, in particular surrounding the traceroute family of tools, where there was some duplication as well as some missing information. There are still gaps that I'd like others to work on, for example some demos of Windows tools like PingPlotter or PathPing, as well as more introductory text on when and how to use traceroute, its limitations, etc.

Revision 32006-02-28 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>

SimonLeinen - 28 Feb 2006

The deadline for the second release of the "performance guides" deliverable (DS3.3.2v2) is drawing close - it should be ready for internal review by the end of March. In preparation, I made some modifications to this Knowledge Base, in particular surrounding the traceroute family of tools, where there was some duplication as well as some missing information. There are still gaps that I'd like others to work on, for example some demos of Windows tools like PingPlotter or PathPing, as well as more introductory text on when and how to use traceroute, its limitations, etc.

I'd also get rid of the GeneralTools topic, which I consider a catch-all category that can only confuse readers. By moving traceroute out of it, it's already much smaller. The remaining text on "ping" could be moved to a separate topic, maybe under "active measurement" (but I'm not thrilled about that category either :-), and the other small sections on Linux-specific end-system monitoring and configuration tools could be moved to LinuxOSSpecific.

SimonLeinen - 17 Feb 2006

 ACM Queue has started to publish "QueueCasts" on their Web server. These are interviews in the form of "podcasts", i.e. MP3 files that you can put on your portable music player and listen to while on the bus etc. I found the interview with Jarod Jensen, "Large Scale Systems: Best Practices", particularily interesting from a network performance point of view. Jarod talks about modern system instrumentation such as Sun's DTrace, the general problem of "finger-pointing", and many other performance issues in the context of large distributed systems. On the ACM Queue site, there is also a text version of an interview with Jarod Jensen by Kirk McCusick, "A Conversation with Jarod Jenson".
Changed:
<
<
-- SimonLeinen - 17 Feb 2006
>
>

SimonLeinen - 04 Feb 2006

  Continued my investigation of Van Jacobson's networking rearchitecture ideas (see yesterday's entry). Added the pointers to the VanJacobson topic, and sent them to Jamal Hadi Salim, who is favorably mentioned in Van's slides.
Changed:
<
<
-- SimonLeinen - 04 Feb 2006
>
>

SimonLeinen - 03 Feb 2006

  Today there was quite a bit of traffic on the IRTF End-to-end mailing list on comparative evaluation of TcpHighSpeedVariants, including a somewhat heated discussion between Doug Leith and Injong Rhee, who were both posting from PFLDnet 2006 in Tokyo. This prompted me to look at Injong et al.'s A step toward realistic evaluation of high-speed TCP protocols again (already referenced from TcpHighSpeedVariants), and print it out. While a was at the site, I surfed up to the BIC TCP Web page and noticed a broken reference to the abovementioned comparison paper, which I pointed out to Injong (who was probably on his way back from Tokyo).

Surfing on from there I somehow stumbled over Dave S. Miller's (DaveM) blog entry about a talk at LCA 2006 by Van Jacobson on rearchitecting device drivers and buffer management for networking. Very intriguing. Googled through the Blogosphere until I found the slides of this presentation.

Changed:
<
<
-- SimonLeinen - 03 Feb 2006
>
>

SimonLeinen - 31 Jan 2006

  TobyRodwell suggested to add a "Latest News" section to the PERT Knowledge Base, so let's try this.
Deleted:
<
<
-- SimonLeinen - 31 Jan 2006

Revision 22006-02-17 - SimonLeinen

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Latest News

Added:
>
>
ACM Queue has started to publish "QueueCasts" on their Web server. These are interviews in the form of "podcasts", i.e. MP3 files that you can put on your portable music player and listen to while on the bus etc. I found the interview with Jarod Jensen, "Large Scale Systems: Best Practices", particularily interesting from a network performance point of view. Jarod talks about modern system instrumentation such as Sun's DTrace, the general problem of "finger-pointing", and many other performance issues in the context of large distributed systems. On the ACM Queue site, there is also a text version of an interview with Jarod Jensen by Kirk McCusick, "A Conversation with Jarod Jenson".

-- SimonLeinen - 17 Feb 2006

 Continued my investigation of Van Jacobson's networking rearchitecture ideas (see yesterday's entry). Added the pointers to the VanJacobson topic, and sent them to Jamal Hadi Salim, who is favorably mentioned in Van's slides.

-- SimonLeinen - 04 Feb 2006

Revision 12006-02-05 - SimonLeinen

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="WebHome"

Latest News

Continued my investigation of Van Jacobson's networking rearchitecture ideas (see yesterday's entry). Added the pointers to the VanJacobson topic, and sent them to Jamal Hadi Salim, who is favorably mentioned in Van's slides.

-- SimonLeinen - 04 Feb 2006

Today there was quite a bit of traffic on the IRTF End-to-end mailing list on comparative evaluation of TcpHighSpeedVariants, including a somewhat heated discussion between Doug Leith and Injong Rhee, who were both posting from PFLDnet 2006 in Tokyo. This prompted me to look at Injong et al.'s A step toward realistic evaluation of high-speed TCP protocols again (already referenced from TcpHighSpeedVariants), and print it out. While a was at the site, I surfed up to the BIC TCP Web page and noticed a broken reference to the abovementioned comparison paper, which I pointed out to Injong (who was probably on his way back from Tokyo).

Surfing on from there I somehow stumbled over Dave S. Miller's (DaveM) blog entry about a talk at LCA 2006 by Van Jacobson on rearchitecting device drivers and buffer management for networking. Very intriguing. Googled through the Blogosphere until I found the slides of this presentation.

-- SimonLeinen - 03 Feb 2006

TobyRodwell suggested to add a "Latest News" section to the PERT Knowledge Base, so let's try this.

-- SimonLeinen - 31 Jan 2006

 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2004-2009 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.