HPCwire: Enhanced pS-Performance Toolkit Version Released

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

July 22, 2009

Enhanced pS-Performance Toolkit Version Released

Easy to install suite of tools provides performance metrics, monitoring and diagnostics to significantly improve users' network experience

INDIANAPOLIS, July 21 -- The perfSONAR-PS development team today announced at the ESCC/Internet2 Joint Techs Workshop, the release of an enhanced version of the pS-Performance Toolkit. The pS- Performance Toolkit (formerly called the pS-Network Performance Toolkit) is available as a live CD download, which can be burned to disc and used to turn any Intel-based computer into an active network performance node. Based on feedback from early adopters, the latest release bundles the suite of recommended network performance tools in an easier-to-use package with new functionality. With these tools, researchers and network engineers can not only monitor and debug their own networks, but access information from external networks.

The latest pS-Performance Toolkit release aligns with the project's roadmap toward greater ease of use and more graphical displays. Users will immediately notice a more streamlined startup process, and will be able to use the Toolkit's functionality even before it is configured and customized. The tools are available as soon as the CD boots up, but the configuration process lets users register their performance node with the global perfSONAR infrastructure and locate and share information with other registered performance nodes. It also allows users to set up appropriate access and use policies to ensure the Performance Node (created using the pS-Performance Toolkit) only consumes the amount of network resources and shares the information the administrator deems appropriate. The configuration process is now driven by a wizard-like interface rather than command-line prompts, and typically takes under 30 minutes to complete.

The Toolkit interface has more graphical displays and new administration screens which make it easier to fine-tune configuration settings. Beyond the interface enhancements, significant functionalities have also been added. The most important is OWAMP support, added within perfSONAR-BUOY, which enables regular one-way latency monitoring, a long-standing request from the network research community. Driver enhancements have also been added in response to user requests, so setup is smoother, with more equipment brands supported out of the box and less troubleshooting required.

"The pS-Performance Toolkit provides an easy means for sites to deploy and configure perfSONAR monitoring and measurement tools," said Phil DeMar, head of wide-area networking at Fermi National Accelerator Laboratory (Fermilab). "We expect the Toolkit's ease-of-use factor to facilitate rapid development of a cross-domain monitoring infrastructure in support of our collaborations."

Developed through a collaboration and integration of tools from ESnet, Fermilab, Georgia Institute of Technology, Indiana University, Internet2, the University of Delaware, Pittsburgh Supercomputing Center and SLAC National Accelerator Laboratory, the perfSONAR-enabled Toolkit was first introduced in the summer of 2008 at the ESCC/ Internet2 Joint Techs Workshop and has quickly become a key enabler of several high-profile science projects that depend on high-performance, distributed computing.

Research and Education Data Depot Network (REDDnet), a National Science Foundation-funded infrastructure project, provides distributed storage for data-intensive collaboration among U.S. researchers and educators in a wide variety of application areas, including the Large Hadron Collider (LHC) project. Scientists involved in the LHC CMS or ATLAS experiments use REDDnet "depots" to access their LHC data. In collaboration with Internet2, REDDnet deployed tools from the pS- Performance Toolkit at several of its depots, enabling the diagnosis of sub-optimal network performance problems and the establishment of a long-term monitoring and analysis infrastructure to help guarantee future performance.

Paul Sheldon, physicist with Vanderbilt University and principal investigator for REDDnet, noted the importance of constantly monitoring and debugging network performance to and from the data depots: "The tools in the pS-Performance Toolkit have been extremely valuable in our work. By leveraging these performance technologies, we have been able to better isolate problems on the network and resolve them quickly resulting in better data throughput and better use of our network capacity investments to support our scientific research."

Telepresence initiatives represent another application area especially dependent on optimal network performance. Tools from the pS- Performance Toolkit were used to support a telehealth demonstration leveraging telepresence technology at Internet2's recent Spring Member Meeting. Tools such as perfSONAR-BUOY and BWCTL -- both available via the Toolkit -- showed network engineers the "big picture" of network performance over the long-haul network route, and also helped them isolate network segments that did not meet the service's demanding jitter and latency requirements. The demonstration underlined why the pS-Performance Toolkit will be a critical part of the Internet2 community's implementation plan for telepresence services.

Another major pS-Performance Toolkit release is anticipated by the end of the last quarter, 2009, with even more streamlined user interface and further enhanced graphic visualizations. Additional functionality will be included, and the Toolkit will be installable directly from network accessible software repositories.

To download the pS-Performance Toolkit version 3.1rc1, visit http://www.internet2.edu/performance/toolkit/

Questions and feedback on the pS-Performance Toolkit can be directed to the performance-node-users@internet2.edu email list. Feedback is encouraged from both inside and outside the performance measurement community.

About perfSONAR

perfSONAR is developed through a global collaboration consortium led by ESnet, GÉANT, Internet2 and RNP, and in the U.S. via partnerships between ESnet, Fermi National Accelerator Laboratory, Georgia Institute of Technology, Indiana University, Internet2, the University of Delaware and SLAC National Accelerator Laboratory with support from the U.S. Department of Energy, the National Institutes of Health, and the National Science Foundation. To find out what's ahead on perfSONAR's development roadmap, visit perfsonar.net. For more on Internet2 performance tools and the pS-Performance Toolkit, visit internet2.edu/performance.

-----

Source: Internet2

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 23, 2012

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

May 11, 2012

Feature Articles

NVIDIA Works On CPU Co-Dependency Issues with Kepler GPU

NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...