IEPM

Status Report for SLAC/DOE IEPM Project

For February 2003

SLAC Home Page
Work completed or in progress since the last report.

Work Related to Analysis and Development

IEPM

Fabrizio and Les worked on characterizing how the various TCP stacks and stock TCP behave on our high speed tetsebed. See TCP Stack measurements. As part of this we installed the scalable and HS TCP/web100 stacks at Sunnyvale. We also installed a Linux filtering package to look into the effects of packet loss. Unfortunately due to other pressures (the 10GE testbed) we have not had a chance to play with the filtering yet. Much time was spent by Les negotiating and coordinating loans of 10GE GSR blades from Cisco and loaning a 10GE card from Intel. This was for the 10GE testbed which finally came into operation around the 25th February. Results from the the testbed can be found at 10GE end-to-end TCP tests.
We had meetings with Level(3), CENIC, Stanford, Yotta Yotta, Riverstone and QWest to gather information for a proposal for a high speed testbed for SLAC.

MAGGIE

Warren and Les worked with Vern Paxson (ICIR), Andy Adams (PSC) and Brian Tierney (LBL) to write a pre-proposal for the Measurement and Analysis for the Global Grid and Internet End-to-end performance (MAGGIE) project. kc claffy (CAIDA) provided a letter of support. The paper submitted for review by the DoE program manager is available in DOC or PDF format.

PIPES

Warren talked with Jim Ferguson of NLANR/DAST about their Advisor tool, and worked with the developer to get an advanced copy. Warren is learning the python programming language in order to add a web service client to Advisor and extract PingER, IEPM-BW and other data.

Warren worked with Eric Boyd and Jeff Boote from Internet2 to put a web service front end on their one-way measurement (OWAMP) data. The WSDL is available at http://thunderbird.internet2.edu/~cottrell/wsdl/get_owamp.wsdl. Warren is working on modifying Advisor to access the data.

Warren is attempting to build a similar front end to the NASA iperf data.

Warren is working with the developer of the MonALISA monitoring system to access data from his GLUE based web service.

Warren is working with Yee-Ting Li to write an "Implementation" document.

Warren talked with Tony McGregor about his fault detection work.

ICTP/eJDS

Warren talked to Ana Preston, manager of the Internet2 International group about presenting some results at the "Hard to Network Places" BOF which will take place at the forthcoming Internet2 Members meeting.

Warren contacted the ICTP participants in Nigeria, Guatemala, Ghana and Uganda for information about networking in those countries.

Warren talked with Heather Boyles, Director of the Internet2 International group about networking in Egpyt.

Warren worked with Enrique Canessa on the poster for CHEP, and prepared graphs for Enrique's to prepare the talk for CHEP.

Warren talked with Julio Ibarra, from AMPATH, and Michael Stanton, from RNP, about networking and extending the monitoring in Latin America.

Warren and Les discussed with the team at ICTP about a proposal for funding the 'quantifying the digital divide' work.

Animated GIFS

Warren discussed further development on the performance maps animated GIFs with Andy Germain from NASA.

Globus

Warren reviewed a draft of a document by Yee Ting Li from UCL (at his request) on OGSA compliant Network Monitoring.

NIMI

Warren continued to work with Fabrizio on reviving the NIMI probe at SLAC.

PingER

Warren restored PingER data gathering and collection at LANL and BNL.

Warren worked with the security at JLAB to identify nodes requiring access to the PingER web server and restrict access.

Warren worked on designing a re-write for the PingER software, using modules, for easy installation at the new sites expected to join as part of the ICTP project.

Web100

Warren install the new userland tools release. Fabrizio installed the new net100 private patch.

IPv6

Warren and Paola revived the IPv6 network. ESnet has changed the SLAC allocation to 2001:404:1010::/48.

Warren worked with project co-ordinators to get another AMP box, specifically for the IPv6 network.

Work related to Publicity and Information

Internet2/ESCC

Warren attended the Internet2 Joint Tech Meeting and the end-to-end workshop. Warren made presentations at the end-to-end workshop, the ESCC/SciDAC meeting, and gave a review of the workshop to the group on return to SLAC.

BABAR

Warren and Les worked on a presentation to the data distribution working group during the BaBar collaboration meeting at SLAC. PPT, PDF or HTML.

CHEP

The papers and poster for the Computing in High Energy Physics (CHEP) conference were accepted.

Warren arranged with the travel office to attend the conference.

Warren arranged meetings with kc claffy of CAIDA and Kevin Walsh of San Diego SuperComputing Center while at CHEP.

PAM 2003

We had 4 papers accepted to PAM03, Jiri & Les Available Bandwidth Estimation, Ajay, Tom Dunnigan & Les on Quick Iperf, Les, Connie and Heng on Experiences and Results from a New High Performance Network and Application Monitoring Toolkit and Jiri and Les and the INCITE team on Pathchirp.

Land Speed Record

Les spent much time talking to reporters and preparing a SLAC publicty release and web page on the Land Speed Record. There are articles in English, French, Portuguese, Spanish, French and Dutch. In Neil Calder's absence talked to journalists from Wired, the Stanford Report and the Stanford Daily.

Internet2 Spring Meeting

Warren arranged with the travel office to attend the Internet2 Spring Members Meeting.

Warren arranged to make presentations at the 'hard to network places' BOF and for the PIPEfitters meeting. Warren also arranged to present the poster being prepared for CHEP.

ESnet

Connie, Paola and Les worked on a report for Bill Wing of the ESCC characterizing SLAC's ESnet traffic to show how sciences use networking. We also worked on characterizing the effects of the slammer worm.

GGF

The Network Monitoring Working Group (NMWG) of which Les is a member finished their draft of "A Hierarchy of Network Measurements for Grid Applications and Services," to be discussed at GGF7 and submitted it.

Europe

Les traveled to Europe and made presentations on High Performance Active End-to-end Network Monitoring, at the Protocols for Long Distance Networks Workshop, CERN, February 2003, and the SCAMPI workshop in Amsterdam. at the SCAMPI

Student

Warren talked with 2 students about summer jobs at SLAC.

Created February 3, 2003. Last Updated February 28, 2003.
Comments to iepm-l@slac.stanford.edu