IEPM

Status Report for SLAC/DOE IEPM Project

For December & January 2001

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

Work related to Analysis and Collaborations

QoS

The project collaborators at Daresbury Lab created a mailing list. Warren installed the pingER software on loggy. loggy had some hardware problems that have been temporarily fixed by the Stanford administrators.

ping_data_plot.pl

Les modified ping_data_plot.pl to insert today's date as default into the date selection tables. Sent mod to Robin Tasker of Daresbury and it is now incorporated in the production version.

Passive Analysis

Warren installed another passive monitoring package called Cflowd and FlowScan, an analysis package from the University of Wisconsin.

Warren installed a new 18G hard drive on the passive monitoring machine to enable around 9 months of data to be stored.

Web100

Doug reconfigured epahus and reinstalled the web100 kernel patch. Work is under way to modify a SLAC file-transfer program to assess the promise of web100 to improve throughput.

Jplot

Imran worked on a java version of the xplot program. The program uses classes from java2 and only works with the very latest browsers such as netscape6.

Netpredict

Doug and Les met with Bjorn Frogner to discuss Doug's review of the Netpredict software.

IPv6

Warren installed IPv6 on a machine running Windows2000 and was pleased to find the explorer web browser was already Ipv6-aware.

Warren reported a routing problem to ESnet. It appears the Canadian CRC is multi-homed and some european networks and mistakenly advertising the CRC AS.

Warren contacted the Viagenie IPv6 group about re-establishing an account on their machine at the 6TAP after it was relocated.

Upgrades

Warren upgraded the Tiresias machine to FreeBSD 4.2-REL with an eye to gaining detailed knowledge of the NIMI design.

Warren installed Netscape6 and the VRVS video conferencing system.

Data

A few occurences of lost data this month, but less than usual except for DESY which seemed to be down for over a week.

Warren retrieved the data from Surveyor, the QoS project with Darebury Lab and the IPv6 monitoring for December and ran them all through the analysis code.

Warren worked with Raghu Kacker from NIST to resolve a problem they were having understanding timestamps and Warren obtained an account at CNRI to access the XIWT database. The Mathsoft developer working on the statistical analysis of the XIWT data for NIST contacted Les and Warren for details of the project and available data.

Work Related to Publicity and Information

HEP

Les has been working with Joe Izen of UT Dallas to get an account there, in order to assist with concerns over performance between SLAC & UT Dallas. Worked with The Novosibisrk people and BaBar to put together a recommendation for upgrading the NSK-KEK link. Worked to understand an asymmetric throughput performance problem between SLAC and IN2P3 (see Network connectivity between SLAC and IN2P3, Lyon, France, Jan 2001).

DoE

Les has been working with someone from George Riley Georgia Tech and Rich Wolski from UTK to put together a proposal for "Optimizating Performance of Throughput with help from Simulation (OPTS)". Les is the PI and submitted the pre-propoasl to DoE on January 31st. In order to gather information for the above proposal Les made and compared simulator measurements with observed throughputs.

Les had many dicussions and emails with people from ANL, LBL, UTK, ORNL and PSC to put together the Active Measurement Infrastructure for ESnet proposal for which Les is the PI. Sent in pre-proposal on Jan 31st. Attended NTON meeting to assist with putting together an NTON proposal.

PAM2001

Warren worked on the paper for PAM2001

ITU

Les accepted an invitation to make a presentation and serve on a panel at the ITU meeting in Geneva in April. The topic is QoS and voice over IP.

I2

Les travelled to and attended the Internet 2 End-to-end performance initiative planning meeting in Ann Arbor Michigan.

XIWT

Coordinating between SLAC technology transfer office and CNRI to formalize the terms under which SLAC/IEPM is paid as a subcontractor to CNRI for a DARPA grant.

Back to Top


Revised 31 January, 2001.
Comments to iepm-l@slac.stanford.edu