Status Report for SLAC/DOE IEPM Project
For September & October 2000
Work completed or in progress since the last report.
PingER Frank Nagy and Al Thomas visited SLAC and met with Warren and Les to review the PingER project, revitalize FNAL's involvement and discuss development and future direction. See notes from the meeting.
ToolsWarren installed the latest version of pchar, a tool for discovering network bottlenecks.
Warren reconfigured JNFlows, the tool for analysing protocols in and out of SLAC, for the new DMZ configuration.
Les put a fix into the traceroute server to fix a bug when the client is on the other side of a proxy.
JplotThe new RA, Imran had begun work on a new java applet for displaying the results from PingER.
IPv6Warren installed two new IPv6 machines, these are running Solaris 8.
Warren contributed to discussion on future direction for the Internet2 IPv6 WG and will attend the next face-to-face meeting.
QoSWarren and Les Video Conferenced with the Daresbury team on the new QoS collaboration. Warren organised Computer accounts at SLAC and Stanford.
Thruput MeasurementsFollowing an email discussion with Olivier martin of CERN, Les studied the relative impact of increasing the number of streams and window sizes on thruput achievable (see Bulk thruput: windows vs. streams. Talked to some people about how one might use measurements of RTT to provide feedback for adjusting the number of parallel streams to get optimum thruput while impacting others minimally.
Missing dataThis month has been heavy in lost PingER data with about 30% of sites have some form of outage for at least a day.
SC2000Further effort has been dedicated to the exhibit for Super Computing 2000. We will rpesent information on IEPM/PingER and high thruput measurements etc. All that remains is the conference!
BaBarWarren presented a report on network performance to the BaBar collaboration meeting.
Les met with Davide Salomoni of SLAC and visitor from INFN to discuss high performance thruput measurements etc.
Les has been assisting LBL and LBNL with understanding network problems due to the current routing between SLAC and LBNL/LLNL that goes via Chicago.
IPv6ForumWarren presented a report on "IPv6 Reliability and Performance" at the IPv6-2000 Conference.
NANOGWarren attended the North American network Operators (NANOG) meeting.
XIWT/DARPALes attended DARPA Network Modeling and Simulatiion workshop in Albuquerque. See trip report. Gave short presentation on PingER. Discussed possible collaborations with modeling folks, and also ways to validate their results.
Discussing a joint collaboration with Rice U to make some fractal measurements on the Internet. Sent email to NIKHEF ping author to request a couple of options that would help in the collaboration.
Les discussed making PingER data available to some ORNL statisticians and responded to NIST folks concerning PingER data format.
We are working with the GATech people to look at effect of packet reordering and streams vs windows by using simulation.
StanfordWe are working with Stanford to follow up on an invite for the sound ping stuff to become an installation at the SF Museum of Modern Art starting next March (who can predict these things?)
CybergeographyOur animated gifs of performance will appear in the next update of the Atlas of Cyberspace (http://www.cybergeography.org/atlas/)
Back to Top
Revised 25 July, 2000.
Comments to firstname.lastname@example.org