Status Report for SLAC/DOE IEPM Project
For May 2002
Work completed or in progress since the last report.
PingERWarren worked with UCL on providing node details and pinger data as a web service.
Warren worked on amending the pinger code to directly access the database for configuration details. Les reviewed the Eu DataGrid modifications to PingER witha view to incorprating them in the standard release.
Les applied a fix from NASA to the connectivity.pl program and made available a new public version.
BandwidthWarren worked with UCL to understand the European Datagrid (EDG) extensions to pingER and how the IEPM-BW project and the UCL bandwidth monitoring work can work together without installing the EDG extensions at SLAC.
Warren looked at getting a GE interface for the machine on campus.
Warren worked on a new proposal to be called Maggie ( Measurement and Analysis for the Global Grid and Internet End-to-end Performance ).
Les worked with Richard Hughes-Jones of Manchester to set up a remote host at Manchester. We eventually measured over 400Mbits/s to it (70% of the Internet 2 Land Speed record), see Bulk Throughput Measurements - Manchester University, England. THis in turn initiated an interesting discussion on jumbo frames on a mailing list. Les also set up a remote host at UCL in London.
Andy Hanushevsky and Les submitted a paper on compression to a conference in Paris later this year. Les and Connie submitted a paper on "A New High Performance network and Application Monitoring Infrastructure" to the Internet Measurement Workshop in Marseille later thus year.
Connie made a major rewrite of the IEPM-BW software (this is the 2nd rewrite) as part of successfully porting the measurement, extraction, anlaysis and reporting code to Linux. Les ported the program for pushing updates to the remote hosts, and the script to analyze and compare passive and active measurements.
INCITEJiri and Les worked on topology/tomography visualization. Les worked with Jiri and the Rice/LANL collaborators to put together the annual INCITE report for DoE.
IPEXWarren is working on an interface for requesting IPEX data. The user will be able to request a specific set of passive measurements (not summaries), for example 'all port 80 traffic between 3:00pm and 5:30pm collected at SLAC'. The user will also be able to request an active measurement, for example 'traceroute from SLAC to Telcordia'. Allen McIntosh from Telcordia will determine a required or preferred Certification Authority (CA) with which to authenticate users.
GlobusWarren and Les worked with the administrators to overcome difficulties applying for a DoE certificate for Les to use with the Globus toolkit.
Warren worked with the administrator at Fermilab to get kerberized-ssh access to the monitoring machine.
Web100Upgraded to newest alpha1.2. Les and Heng Mei are working on adding Web100 measurements to IEPM-BW.
AMP/IPMPStill no progress with bringing up the external-monitoring network and implementing the IPMP protocol at SLAC.
IPv6Warren provided some tips on the IPv6 port of PingER for the developer at Dante who is taking over the project.
TracerouteLes followed up on a large number of traceroute requests from Wisconsin that turned out to be caused by a program bug at Wisconsin.
Missing DataThe PingER site at INFN is up, but the script that serves the data did not work. Warren worked with the local administrator to resolve the issue.
Still no estimate from BNL as to when the new machine will be operational.
The PingER sites at the University of Wisconsin and Los Alamos National Lab both experienced NFS problems which disrupted data taking briefly this month.
Jerrod amended the data taking process to use the new ESnet site. Warren assisted trouble shooting a data collection problem.
MiscThe usual paperwork, installations and administration.
Warren received email from a developer at the National Grid USA regarding the use of the jplot code.
Travel, meetings, talks, reportsLes wrote and submitted the Annual IEPM report to DoE.
Warren will attend the Global Grid Forum in July. Les attended the CENIC meeting in San Diego (see the Rough Notes, and an INCITE review in Santa Fe (see Rough Notes).. Connie attended the Internet 2 meeting in Arlington Virginia. Les and Connie are working on putting together presentations on network performance for the SLAC/FNAL booth for SC2002. Les attended a meeting at FNAL to plan the presentations for the whole booth. Les met with the FNAL/IEPM folks to discuss porting IEPM-BW to FNAL.
The following talks were presented:
Overview of IEPM-BW - Bandwidth Testing of Bulk Data Transfer Tools, presented by Connie Logg at the May 2002 Internet2 Conference in Arlington, VA. Comparing Throughputs as Measured by Active and Passive Methods, presented by Connie Logg at the May 2002 Internet2 Conference in Arlington, VA. International Connectivity and Performance for the HENP Community, presented by Connie Logg at the May 2002 Internet2 Conference in Arlington, VA. INCITE Review, presented by Les Cottrell at INCITE Review at Santa Fe, May 2002. Experiences and results from implementing the QBone Scavenger, presented by Les Cottrell at the CENIC meeting May '02 in San Diego.
Hires, visitors etc.Richard Hughes-Jones of Manchester and the Eu DataGrid spent a week and a half at SLAC working on porting the measurement software to Manchester. Yolanda Tsang a PhD student from Rice is at SLAC for 3 months this summer she is working with Jiri. Ajay Tirumala a PhD student from UIUC is at SLAC for the summer working with Les on bandwidth measurement issues. Warren will work with a student from Arkansas who will visit SLAC for 8 weeks this summer. Heng Mei a Stanford student is working with Connie this summer.
Working groups etc.Les put together a draft web page for the Internet 2 HENP network WG for high performance throughput verification.
Les attended 3 GGF Network Monitoring Working Group (NMWG) phone meeetings and worked on the Network Measurements for Grid Applications document.
Created May 7, 2002, last updated June 8, 2002.
Comments to email@example.com