IEPM

Status Report for SLAC/DOE IEPM Project

For April 2002

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

Work Related to Analysis and Development

PingER

Warren modified the PingER table to include details of the top level domain (tld) and region (continent). He is also working on producing 48-hour reports in addition to the 24-hour report.

Warren worked with Jerrod Williams to help him understand how the data gathering and analysis tools are maintained.

INCITE

Warren worked on making Pinger-like reports for INCITE results and developing new code to implement the back-to-back and sandwich packets required by the INCITE methodology.

Jiri is working on providing traceroute maps identifying the network, see for example: http://www-iepm.slac.stanford.edu/bw/incite/TOMO/IEPM/.

IPEX

Warren set up access to the IPEX machines for the new student.

Traceping & Traceroute

Warren rewrote traceping to improve stability. He also modified forking to avoid erroneously tripping alarm systems at some sites.

Les added more blocks to abusers (sending many tens of traceroute requests in a burst, and not responding to emails to them) to the traceroute server. Also added a link to the Abilene router access so can use for traceroutes from the interior of Abilene network.

Globus

Warren worked with the DOE-SG certification authority (CA) to fix a problem with the certificate they had issued for a SLAC machine. Warren worked with Rod Walker from Imperial College London to set up GridFTP transfers using DOE certificate authentication. Warren brought up MDS and is working to add PingER data.

Web100

Warren Web-100 enabled several machines for IEPM-BW and BBCP-testbed. Warren worked with Tom Dunigan to implement his patch to allow systcl to disable the Linux 2.4 caching of window sizes.

Les modified tcpload.pl to incorporate Web100 to get PktsRetrans, PktsOut, SmoothedRTT etc at the end of each iperf test. Had several discussions on the web100 email list and with Matt Mathis and others in person concerning use of the variables.

AMP/IPMP

The new external monitoring network has been created to allow the SLAC AMP machine to send/receive IPMP (protocol 169) packets. However we are still waiting on security to implement the ACLs in the new router to permit this traffic from the participating sites.

IPv6

Warren was informed the Euro6IX and 6net projects are interested in picking up the IPv6 Pinger Monitoring.

IEPM-BW

Worked with NASA Goddard folks to get a new Linux (instead of BSD) remote host to monitor. Ran into problems with getting OpenSHH versions 2 (at SLAC) and 3 (at Goddard) to work together. Gary, Les & Jerrod tried many things without success (later 4/26/02, after essentially giving up, it was found to be working, not sure what magic did the trick).

Connie, Manish and Les worked on comparing passive (netflow) measurements with active iper, bbcp and bbftp measurements. We are arranging to make a presentation on this work at the Internet 2 E2Epi BOF meeting in May.

Warren worked with Paola Grosso to set up a bbcp testbed to examine the effect of cross traffic.

Warren installed a 'proof-of-concept' windows and streams server. Applications can consult the server to find the recommended number of parallel streams and window size to use for a certain remote site.

Connie and Jerrod are working on porting run-bw-tests to Linux. This is pre-requisite\ for incorporating Web100, and porting the monitoring to FNAL and Milan.

Missing Data and problems

Still no data from BNL nor INFN.

Les worked with INFN to pin point some network problems they were having that resulted in poor throughput between SLAC & Rome.

Misc

The usual paperwork, installations and administration.

Work related to Publicity and Information

Travel, talks and publications

Warren worked on a talk for Connie Logg to present at the Internet2 members meeting.

Les & Connie attended the Passive and Active Measurements workshop in Fort Collins Colorado. Les was a program advisor and chaired a session. Wrote trip report.

We had a phone meeting with Rice.

Les visited Romania at the invitation of the Romanian Ministry of Telecommunications and Information Technology and the Ministry og Education of Research. He was part of a team from the west (including Harvey Mewman, Ian Foster and David Williams) to provide assistance for "The Romanian Potential regarding Grid Activities and Distributed Computing". Presented talk on Worldwide Network Performance and Monitoring.

On the way to and from Romania, Les vsited with Richard Hughes-Jones in Manchester and Peter Clarke of UCL to discuss bandwidth measurements, various collaborations etc. See the trip report for more details.

Les and Connie worked on a paper on IEPM-BW to be submitted to SC2002.

Warren's article on QBone Scavenger Service (entitled "Network Scavengers") was published as the first article in the CENIC InterAct Network Applications magazine.

DoE

In response to a request from Thomas Ndousse, we sent him some publicity graphics for a publication he is working on.

Les wrote the yearly Field Task Proposal Agreement/Field Work Proposal for DoE.

Jiri wrote the half-yearly report for SLAC's contribution to the INCITE project report.

NSF & Other funding agencies

Les wrote a letter of recommendationfor a Rice/Michigan NSF prposal.

Les reviewed a UK DataGrid proposal and sent in a recommendation.

PPDG & HENP

Warren & Les put together the quarterly PPDG-IEPM report

Les met with Mauro Morandin of Padova, to discuss increased monitoring since they will become a Babar tier 1 site later this year.

NetPredict

We met with NetPredict to discuss their new product and where it might fit for SLAC. We also shared information on QoS requirements for Voice over IP.

Created April 28, 2002, last update April 28, 2002.
Comments to iepm-l@slac.stanford.edu