Warren prepared and presented a
talk to the CCIRN group
to update them on the present status
of the Pinger tools.
We discussed helping the XIWT port the PingER code from SAS to SPSS.
Unfortunately it does not look like a good match for our resources and skills.
Warren prepared a
talk and a
poster to present to the
CHEP98 conference in Chicago at the end of August.
Coordination of IEPM Project
We had a very fruitful meeting at SLAC of
Warren, Charley Granieri and Les with Dave Martin of HEPNRC to go
over the current IEPM status, decide on how to parse the CHEP98 presentations,
and what should be our next focii.
Notes of the meeting
We met with a group of people from NetPredict (a start up whom we assisted and
recently was awarded an SBIR award).
Business Communications Review (BCR)
We exchanged a lot of Email with an editor of BCR who was looking for detailed
information on how the Internet is performing. We provided her with many
references, as well as charts of performance and explanations. She also kindly
sent us a draft of her document to review. The article should be published next
Warren has done a lot of work remaking the report generator programs
based on gathering the data daily and making daily reports with
The new programs also store the data in gzipped format, and the data
is unzipped by the program when required. It has been suggested using
(g)zcat instead of gzip, but we are concerned about memory issues. Also
we will re-write the data gathering program to gather data from the
archive site at HEPNRC instead of each monitoring site.
Surveyor is working well and taking data. Some of the surveyor nodes have
been added to the nodes pinged from SLAC, so we can compare the performance
reported by Surveyor to performance reported by Pinger.
We have made maps detailing the beacon sites in North America and Europe.
These are useful for presentations, but we also hope they will be useful
when the netmaps work develops.
We worked with Marek Karliner, Hank Nussbacher and others from Tel Aviv
university to understand and improve the ping monitoring of their hosts. Previously they
were blocking hosts that made ICMP requests requiring > 10kbps, which
was causing problems for the 1000 byte pings. They put in a
filter to allow PingER sites through.
We made several updates to the
Tutorial on Internet Monitoring to add information on PingER pathologies, to show
the impact of line upgrades and holidays on the performance of the RAL-SLAC link,
to add information on one way pings,
and to add information on the grouping.
Back to Top
Revised 24 August 1998