bandwidth estimation tools that `suck the least' [ippm] pathrate & pathload]
k claffy
kc at caida.org
Sat Apr 12 00:53:41 UTC 2003
anyone willing to help by sending unsatisfying output
back to the knightly development team
(with as much information as you have about the
infrastructure being gauged)
will be of great help to some incredibly talented members
of the research community beating at this holy grail
trying it out and deciding it doesn't work as well as you need
and not sending any feedback that might help the authors improve it
doesn't help anyone at all
but you knew that already
k
----- Forwarded message from Constantine Dovrolis <dovrolis at cc.gatech.edu> -----
Date: Thu, 10 Apr 2003 16:47:07 -0400 (EDT)
From: Constantine Dovrolis <dovrolis at cc.gatech.edu>
Subject: [ippm] pathrate & pathload
To: ippm at advanced.org
cc: Manish Jain <jain at cc.gatech.edu>, Ravi Shanker Prasad <ravi at cc.gatech.edu>
We would like to announce a new release of our bandwidth
estimation tools: pathrate and pathload. Pathrate measures
end-to-end capacity (aka bottleneck bandwidth), while
Pathload measures end-to-end available bandwidth.
Both tools are available at:
http://www.pathrate.org
The major differences in the new versions:
Pathrate - 2.3.0
----------------
* The tool has been also tested in Gigabit Ethernet paths. We added
functionality to deal with interrupt coalescion at the receiver.
* A "quick termination" mode (option -Q) was added for an estimate
after just a few seconds. Useful for frequent light-weight measurements.
* Support for netlogger output format.
Pathload - 1.1.0
----------------
* The previous version was able to measure available bandwidth
only in the 2Mbps-120Mbps range. The latest version has been
successfully tested in both low bandwidth paths (dial-up, DSL,
cable modems), and in high bandwidth paths (OC-3, OC-12, GigEthernet).
If the receiver does interrupt coalescion, the tool can only
report a lower bound on the available bandwidth.
* The tool can now automatically choose an appropriate bandwidth
resolution (previously this was a user-specified parameter).
* Support for netlogger output format.
Enjoy,
Constantinos, Manish, and Ravi.
_______________________________________________
ippm mailing list
ippm at ietf.org
https://www1.ietf.org/mailman/listinfo/ippm
----- End forwarded message -----
More information about the NANOG
mailing list