Windowed Ping: An IP Layer Performance Diagnostic
Windowed Ping is also known as mping.
From the abstract of the 1994 INET paper:
In this paper we present a diagnostic tool
that provides direct measurement of IP performance, including queue dynamics
at or beyond the onset of congestion. It uses a transport style sliding
window algorithm combined with ping or traceroute to sustain packet queues in
the network. It can directly measure such parameters as throughput, packet
loss rates and queue size as functions of packet and window sizes. Other
parameters, such as switching time per packet or per byte, can also be
derived. The measurements can be performed either in a test bed environment
(yielding the most accurate results), on single routers in situ in the
Internet, or along specific paths in the production Internet. We will
illustrate several measurement techniques.
Please read the full INET'94 paper before trying to use this tool.
A number of things have changed since 1994:
Down load source here [Compressed tar].
- This tool was originally called 'mping' but due to a rumor of another
mping it was published in 1994 as 'windowed ping'. In this release it has been
restored to its original name.
- In 1994 there was the fear the mping could be abused. This fear is
still present but mping is very tame compared to tools designed for DoS. It
still is not appropriate to use mping in the Internet at large: Its normal
operation is quite abusive, and can disrupt normal traffic.
- The vast majority of the ICMP implementation in the Internet are either
at very low priority or explicitly throttled. As a consequence ICMP
performance is often so much worse than IP performance that you must
explicitly confirm or calibrate ICMP performance before you can trust any
- This release includes a number of bug fixes relating to changes in
standard library semantics, however the code has only received limited testing,
and all of it on Linux 2.4. It is quite likely that there are more bugs.
Please send patches.
- The original graphing tool used in the paper, xgraph, is no longer well
supported. Further complicating matters there are now multiple programs
called xgraph obscuring the original by David Harrison.
- The output syntax has been change since the original release to provide
more emphasis on packet losses.
- Due to the the changes in output syntax and xgraph the post processing
scripts included in this release no longer work. It would be greatly
appreciated if somebody re-wrote or updated the scripts to use one of the newer
tools such as gnuplot or xmgr.
This page is http://www.psc.edu/~mathis/wping/index.html.
For additional information check out these pages:
Pittsburgh Supercomputing Center,
Network research at PSC, or
Please send comments and suggestions to firstname.lastname@example.org.