TCP performance analysis

Hello,
Does anyone knows how to interpret results from TTCP performance
tool?
I executed ttcp -r -s on PC1 and ttcp -t -s 10.0.0.20. My results on
PC1 is
ttcp -r -s
ttcp-r: buflen=8192, nbuf=2048, align=16384/0, port=5001 tcp
ttcp-r: socket
ttcp-r: accept from 10.0.0.20
ttcp-r: 16777216 bytes in 14.89 real seconds = 1100.66 KB/sec +++
ttcp-r: 11573 I/O calls, msec/call = 1.32, calls/sec = 777.46
ttcp-r: 0.0user 0.0sys 0:14real 0% 0i+0d 0maxrss 0+2pf 0+0csw
I am unable to get last two lines. can anybody help me to understand
it?

(E-Mail Removed) wrote:
> Hello,
> Does anyone knows how to interpret results from TTCP performance
> tool?
> I executed ttcp -r -s on PC1 and ttcp -t -s 10.0.0.20. My results on
> PC1 is
> ttcp -r -s
> ttcp-r: buflen=8192, nbuf=2048, align=16384/0, port=5001 tcp
> ttcp-r: socket
> ttcp-r: accept from 10.0.0.20
> ttcp-r: 16777216 bytes in 14.89 real seconds = 1100.66 KB/sec +++
> ttcp-r: 11573 I/O calls, msec/call = 1.32, calls/sec = 777.46
> ttcp-r: 0.0user 0.0sys 0:14real 0% 0i+0d 0maxrss 0+2pf 0+0csw
> I am unable to get last two lines. can anybody help me to understand
> it?

The second to last line claims that ttcp make 11573 I/O calls (eg
send/recv), that each call took (on average I presume) 1.32
milliseconds, and that it was making those calls at a rate of 777.46
per second.

The last like is (perhaps bogusly depending on what calls it made to
get it) that there was 0.0 seconds of CPU time in user mode, 0.0
seconds of CPU time in "system" (kernel) mode and 14 seconds of wall
clock time. Alas I'm not familiar with the last bits on the last line.

rick jones

shameless plug time - you might want to look into netperf -
http://www.netperf.org/

--
The glass is neither half-empty nor half-full. The glass has a leak.
The real question is "Can it be patched?"
these opinions are mine, all mine; HP might not want them anyway...
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...