[netperf-talk] new test in netperf2 top-of-trunk

Rick Jones rick.jones2 at hp.com
Wed Nov 7 18:30:02 PST 2007


Folks -

After suffering through "problems" running the packet_byte_script, not all of 
which were related to netperf :)  I decided to add a "TCP_MSS" test to netperf 
which can be used to easily (I never have learned much sed) get the MSS of a 
data connection (a la TCP_STREAM) between netperf and the remote.  Basically, it 
is a TCP_STREAM test with all the actual data transfer and cpu util stuff 
ripped-out.  The output with various options looks like:

raj at tardy:~/netperf2_trunk$ src/netperf -t TCP_MSS -H lart.fc.hp.com
TCP MSS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lart.fc.hp.com 
(15.11.146.31) port 0 AF_INET
Maximum
Segment
Size (bytes)

1448
raj at tardy:~/netperf2_trunk$ src/netperf -t TCP_MSS -H lart.fc.hp.com -v 0
TCP MSS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lart.fc.hp.com 
(15.11.146.31) port 0 AF_INET
1448
raj at tardy:~/netperf2_trunk$ src/netperf -t TCP_MSS -H lart.fc.hp.com -v 0 -P 0
1448
raj at tardy:~/netperf2_trunk$ src/netperf -t TCP_MSS -H lart.fc.hp.com -v 0 -P 0 
-B "foo"
1448 foo

I suspect the third version is the one most useful for scripting.  You can use 
all the options one can use in a TCP_STREAM test.  For example, one that will 
affect the MSS:

raj at tardy:~/netperf2_trunk$ src/netperf -t TCP_MSS -H lart.fc.hp.com -- -s 1024 
-S 1024
TCP MSS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lart.fc.hp.com 
(15.11.146.31) port 0 AF_INET
Maximum
Segment
Size (bytes)

724

Netperf2 top-of-trunk bits can be pulled via subversion/webdav (iirc) from:

http://www.netperf.org/svn/netperf2/trunk/

happy benchmarking,

rick jones


More information about the netperf-talk mailing list