[SNMP4J] Performance Benchmarks
Frank Fock
fock at agentpp.com
Mon Oct 22 18:44:32 CEST 2007
Hi Adam,
It seems that you have done the test with debugging
enabled in the trap receiver. Have you tried the
same with the "-d OFF" option?
Are you sure that the UDP buffer of the target OS
is big enough to handle 1000 trap packets per
second?
Best regards,
Frank
Adam Brons wrote:
> I am wondering if any performance benchmarks have been done using the
> SNMP4J API to receive v1 and v2c traps. We have a test application to
> send 1000 v1 traps per second to the SNMP4j command-line tool using the
> following command-line options:
> SNMP4J -Ol 0.0.0.0/162
>
> What we have noticed it that over a period of 9 seconds only 1348 v1 traps
> were processed by the SNMP4j command-line tool. This works out to 149
> traps per second. Does this sound on pair or should I be expecting a
> great number more or less.
>
> Adam Brons
> TSOM Software Engineer
> IBM Tivoli Software
> _______________________________________________
> SNMP4J mailing list
> SNMP4J at agentpp.org
> http://lists.agentpp.org/mailman/listinfo/snmp4j
--
AGENT++
http://www.agentpp.com
http://www.mibexplorer.com
http://www.mibdesigner.com
More information about the SNMP4J
mailing list