Benchmarking Software
HP generated test results using the netperf Version 2.4.1 benchmarking software and the following options:
Specifies to gather local CPU utilization for the test. You can calibrate the CPU rates separately using LOC_CPU and REM_CPU tests. HP used
Specifies to gather remote CPU utilization for the test. The CPU rates can be calibrated separately using LOC_CPU and REM_CPU tests. HP used
Specifies the remote client IP address.
Specifies duration (in seconds) of the test. HP typically used 60 seconds per test run.
Specifies the remote socket buffer sizes. HP used 128K buffer sizes. Specifies the local socket buffer sizes. HP used 128K buffer sizes. Allows the binding of netperf/netserver processes on the local (server) and remote (client) CPUs.
Use TCP_STREAM for transmit tests and TCP_MAERTS for receive tests. Transmit and receive tests were run concurrently to generate bidirectional workloads. The TCP_MAERTS test is new. You can find more information about this test at:
http://www.netperf.org/netperf/NetperfNew.html
Specifies the message size for the test. We used 32K message size.
You can find more information about netperf, documentation, and obtain a free copy at:
http://www.netperf.org
For More Information
•HP APA documentation: http://docs/hp/com/en/netcom.html#Auto%20Port%20Aggregation%20%28APA%29
•Transport Optional Upgrade Release Version 3.0 For information:
•IP Filter product information and free download: http://h20293.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=B9901AA
•HP ProCurve Network products: http://www.hp.com/rnd/index.htm
Benchmarking Software 13
