Re: Wanted: RAID Benchmarks

Gary M. Watson writes:

>The RAID Advisory Board (HP is a member) has commissioned Peer Protocols
>to write a benchmarking suite for RAID arrays, as well as a test suite.
>RAB members get the software for about 50% off, but it is still in the
>kilobuck range. The idea is that the RAID vendors run this test,
>then report the (hopefully) fair results in their literature. Cheating
>would be impractical as the RAIDMarks(tm) results are a registered
>trademark and the board can vote to stomp on anyone screwing around
>with the stats.

This is interesting. Who might Peer Protocols competitors be and how do they feel about Peer Protocols' ability to move from SCSI protocol testing to performance testing of applications and subsystems? How did Peer Protocols get selected over AIM and TPC? How do AIM and TPC feel about this? Does this make sense from the user's point of view?

I was just reading another posting about benchmarks by Alan Goodman at Tandem (goodman_alan_at_tandem.com) which reads in part:

>Looks like we are back into questionable and misleading benchmarks again.
>The COMDEX Server challenge was not a recognised industry benchmark and
>the results should be treated with deep suspicion...The Transaction
>Processing Council has been trying to create meaningful and comparable
>benchmarks that can be used with some reliability and credibility. We
>should support this attempt. TPC benchmarks are based on performance and
>price-performance including a 5 year cost of ownership...TPC-C is the
>current best available benchmark for servers.

Have Oracle, Sybase, etc. users been polled to see what RAID benchmark tests they would like to use?

Mike Willett
Storage Computer Corp.
11 Riverside Street
Nashua, NH 03062
Tel. 603-880-3005
Fax 603-889-7232

::::::::::::::::I/O-accelerated storage to more than a terabyte:::::::::::::