🚀 go-pugleaf

In article <CyC9L1.5Hx@world.std.com>, mew@world.std.com (Michael E Willett) writes:
>Andrew Rothwell writes:
>
>>get some appropriate and useful comparative (RAID 7) benchmark results and
>>post them here...
>
>I've just been reading the new bit.listserv.banyan-l posting from John
>Krogh at Rockwell (Re: Preferred superserver for LANs: NetFrame or Tricord?).
>He just posted "Take benchmarks with a grain of salt.  In most cases, the
>testing environment almost always leans toward getting specific results
>(specific examples posted)...Statistics (benchmarks) can be used to prove
>just about any point of view you want them to...

If the vendor has composed the benchmark, that may well be true. There are,
however, benchmarks devised by independent organizations unaffiliated with
any vendor that can be employed to determine a base level of capability.

>Our own customers do seem to require being shown at their own sites exactly
>how much their own particular applications will be speeded up, rather
>than relying on prior benchmark data.

Agreed, but this sounds like an excuse for not releasing benchmark data.

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

-------------------------------------------------------------------------------
Ramon L. Tate             |  "Is it not strange that sheeps' guts should hale
DSB, DCRT                 |   souls out of men's bodies?"
Nat.Insts. of Health      |     - Act II, scene 3, Much Ado About Nothing
Bethesda, MD 20892        |               Wm. Shakspere
tate@faxcsl.dcrt.nih.gov  |