benchmarking SCSI host adapters

[Chip Rosenthal]

I have in my hot little hands two SCSI controllers - an Adaptec AHA-1542B
and a BusTek BT-542B.  I've been trying to do some benchmark comparisons
of the two using the "bonnie" program.  Both cards were tested in the
same system, all running with their default settings (5.0MB DMA, Sync
Negot disab, etc.).  I was told the BusTek gives about 3X the performance
of the Adaptec.  I'm not seeing this - I see 1X performance.

Numbers as follows:

            -------Sequential Output-------- ---Sequential Input-- --Random--
            -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine  MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
Adaptec 100   116 94.3   305 47.8   165 28.1   112 98.2   321 26.0  11.4 14.9
Bustek  100   116 94.3   305 48.1   164 26.9   112 98.3   321 24.8  11.5 14.6

For reference, these were run on a 20MHz non-cacheing Micronics in single
user mode, 14MB main memory, Fujitsu M2248SA hard disk, and SCO ODT 1.0
(with no kernel tweaking).

My questions are twofold:  (1) has anybody looked at the BusTek adapter
critically, and what results did you obtain, and (2) how useful is "bonnie"
for disk system benchmarking?

Right now, my guess is that either I was given marketing performance
numbers for the BusTek card rather than real performance numbers, or any
additional performance of this card is just wasted potential on this 20MHz
machine.  Comments anyone?

-- 
Chip Rosenthal  <chip at chinacat.Unicom.COM>
Unicom Systems Development, 512-482-8260 
Our motto is:  We never say, "But it works with DOS."