Re: NCR benchmarking
Sun, 7 Jul 1996 01:12:25 +0200 (MET DST)

On Sat, 6 Jul 1996, Gerard Roudier wrote:

> IMHO, in order to get acceptables results, the file size must be at least 3
> times the main memory size.
> Can you try again your benchmark with a file size of about 200 MB.

Yes, of course, there should be about 300 megs free on the same drive.

> I get about 3.6 MB/s for input "per char" and 4.3 MB/s for block input
> with my 2 years old IBM S12, 100MB file size and 32MB main memory.
> I suspect these results to be wrong due to delayed cache flush that
> Bonnie cannot see and that the both speed should be the same.

The IBM DPES-31080 isn't a very fast drive. In fact, it is slower than
most modern EIDE drives, so the values I got from Bonnie are looking o.k.
I'am not so interested in absolute numbers (because I already know,
that the DPES is not fast as hell), I was more interested in comparing the
two drivers on exactly the same hardware.

> You were not in the same conditions.
> The effects of tagged queue depends on hard disk firmware.
> Generally, performances are not affected or are a little increased under
> Bonnie. With some HDs, performances are lower with tagged queue.
> On the other hand, IMHO, bonnie benchmark is not a serious test for tagged
> command queueing.

Well, I don't know anything about the firmware of this drive, but I will
repeat the test with tagged queuing disabled. I'll send you the results
with the results, I'll get from the 200 MB Bonnie tests.

What about some other benchmarks. I have the byte benchmarks here on my
system. Is this suite more serious than Bonnie ?

# /AS/ Linux 2.0 #