On 16/11/04 5:43 pm, in article
29a7918d.0411160943.48dd2444@posting.google.com, "Florian Heigl"
<florian.heigl@gmx.de> wrote:

> dean.sun@gmail.com (Dean) wrote in message
> news:<a2efdbbd.0411112105.d210392@posting.google.c om>
> ...
>
>> I did a write IO test in IBM P690 and HP SUPERDOME.
>> When I tested the local disks in P690,use this:
>> I used the same method to test the Storage of HDS 9980,which is
>> connected to the P690 and SUPERDOME.

> To some of the other posters:
> I agree in that dd really is not a reasonable benchmark. But first
> things come first! Horrible performance on /tmp? (and 7MB/s is far
> from what one should expect today, not even 4 years ago). I wouldn't
> care for the difference in between 30 and 40 MB/s, but to get down to
> 7MB/s something has to be wrong.
> In addition to that, if it were a memory filesystem and only made
> 100MB/s it would be even worse
> I get 3-7MB/s on my stone-age 4.3GB seagate disks in a B50 at home.
> both the p690 and the superdome a bit heavier on the wallet, and it's
> just
> fair Dean expects a bit more performance.
>
> Regards,
> Florian

That is correct - but as you know if you work with big systems, there are so
many factors that it can be impossible to diagnose without looking at
exactly how the system is set up and configured. For example, I was asked to
look at a large Sun installation where the backup (of all things) was
running incredibly slow. After investigating, it appeared that the new
version of Solaris combined with an SDLT 320 drive in the tape array and a
particular type of ultra-SCSI-3 card meant that the Solaris Volume Manager
internal mirroring strategy had to be changed from round-robin to
concurrent. A simple command in Solaris and this changed output from 3Mb/s
to 30Mb/s. Sun were not aware of this but they are now.

Incredible performance issues can have many causes and sometimes have a
simple solution. Sometimes the don't. My point is that running arbitrary
performance benchmarks is not the correct way to diagnose a problem. As you
quite rightly demonstrate Florian, experience is a big player here and in my
opinion, that is the most important factor that differentiates a great
system administrator from a mediocre one, but also luck ( I know, hope
management aren't reading this but haven't you investigated a major problem
to discover that it's suddenly solved itself but you take the credit!!!) and
the ability to think laterally but also very importantly to really
understand what is going on!

That's why they pay us lots of money to run their computers IMHO!