Re: Performance of large LF access path builds - SSD vs 10K spinny

Hi Sue

You said this:
"Feature 5913 has 1.8 GB of write cache. However, if your SSDs
are in a single parity set, you won't be getting the best
possible performance out of them."

Could you please expand on that comment a little to explain why that is ?

I'd just like to understand this a little better.


On Tue, Sep 23, 2014 at 5:16 AM, Sue Baker <sue.baker@xxxxxxxxxx> wrote:

"Steinmetz, Paul" <PSteinmetz@xxxxxxxxxx> wrote on Sat, 20 Sep
2014 00:38:07 GMT:

Could this be due to 5913, 0 read cache?

Feature 5913 has 1.8 GB of write cache. However, if your SSDs
are in a single parity set, you won't be getting the best
possible performance out of them.

Is there anything else I'm missing?

As I said before there are many variables. I can't recall if
anyone has asked about expert cache being turned on for the
memory pool where the reorg jobs run.

As I said, without looking at PEX trace data and standard
collection services data, we can speculate and point at things
to look for but are unlikely to come up with the exact reasons
for the performance you are seeing.




--
Sue
IBM Americas Advanced Technical Sales Support (ATS) Power
Systems
Rochester, MN
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.