Sue,
Is there a way of seeing the results of a TRCASPBAL % of hot / cold data, without actually running a STRASPBAL?
I would like to run a TRCASPBAL just to find out %, if possible.
Most of our PF/LF are 3rd party, I would need to get ok from 3rd party to make the unit change, if I went that HDD direction.
At this point, I don't think my DASD subsystem is large enough to justify 2 EXP24S, one for SSD and one for HDD.
Can a single pair of 5913 handle both a SSD EXP24S and a HDD EXP24S drawer, or not recommended.
Would 2 pair of 5913 be needed, if I went that direction?
Paul
-----Original Message-----
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Sue Baker
Sent: Monday, December 29, 2014 9:29 AM
To: midrange-l@xxxxxxxxxxxx
Subject: Re: Is STRASPBAL the command to manage/move hot/cold data, P7 V7R1, with internal disks similar to Easy Tier
"Steinmetz, Paul" <PSteinmetz@xxxxxxxxxx> wrote on Sat, 27 Dec
2014 08:35:48 GMT:
1) Is STRASPBAL the command to manage/move hot/cold data, P7 V7R1,
with internal disks similar to Easy Tier?
Conceptually it is similar but implementation is quite different. Easy Tier constantly monitors the hot/cold data and rearranges appropriately. ASP balancing requires you to trace to determine what is hot while the trace is running. You run STRASPBAL *HSM to remove the cold data and move the hot data from/to the SSDs.
2) What is the
difference between STRASPBAL HSM (Hierarchical Storage
Management) and STRASPBAL MP (Media Preference), end results appear to
be the same?
*HSM is used after a trace to rearrange the data.
*MP is used to move DB2 tables and indexes or IFS UDFS marked objects to SSD and remove unmarked objects from SSD.
3) Can very large PF and LF be both hot and cold data? In other words,
the old data which may not be accessed be cold data?
Yes. ASP balancing has no awareness of PF/LF,STMF and looks at the hot logical block addresses.
4) Does this also apply to IFS data?
Yes.
5) Does STRASPBAL need to be constantly running?
It does not need to be run constantly. But you can use a CL program to trace for a given amount of time (at least 30
minutes) followed by the STRASPBAL *HSM in a continuous loop.
6) Does anyone have a SSD/HDD mixture and using STRASPBAL?
There are various i shops who have done this. Most run TRCASPBAL at specific times of day on multiple days then run the STRASPBAL *HSM during a relatively low usage period on Saturday or Sunday. By far most use UNIT(*SSD) to specifically place data on the SSDs.
You can combine using UNIT(*SSD) with trace & balance. Objects with media preference indicated do not get removed even if cold with STRASPBAL *HSM.
--
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.
midrange.com
