MIDRANGE dot COM Mailing List Archive



Home » MIDRANGE-L » November 2012

Re: Using SSD with IBMi V7R1



fixed

Helo Luis

Thanks,
That helps a lot, I'll keep digging thought.

Regards
Gad


message: 4
date: Wed, 14 Nov 2012 08:10:36 -0430
from: Luis Rodriguez <luisro58@xxxxxxxxx>
subject: Re: Using SSD with IBMi V7R1

Gad,

I would like to help you more regarding SSDs. Nevertheless, unfortunately
we don't have them in our shop. Although we still have an old i5-520 (V5R3)
for legacy and BI purposes, about 3-4 years ago our company converted to
the SAP/Oracle/HP environment.

All the above stated notwithstanding, the IBM i is still my favorite
platform, and I like to keep abreast on new developments.
Regarding to the KEEPINMEM parameter, it was introduced (AFAIK) with 7.1.
Let me quote you the IBM 7.1 Technical Overview Redbook:

**********

*The KEEPINMEM parameter specifies whether the data or an access for a file
member is*

*brought into a main storage pool by the SQL Query Engine (SQE) when the
data is used in*

*the query to improve the performance. When you specify *YES for this
parameter, the Query*

*Options File (QAQQINI) parameter MEMORY_POOL_PREFERENCE (see 6.4.15,
?QAQQINI*

*properties? on page 150) specifies the preferred main storage pool to be
used.*

* *

*This function applies only during the run time of a query, and might
therefore be substituted*

*for the Set Object Access (SETOBJACC) command that puts the table or index
in memory in a*

*static function. After the query completes, the memory might be freed
again, contrary to the*

*effects of the Set Object Access (SETOBJACC), where you need to clear it
using the *PURGE*

*option on the Storage Pool (POOL) parameter of the command.*
**********
So, while it has certain similarities to the SETOBJACC it?s not quite the
same thing, as it seems to be a little more dynamic and specific to a
particular query. Also, this parameter would work in any kind of DASD, SSD
or not (and, although a SSD is quite fast, I suppose that anything residing
in main memory should be faster still, so *maybe* this still is valid in a
SSD world).


Best Regards,

Luis Rodriguez
IBM Certified Systems Expert ? eServer i5 iSeries
--




On Wed, Nov 14, 2012 at 6:53 AM, Gad Miron <gadmiron@xxxxxxxxx> wrote:

Hello Luis

I read with interest the http://tinyurl.com/6aqt5xl
and would like to enquire further:

1. Regarding SSD object placement,
Isn't there a system utility that identify "Hot" objects and
automatically moves them
to SSD units?
2. while looking at CHGPF/CHGLF CMDs I noticed a KEEPINMEM
(*SAME/*NO/*YES)
parameter I haven't noticed before.
Is it new to 7.1 ?
Is it equivalent/substitute/enhancement to SETOBJACC ?
Isn't it actually unnecessary once you have a SSD unit?

Thanks
Gad










Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2014 by MIDRANGE dot COM and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available here. If you have questions about this, please contact