+1 for Paul's response, IIRC I did use the SMF dump program at one time
to copy SMF data, the SMF data for me was a good choice because you can
count on the data to have tons of records, good point to turn off zHPF,
run the test multiple times, your first run will prolly have the best
stats since the data has not been CACHED, or on some modern subsystems,
move to solid state storage.
some new subsystem have a hierarchy for data, that hierarchy can throw
off your test results
Carmen
On 8/25/2022 12:40 PM, Feller, Paul wrote:
Well depending on how technical you want to get. You could write some code to
write/read data and then look at the RMF records.
I created some code to do that using VSAM KSDS. A poor persons benchmark
software. I did not go as far as Carmen around turning stuff off/on. I would
run the jobs before a change and then after a change. I would run the stuff in
our sand box lpar to avoid other I/O from getting mixed in with my tests. Not
the most scientific approach but it worked enough to show what the change might
do. Last time I did something like this this was when we turned on zHPF.
Paul Feller
GTS Mainframe Technical Support
-----Original Message-----
From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> On Behalf Of
Carmen Vitullo
Sent: Thursday, August 25, 2022 12:25 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Modern DASD benchmark testing [EXTERNAL]
got ya, so sorry, It's been so many years since I've done any kind of
benchmark, we used a tool back in the day that generated data, then read that
data back over and over, the problem today, is I don't think there's so much
control over the DASD subsystem as there was when I was benchmark spinning DASD
or RVA's
we'd make sure we'd clear cache, turn off cache, DFW and run some benchmarks,
then do the same on the new subsystem(s)
maybe someone here still performs DASD benchmarks and can offer a solution
Carmen
On 8/25/2022 12:19 PM, rpinion865 wrote:
Not so much as to where to find the statistics. Rather, a set of benchmarks to
produce the statistics. Thanks.
Sent with Proton Mail secure email.
------- Original Message -------
On Thursday, August 25th, 2022 at 1:16 PM, Carmen Vitullo <petevi...@gmail.com>
wrote:
IIRC - the RMF/SMF record type 74-5 can you provide the I/O rates,
Cache Statistics.....etc
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ibm.com_serv
ers_resourcelink_svc00100.nsf_pages_zOSV2R4sc342664-3FOpenDocument&d=
DwICaQ&c=9g4MJkl2VjLjS6R4ei18BA&r=eUhu3PeeWy6RTndlJVKembFjFsvwCa8eeU_
gm45NyOc&m=Sfu9QUeD61DPi8IxmW4PEpaI9FONEd0zrNwb9t8PpxbGzSChwY9aFZZCy5
YUor0m&s=OeiFfcNCAm5sDtxmZe0oIBIKM9aS1iulszVqPhFFAT4&e=
Carmen
On 8/25/2022 12:11 PM, rpinion865 wrote:
I have been asked for ideas in regards to DASD benchmark testing. We would like
to compare IBM DS88xx's against competitors. Something that would generate
statistics for I/O rate, response time, and exercise track change rate
(replication performance).
Something free of course. Thanks in advance.
Sent with Proton Mail secure email.
--------------------------------------------------------------------
-- For IBM-MAIN subscribe / signoff / archive access instructions,
send email tolists...@listserv.ua.edu with the message: INFO
IBM-MAIN
---------------------------------------------------------------------
- For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO
IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send
email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to
lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
Please note: This message originated outside your organization. Please use
caution when opening links or attachments.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
