I've rolled it out into 2 of our test LPARs and have been using the HIS 113 records to try and measure the improvement. Specifically looking at the percentage of TLB miss processor cycles. I haven't seen an improvement using that metric, going into production this month so stay tuned.
On Thu, May 2, 2013 at 5:07 AM, Markus Haselbach < markus.haselb...@credit-suisse.com> wrote: > Hello, > > we start using LFAREA (1M pages) with z/OS 1.13 and DB2 10 on z196. > From IBM I read: "Using large pages can improve performance for some > applications by reducing the overhead of dynamic address translation". > We wonder how to measure this performance improvement. What we're trying > now is measuring and comparing CPU usage/ SQL with and without 1M pages. > Does someone have experience with this performance improvement and can > anybody advice me on what should/could be measured? > > kind regards > Markus > > > > ---------------------------------------------------------------------- > 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
