Found it: https://www.ibm.com/support/pages/node/6355789 https://www.ibm.com/support/pages/system/files/inline-files/IBM_z15_FEx16SA_Performance_2020.pdf
- KB ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ On Thursday, December 9th, 2021 at 9:28 AM, kekronbekron <[email protected]> wrote: > Hi Attila, > > For the curious, a link to the study will be helpful. > > Is it in TechDocs? > > - KB > > ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ > > On Thursday, December 9th, 2021 at 9:05 AM, Attila Fogarasi > [email protected] wrote: > > > I don't know about that particular benchmark, but IBM has a published z15 > > > > zHPF benchmark which does 300k+ IO/sec over a SINGLE FEx16SA channel (using > > > > multiple ports). That was published almost 2 years ago (jan 2020). > > > > On Thu, Dec 9, 2021 at 2:18 PM Clark F. Morris < > > > > [email protected]> wrote: > > > > > On Thu, 9 Dec 2021 02:21:28 +0000, Seymour J Metz [email protected] > > > > > > wrote: > > > > > > > Going from 5 milliseconds to one tenth of that is certainly quite > > > > > > > > significant, and I have no doubt that one day we will get to the > > > > > > > > performance that BJ claimed, but we're not there yet. > > > > > > When it is claimed a Z15 LPA maxes out at 950K I-Os, how many > > > > > > concurrent I-Os over how many paths are involved? > > > > > > Clark Morris > > > > > > For IBM-MAIN subscribe / signoff / archive access instructions, > > > > > > send email to [email protected] with the message: INFO IBM-MAIN > > > > For IBM-MAIN subscribe / signoff / archive access instructions, > > > > send email to [email protected] with the message: INFO IBM-MAIN > > For IBM-MAIN subscribe / signoff / archive access instructions, > > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
