Ed, Actually the TCW has a shorter path length to process on the storage side, at least on HDS kit.
On the VSP we see an improvement in response time, tens of microseconds, and an increase in max throughput of a VSD (Virtual Storage Director - the MP that handles almost everything except data transfer). Noticeable for cache hit workloads, but not so much for cache miss. Ron > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of Ed Jaffe > Sent: Tuesday, August 06, 2013 12:48 PM > To: [email protected] > Subject: Re: [IBM-MAIN] zHPF > > On 8/6/2013 12:35 PM, gsg wrote: > > Would like to turn on zHPF. Is there a way to benchmark this to see what > we're getting out of it? Would like to do this in test and then again in > production. > > The main advantage to zHPF is more (overlapped) I/Os per second, not faster > per-I/O speed. Therefore, to observe a measurable difference, you will need > to run one "hellacious" benchmark! :) > > -- > Edward E Jaffe > Phoenix Software International, Inc > 831 Parkview Drive North > El Segundo, CA 90245 > http://www.phoenixsoftware.com/ > > ---------------------------------------------------------------------- > 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
