"Vernooy, C.P. - SPLXM" <[EMAIL PROTECTED]> wrote: > > > "Staller, Allan" <[EMAIL PROTECTED]> wrote in message > news:<[EMAIL PROTECTED]>... > > Kees wrote: > > <<snip>> > > > AFAIK, No. > > > > > > BTW Why are you still doing VIO at all? > > > It made a lot of sense when disk access times were 17-30 ms, but not > > > with 3-4ms (Escon/Cache) or < 1 ms (FICON/Cache) > > > > > > > Is this that simple? The fastest I/O is the I/O you don't do. With VIO > > you have a good chance to keep all data in storage, depending on the > > amount of real storege, your UIC and your paging. But in modern > systems > > I think you have a good chance to eliminate all I/O with VIO datasets. > > <</snip>> > > > > Everything is relative my son! > > > > VIO only eliminates some of the I/O (associated w/allocation). Under > the > > covers VIO is a data windowing system backed by aux storage. > > > Don't think so. This used to be the case in the early days, when systems > had 4 MB storage and keeping the vio window in storage could just be > toleated. Today I a sure VIO paging will not be that dumb to page in and > out while GB's of storage are available to keep the data at hand. So I > state that with VIO the I/O is eliminated. Anyone from MVS Internals who > can give the current state of facts? > > As for Ted's response that VIO overhead and fast Dasd are approacing > each other such that the don't differ much in performance: I don't have > figures to prove the opposite and I accept it if someone has measured > this, although it surprised me at first hand.
No numbers but a story from personal experience. When I first implemented SMS, I setup a VIO storage group with a limit of about 10 cylinders on a 3390 and made any requests over 10 cylinders go to dasd. About six years later, the tape guy was making changes to STK HSC database which caused it to increase in size to around 12 cylinders (hsc was created using vio and then gener'ed to dasd, IIRC). He noticed that his runtime on the hsc gen increased more than triple. This was on a 9672 G5 box using EMC 8830 escon. I can not imagine more recent equipment could make up this difference. Regards, John ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
