What I am trying to find out (or avoid) is that this rule starts leading its own life. When asking why will get the answer: just because this is the rule.
As I remember contiguous slot allocation was introduced in the 3380 period. The first 3380's had a 625 MB capacity, it was probably reasonable to keep a pageds fairly empty to allow ASM to always find 30 contiguous slots (a number I remember from then). This basic ruled lead to a 30% max utilization recommandation on a 3380-A, but does this still lead to the same utilization on 4GB 3390-09 pagedatasets? Or can I fill these up to 70% (the other ASM limit) and still leave enough free space for ASM to always find 30 contiguous blocks? Hopefully Greg can shed a light on this. Kees. "Shane Ginnane" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>. .. > I've always seen it as a recommendation not to fill AUX above 30% (or 25%, > lets not quibble) for best performance. > Generally would equate to the same thing in practice, but I've never > specifically seen it as a disk (aka "dataset") recommendation. > > As such, I always try to keep the AUX highwater mark at 30%. This means > (for me) the same number of total slots. > Although I always seem to wind up adding more anyway. > > However ... > it is reasonable to suppose that this recommendation came about because of > ASM algorithm design that may well have been predicated on the limitations > imposed by SLED. > See Steve Samsons post from Sept 1997 > "There is another implicit threshold, namely an occupancy of 30-35% at > which ASM stops trying to allocate contiguous slots for multi-page > transfers. This may not be a performance problem on the EMC box, but ASM > does work harder and the channel programs become less efficient. For > best results, try to keep the occupancy at less than 35% most of the > time." > > With the newer disk controllers, this may have all gone out the door, and > maybe ASM *can* tolerate higher occupany rates. > Greg is probably the only person to answer that one - maybe Jim as well. > > > > > I am referring to the *recommandation* to not fill a pageds more than > 25%. > > AFAIR this recommandation dates from the 3380 era. > > > > My question is, is this ROT still correct for large 3390-09 pageds (and > was > > it still correct for 3390-03). > > > > If I replace a couple of (25%) filled 3390-03 pageds's with 3390-09's, > > should I keep the total space occupied for a max of 25% or could this > value > > be raised, so I need less pageds's? > ********************************************************************** For information, services and offers, please visit our web site: http://www.klm.com. This e-mail and any attachment may contain confidential and privileged material intended for the addressee only. If you are not the addressee, you are notified that no part of the e-mail or any attachment may be disclosed, copied or distributed, and that any other action related to this e-mail or attachment is strictly prohibited, and may be unlawful. If you have received this e-mail by error, please notify the sender immediately by return e-mail, and delete this message. Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its employees shall not be liable for the incorrect or incomplete transmission of this e-mail or any attachments, nor responsible for any delay in receipt. ********************************************************************** ---------------------------------------------------------------------- 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
