George, Splitting the Index and Data components of a KSDS has not become an unnecessary or bad practice. I've used rules in ACC/SRS in the past to do this, and I like the earlier suggestion of processing a psuedo volser in the ACS routines to accomplish the same thing. The benfits of this practice are decreasing though.
Typically this was done because the default buffering for NSR VSAM on a two or three level index would see the index thrashing as the index set was constantly reloaded after being pushed out by the sequence set. A listcat or stats in the Type 64 SMF record can tell you quickly if there is a problem by seeing if the index EXCP count is roughly equivalent to the DATA EXCP count multiplied by the number of index levels. Splitting the index and data onto separate volumes meant that the IO to the data component, 25% of IO for a three level index KSDS, would be happening on another volume. It looks good in the RMF reports, but NSR processing is usually single string and the index and data components are accessed serially. The main benefit was that it reduced skew a bit so that impact for other work on those volumes was reduced. HyperPAV and wide striping has decreased the benefit of this practice. It is probably still a good thing to do for IBM storage because the relatively large 2GB chunk size of their wide striping formula may see this carry through to a small amount of skew in a parity group. If you don't have HypePAV or WLM PAV with plenty of aliases, then it is still a practice that I would continue with. However if a particular batch job makes this critical to your race to sunrise I would recommend looking at using BLSR to do away with as much INDEX IO as possible, rather than the superficial band-aid of split index and data. I suspect the consultant doing the study understands that you do not have aliases and wide striping, or they are looking at an outdated cheat sheet. Ron ________________________________ From: George Rodriguez <[email protected]> To: [email protected] Sent: Wed, June 20, 2012 4:26:42 PM Subject: Re: [IBM-MAIN] Splitting the DATA and INDEX Components of Extended Addressing Dataset Please! I appreciate everyone's input on this problem... To answer why... This is a very high I/O file with delays. Mainline, the company assisting with the performance study, suggested splitting the DATA and INDEX component. Thanks for all the replies... * * *George Rodriguez* *Specialist II - IT Solutions* *Application Support / Quality Assurance* *PX - 47652* *(561) 357-7652 (office)* *(561) 707-3496 (mobile)* *School District of Palm Beach County* *3348 Forest Hill Blvd.* *Room B-251* *West Palm Beach, FL. 33406-5869* *Florida's Only A-Rated Urban District For Seven Consecutive Years* On Wed, Jun 20, 2012 at 12:53 PM, Baraniecki, Ray < [email protected]> wrote: > John; > > > > At Morgan Stanley we split the index and data all the time. We are > outsourced to IBM Global Services and it is their ACS routines that do > this. On the DEFINE control statements in the INDEX component the VOLUME > parameter looks like VOL=(I * * * * *) where the "I" is a pseudo volser and > the "*" are secondary volsers. On the DATA component we specify VOLUME=(D * > * * * *) where the "D" is also a pseudo volser and the "*" are secondary > volsers. > > > > I will contact IBM to obtain the details if you like. Please let me know. > > > > Thanks; > > > > > > Ray Baraniecki > > Morgan Stanley Smith Barney > > 1 New York Plaza - 18th Floor > > New York, NY 10004 > > Telephone: 212-276-5641 > > Cell: 917-597-5692 > > > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of McKown, John > Sent: Wednesday, June 20, 2012 12:09 PM > To: [email protected] > Subject: Re: Splitting the DATA and INDEX Components of Extended > Addressing Dataset > > > > > -----Original Message----- > > > From: IBM Mainframe Discussion List > > > [mailto:[email protected]] On Behalf Of George Rodriguez > > > Sent: Wednesday, June 20, 2012 10:34 AM > > > To: [email protected] > > > Subject: Splitting the DATA and INDEX Components of Extended > > > Addressing Dataset > > > > > > I'm hoping that someone can help me out here... > > > > > > Is there a way through the ACS routines to allocate the DATA > > > component on > > > one disk volume and the INDEX on another and are you willing > > > to share the > > > code or knowledge? > > > > > > Thanks in Advance! > > > > Interesting question. My "gut feel" is that you can't do this. But I do > wonder what would happen if the STORGRUP ACS routine mapped the different > components of a VSAM cluster into different storage groups, which would put > them on different volumes. I do know that the components must reside in the > same STORAGE CLASS. And that the STORGRUP ACS routine is not forced to map > a given STORCLAS to a single STORGRUP. > > > > -- > > John McKown > > Systems Engineer IV > > IT > > > > Administrative Services Group > > > > HealthMarkets(r) > > > > 9151 Boulevard 26 * N. Richland Hills * TX 76010 > > (817) 255-3225 phone * > > [email protected] * www.HealthMarkets.com > > > > Confidentiality Notice: This e-mail message may contain confidential or > proprietary information. If you are not the intended recipient, please > contact the sender by reply e-mail and destroy all copies of the original > message. HealthMarkets(r) is the brand name for products underwritten and > issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake > Life Insurance Company(r), Mid-West National Life Insurance Company of > TennesseeSM and The MEGA Life and Health Insurance Company.SM > > > > ---------------------------------------------------------------------- > > For IBM-MAIN subscribe / signoff / archive access instructions, > > send email to [email protected] with the message: INFO IBM-MAIN > > -------------------------------------------------------------------------- > NOTICE: Morgan Stanley is not acting as a municipal advisor and the > opinions or views contained herein are not intended to be, and do not > constitute, advice within the meaning of Section 975 of the Dodd-Frank Wall > Street Reform and Consumer Protection Act. If you have received this > communication in error, please destroy all electronic and paper copies and > notify the sender immediately. Mistransmission is not intended to waive > confidentiality or privilege. Morgan Stanley reserves the right, to the > extent permitted under applicable law, to monitor electronic > communications. This message is subject to terms available at the following > link: http://www.morganstanley.com/disclaimers. If you cannot access > these links, please notify us by reply message and we will send the > contents to you. By messaging with Morgan Stanley you consent to the > foregoing. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > Home of Florida's first LEED Gold Certified School Under Florida law, e-mail addresses are public records. If you do not want your e-mail address released in response to a public records request, do not send electronic mail to this entity. Instead, contact this office by phone or in writing. ---------------------------------------------------------------------- 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
