I guess I need to add my $0.02 worth. Since I do not know what Hardware your shop uses, or what you are looking at managing, this is a WAG.
With a storage array, you may think you have placed the datasets across the volumes, but your files inside the array could be where you did not want them to be. As for performance, so long as you are not sharing your storage between open systems and mainframe, you have enough cache on the storage array and enough DA, you have Solid State greater than Spinning disks, you performance should be okay. So can you truly balance files across a string of dasd. My observation is - not really. In the past with real 3390 and 3380 hardware, you needed to do that. But with today's storage array, not so much. True there are some files, like PAGE Volumes, or CA1 TMCs that need to think they are being handled differently. But in fact they could be residing on the same LUN as something you may not wish it to. Now with that said. It sounds you want something like the Serverpac that lays out files based on specific attributes. I would recommend creating a META Database with the file name, its attributes and accumulation process to tell you when that volume is full. Also test whether or not the file exists. However, once the META Database is built and the process works the way you want it to, then you will have (hopefully) less maintenance. Then a REXX process could build the IEBCOPY, IEBGENER, SAS, or whatever other program product you need to use to lay down the files. It will probably be a really extensive process as you need to add up the number of tracks/cyls/MB and see if it will fit that volume and then move on. In my mind this is a lot of work. I would side with others that indicate that SMS is the easiest and least painful way of creating your datasets. However, if you are intent on the other road, take a look at the process in Serverpac to see how it setups the SYSRES volumes. You might get some ideas out of it. Lizette > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Hardee, Chuck > Sent: Saturday, October 10, 2015 5:05 PM > To: [email protected] > Subject: Re: File Placement Utility > > Ron, > > By balanced I mean an even spread across the volumes. > > Our DASD farm is built from SAN type storage and cached so I/O performance, right > now, is not a major concern. > It may be down the road, but not now. > > In addition, the process using these files has its own caching mechanism in place so > all I am looking for is some tool to help me place the files in a somewhat, allocation > based, balanced order. > > C- > > Charles (Chuck) Hardee > Senior Systems Engineer/Database Administration EAS Information Technology > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
