allocating volume M27 before M9



JES, JES2, JCL utilities, IDCAMS, Compile & Run JCLs, PROCs etc...

allocating volume M27 before M9

Postby samb01 » Fri Sep 05, 2014 2:22 pm

Hello,

i have to allocate a huge dataset. About 584130 tracks.

In the pool, there are 18 disk. 16 M9 and 2 M27 (VOL17, VOL18).

When i allocate the dataset, i would licke to use thirst the two disk m27, and after, if there is no place on these volume, the dataset should use the other disk M9.

Could i use this in the JCL :

//FICO    DD DSN=BIG.DATASET,DISP=(,CATLG,DELETE)
//         SPACE=(CYL,(1000,1000),RLSE),VOL=SER=(VOL17,VOL18,other volume...)             



Thank's for your hlep.
samb01
 
Posts: 236
Joined: Mon Nov 16, 2009 7:24 pm
Has thanked: 0 time
Been thanked: 0 time

Re: allocating volume M27 before M9

 

Re: allocating volume M27 before M9

Postby steve-myers » Fri Sep 05, 2014 2:59 pm

I'm not so sure we can provide a meaningful answer without more information.
  • Are the volumes SMS managed?
  • Can all or part of the data set exist in the cylinder managed area of the M27 volumes?
steve-myers
Global moderator
 
Posts: 1885
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 197 times

Re: allocating volume M27 before M9

Postby samb01 » Fri Sep 05, 2014 3:07 pm

Hello steve-myers.

Yes the volume are sms managed.

Can all or part of the data set exist in the cylinder managed area of the M27 volumes?


i'm not sure i andertood the question...

Every dataset can be allocate in the M27 volume.

Acutally the are dataset in the M27 Volume.
samb01
 
Posts: 236
Joined: Mon Nov 16, 2009 7:24 pm
Has thanked: 0 time
Been thanked: 0 time

Re: allocating volume M27 before M9

Postby steve-myers » Fri Sep 05, 2014 4:00 pm

samb01 wrote:Hello steve-myers.

Yes the volume are sms managed.

Can all or part of the data set exist in the cylinder managed area of the M27 volumes?


i'm not sure i andertood the question...

Every dataset can be allocate in the M27 volume.

Acutally the are dataset in the M27 Volume.
The storage space in "M27" volumes exists in two areas.
  • "Track managed" data where the cylinder number in each extent of a data set or free space is less than 65536.
  • "Cylinder managed" space, where the cylinder number of each extent of a data set or free space can be greater than 65536. Many programs have difficulty with this space.
Just as a comment, and it's probably worthless, in my opinion this whole concept for very large volumes is extremely clumsy. The fundamental problem is the 1960s idea of cylinder/head track address assignment where the cylinder number was limited to 65536 does not work well any more.

Although this idea has numerous problems, too, in my opinion IBM should have increased the number of tracks in a logical cylinder, to 512 perhaps, for these volumes. That way they could avoid the whole clumsy issue of two independent space maps. In a sense IBM did just that with "cylinder managed" space; the unit of allocation in "cylinder managed" space is 21 "cylinders," or 315 tracks.
steve-myers
Global moderator
 
Posts: 1885
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 197 times


Return to JCL

 


  • Related topics
    Replies
    Views
    Last post