IBM Mainframe Forum Index
 
Log In
 
IBM Mainframe Forum Index Mainframe: Search IBM Mainframe Forum: FAQ Register
 

To find out size allocated to a sequential dataset


IBM Mainframe Forums -> JCL & VSAM
Post new topic   Reply to topic
View previous topic :: View next topic  
Author Message
ashek15

New User


Joined: 20 Oct 2015
Posts: 48
Location: USA

PostPosted: Thu Apr 27, 2017 9:42 am
Reply with quote

Dear All

Is there a way to check how much actual size was allocated to a sequential dataset which is used in a jcl which ran. Lets say for example allocation was SPACE=(CYL,(200,300),RLSE). Then Is there a way to know that out of total allocated size how much was actually allocated/used during jcl run. Just to mention the dataset is deleted after every run so we can't check actual allocated size in ispf 3.4 option.

Thanks
Back to top
View user's profile Send private message
Willy Jensen

Active Member


Joined: 01 Sep 2015
Posts: 712
Location: Denmark

PostPosted: Thu Apr 27, 2017 1:51 pm
Reply with quote

Not directly as far as I am aware. DCOLLECT dataset data have some information. The SMFDCBBL of SMF type 15 is the count of blocks written, as I read it, and with a known blocksize you can calculate the number of tracks.
Back to top
View user's profile Send private message
vasanthz

Global Moderator


Joined: 28 Aug 2007
Posts: 1742
Location: Tirupur, India

PostPosted: Thu Apr 27, 2017 10:38 pm
Reply with quote

I believe DCOLLECT cannot get that information as the dataset is deleted at the end of the JCL.
If DCOLLECT runs at the exact moment the dataset was allocated to its fullest size, then it would be able to get the information. Which is highly unlikely.

Also SMF 15 does not have the allocated space of a dataset, it does have primary and secondary space, but not the actual allocated space.

As for the answer to ashek15, I don't know :-(
Back to top
View user's profile Send private message
Rohit Umarjikar

Global Moderator


Joined: 21 Sep 2010
Posts: 3049
Location: NYC,USA

PostPosted: Fri Apr 28, 2017 12:01 am
Reply with quote

May be someone could help your better, if you tell us why you need that anyways even after DS is deleted?
Back to top
View user's profile Send private message
Robert Sample

Global Moderator


Joined: 06 Jun 2008
Posts: 8696
Location: Dubuque, Iowa, USA

PostPosted: Fri Apr 28, 2017 1:03 am
Reply with quote

The easiest way to collect this information is to have the application program write out the record count as part of its final tasks.
Back to top
View user's profile Send private message
ashek15

New User


Joined: 20 Oct 2015
Posts: 48
Location: USA

PostPosted: Fri Apr 28, 2017 1:27 am
Reply with quote

Thanks for your replies. To answer Rohit's query, we need this info to ensure we can track/monitor the sequential dataset usage and proactively allocate the appropriate space so that any future job failures can be avoided due to B37, D37 etc.
Back to top
View user's profile Send private message
Willy Jensen

Active Member


Joined: 01 Sep 2015
Posts: 712
Location: Denmark

PostPosted: Fri Apr 28, 2017 1:43 am
Reply with quote

@vasanthz - re 'I believe DCOLLECT cannot get that information as the dataset is deleted at the end of the JCL'
You are of course quite correct, I answered before thinking it through.
Back to top
View user's profile Send private message
mucka

New User


Joined: 11 Apr 2017
Posts: 5
Location: USA

PostPosted: Fri Apr 28, 2017 5:54 pm
Reply with quote

Could you just keep one version of the dataset and get your information and then delete it manually?
Back to top
View user's profile Send private message
ashek15

New User


Joined: 20 Oct 2015
Posts: 48
Location: USA

PostPosted: Sat Apr 29, 2017 2:23 am
Reply with quote

Actually these are many prod jobs for which we need the allocation details and hence it's not feasible to change all of them just for getting dataset allocation details and then again change them to their current state. Moreover, we don't really want to keep these datasets occupying space on dasd as these are temporary datasets and should be deleted as soon as jcl execution is completed. Hence looking for other feasible options. Thanks.
Back to top
View user's profile Send private message
Nic Clouston

Global Moderator


Joined: 10 May 2007
Posts: 2455
Location: Hampshire, UK

PostPosted: Sat Apr 29, 2017 2:29 am
Reply with quote

Why not temporarily add an IDCAMS step at the end of the job (before data set deletion) to list all the data sets and their properties, used in the job?
Back to top
View user's profile Send private message
Robert Sample

Global Moderator


Joined: 06 Jun 2008
Posts: 8696
Location: Dubuque, Iowa, USA

PostPosted: Sat Apr 29, 2017 2:40 am
Reply with quote

Quote:
Hence looking for other feasible options.
The usual method these days is to create a storage pool for work data sets. Allocate an appropriate number of volumes and let the system manage the space. How many x37 abends a week (or month) does your site have that there is a need to do what you're doing?

The basic problem you have is that for temporary data sets, there is NO easy way on z/OS to tell how much disk space they use while the job is executing, and when the data set is deleted when the job is complete you cannot go back and find out the space needed. Your storage management site support group should be involved, and they can size the storage pool based on all the applications running on the system.
Back to top
View user's profile Send private message
Willy Jensen

Active Member


Joined: 01 Sep 2015
Posts: 712
Location: Denmark

PostPosted: Sat Apr 29, 2017 3:35 pm
Reply with quote

If you really dont know the initial size and your goal is to save temporary space, then consider changing the allocation to say from (CYL,(200,300)) to (CYL,(100,400)), so that the initial allocation is smaller. And going back to SMF type 15, if you can find the count of blocks written then you might be able to calculate the space used.
Back to top
View user's profile Send private message
David Robinson

Active User


Joined: 21 Dec 2011
Posts: 199
Location: UK

PostPosted: Sun Apr 30, 2017 9:15 pm
Reply with quote

SMF records will show you how many extents the dataset has used. From that it's easy to assess whether you are likely to be running out of space any time soon.

That's the approach I have used in a couple of different sites. Pick a percentage, say 50% or 75%, and if 's using more than that % of the maximum possible extents it could have, raise an alert.

Note that the SMF record will also show you how many volumes it can extend across. So, if the volume count is 5, you know that the dataset could have 80 extents, for example.
Back to top
View user's profile Send private message
vasanthz

Global Moderator


Joined: 28 Aug 2007
Posts: 1742
Location: Tirupur, India

PostPosted: Mon May 01, 2017 12:18 am
Reply with quote

Hi David what smf record types?
Back to top
View user's profile Send private message
steve-myers

Active Member


Joined: 30 Nov 2013
Posts: 917
Location: The Universe

PostPosted: Mon May 01, 2017 1:13 am
Reply with quote

vasanthz - The only SMF record type we've been discussing here is the type 15 record.

There is one field of possible interest, though I don't know how reliable it is: SMF14NTU. Don't worry about the 14 in the label; type 14 and 15 are basically identical so type 15 records are mapped by type 14 symbols. I use types 14 and 15 in some of my programs, but I have never used SMF14NTU.
Back to top
View user's profile Send private message
Rohit Umarjikar

Global Moderator


Joined: 21 Sep 2010
Posts: 3049
Location: NYC,USA

PostPosted: Mon May 01, 2017 10:56 pm
Reply with quote

Quote:
we need this info to ensure we can track/monitor the sequential dataset usage and proactively allocate the appropriate space so that any future job failures can be avoided due to B37, D37 etc.

1.You can give maximum size in secondary allocations based on the stats of your earlier failures and use MXIG and RLSE to release unused space.
2.However, You also may need to analyze the size of the input dataset and allocate same or more to your work data.
3.DATACLAS are site specific, so you need to speak to your storage people if you could use that and then don't worry about 1,2 or anything for that matter.
Back to top
View user's profile Send private message
View previous topic :: :: View next topic  
Post new topic   Reply to topic View Bookmarks
All times are GMT + 6 Hours
Forum Index -> JCL & VSAM

 


Similar Topics
Topic Forum Replies
No new posts FINDREP - Only first record from give... DFSORT/ICETOOL 3
No new posts To find whether record count are true... DFSORT/ICETOOL 6
No new posts Map Vols and Problem Dataset All Other Mainframe Topics 2
No new posts Allocated cylinders of a dataset DB2 12
No new posts Sort First/last record of a subset th... DFSORT/ICETOOL 7
Search our Forums:

Back to Top