IBM Mainframe Forum

Forget the LRECL. You have the average length (204) so you know how much input and output there will be.

Kolusu won't be back here until start of business on the West Coast, USA. I think that he will suggest that you send information to the DFSORT support team, who would give you the best possible advice with what you can supply to them.

Changing from tape to DASD is only going to affect the reading of the data, and the writing of the data, so is far from being a cure-all.

samb01 wrote:But i would like to know if the sort will work more quiclkly ?

Actually i have 12 tape on entry et 12 tape out.

I think it will be work fatser if the dataset is on disk. What do you think about it.

The problem is the size of the dataset : 712 000 000 records with 624 lrecl.

I calculate 50 disk model 9 (our disk are model 9). Actually we don't have enought disk... but we can ask for having more disk if we can win time by doing the sort ondisk. Only if it worth it.

What does quick mean? CPU time or elapsed time? DASD files doe NOT have the mount times , so you will be saving on that time. If you still need the job to be analyzed, Please add the following to your job so we can see the diagnostic mesages:

//SORTDIAG DD DUMMY

Then rerun the job and post the complete JES log including the //SYSOUT messages to dfsort@us.ibm.com

Hello,

i tryed the sort by getting rid of the sortwork. I want to let the system calculate the sortwork, but it doesn't work. The sysout of the job is in the attached file.



Why it doesn't work.

The value of DYNAUTO is YES.

samb01 wrote:Hello,

i tryed the sort by getting rid of the sortwork. I want to let the system calculate the sortwork, but it doesn't work. The sysout of the job is in the attached file.

Why it doesn't work.

The value of DYNAUTO is YES.

You serious? You haven't used any of my suggestions and your file size is enormous(146,411,286,840) with an average LRECL of 361 and you don't even have enough DASD for a single work dataset. You still wondering why your job failed?

i have DASD but i just want letting DFSORT calculate autmaticly DYNALLOC.
It's not possible ?

So OK, if it's not possible how can i calculate the number eand the values of the optimal sort work ?

Actually i am using 60 sortwork with the space : 100,200 cylinders.



It work but i wonder if it is optimised...

I suggest you actually try implementing what Kolusu specified on the first page of this topic - go back and read and obey. In there he is showing you how to do the dynamic allocation. Why post if you are not going to read and understand the replies?

i have DASD but i just want letting DFSORT calculate autmaticly DYNALLOC.
It's not possible ?

It may or may not be possible -- so far you have not followed instructions you have been given so you cannot say that DYNALLOC cannot be done on your system.

heelo,

i tryed but it doesn't work...

here is the sysout. Thank's for your help..

Just an other think, i thought the number of sortwork is limited to 32... So 60 sortworks is too much ?

samb01 wrote:heelo,

i tryed but it doesn't work...

here is the sysout.

Cut and paste in Code tags, kusomushi; we're not going to contaminate our or our clients' machines with your malware.

Just an other think, i thought the number of sortwork is limited to 32... So 60 sortworks is too much ?

If instead of trying to think you'd read the fine manual, you'd have seen that the maximum is 255 if Blockset is used.

Why are you saying it's a malware ?

It is just the sysout of the job....

The forum allow us to upload it, i am uusing it, that's all.