In my Sas step, i have given SAS.WORK DD SPACE=(CYL,(4000,8000),RLSE) and i am getting space issue.
The address space has used a maximum of 640K below the line and 13924K above the line.
Can you please suggest how much we can declare in work space to process large file.
The SAS code itself will have a lot to do with the amount of storage space needed. There are issues like lengths of variables, compression, removing data after it is no longer needed, using views instead of data sets ... but since you asked a question about the JCL we'll have to skip all of the SAS-code-related issues.
Sometimes space is just not available on an IBM mainframe. You will need to know something about the disk pool clean-up routines at your site. There may be more space available on some days of the week, or there may be alternative disk packs/pools that you can allocate on the WORK DD statement. Also, if you know something about the size of the data set you will work with, you can allocate a larger primary extent and a smaller secondary extent. That way, the job won't even start unless the system can locate a large enough primary extent.
Good luck.
I would suggest removing that
,RLSE
Hi,
The JCL appears to be specifying work space, but the message appears to be related to memory, which is obviously something different.
Please show any SAS "ERROR:" message from the log and what was happening before it. The message you have shown just looks like a normal "NOTE:".
What is the job failure code? E.g. B37
If it is a memory (not space) problem then try adjusting your REGION= parameter on the job card to a higher value and see if that helps.
Regards,
Amir.
I don not see a volcnt definition. Code the multivolume parameter or let your storage admin have defined that using a dataclass definition.
Normally space on a mainframe is limited to 1 volume. 1 volume 3390 is about 3Gb (until ca 2004) I expect the volumesize being set to values like 9Gb today. You can check this value.
If the volume is used by others, it is a multiuser system no pc, less space will be left.
Getting to a Multivolume approach it will spread over many volumes. Please make each file of those not too big. The total amount of workspace can be the amount you have specified mulitplied by the numbers of volumes (volcnt).
Seen your message 640Kb below 13Mb up there is no out of memory resource constraint
The region-size is very confusing.
If you are allowed the specify 0M, that is the easiest as no limit.
IF you specify 4M or 6M (16Mb line) you are getting 32Mb additional above the line (total 38Mb).
The values 8M-16M do not work. Specifying above 16M you will get 8M under the line (total24Mn)
So specify 64Mb or more when running out of memory having specified 6Mb.
Getting java into that you need another 256Mb. From 2Gb-4G another gap in sizing of the memory area.
Small update. Your mainframe support staff could make the HFS options for sas-work available.
SAS(R) 9.3 Companion for z/OS, Second Edition See Ucount wiht tso alloc and the HFS (UFS) datasetnames
Thank all for the response.. i will try using Region= parameter.
My SAS error messages are
NOTE: EOV macro was not able to obtain an additional extent for library data set
number 7. System rc = 00000B37; Reason code = 00000004.
ERROR: Write to WORK.RFM.DATA failed. File is full and may be damaged.
ERROR: Write to WORK.RFM.DATA failed. File is full and may be damaged.
ERROR: Write to WORK.RFM.DATA failed. File is full and may be damaged.
NOTE: The SAS System stopped processing this step because of errors.
If i get this error message my return code is 08..
And another type of error message is
ERROR Writting into output file and I will get the RC as SB37.
Please let me know if you need any more information.
The error you have got is describing problems on the dasd / harddisk / external storaga (SB37) for sas work.
I will not help if you change settings on an other resource like the internal memory - dimm-chips (region size).
Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.
Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.