BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
econ
Quartz | Level 8

Is there an easy way to do this without having to modify the sasv9.cfg file?

 

I'm working with a large table and am running out space in work.

 

Thanks.

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
econ
Quartz | Level 8

This option doesn't change where the util files are created. The util files are created in another directory in work.

 

I found a workaround. These util files are being created due to processes which take advantage of multi-threading, i.e. proc sort and proc sql. I turned off threads in base SAS using options nothreads and now no util files are being created.

 

It's a little crazy you can be working with a 350GB table and can tripple the disk space usage to sort a table. 

View solution in original post

7 REPLIES 7
ballardw
Super User

Identify a location where you have more space with read/write privileges, should be a different disk most likely.

 

Options USER="path to that location";

 

Note that all one level dataset name references will be looking to USER.

 

If you don't have access to another disk then this is likely moot and you need more resources.

 

Also look to see if you have old temporary directories in your work location taking up space. If you have had SAS crashes for any reason you may some. Delete them to free up additional space.

econ
Quartz | Level 8

This option doesn't change where the util files are created. The util files are created in another directory in work.

 

I found a workaround. These util files are being created due to processes which take advantage of multi-threading, i.e. proc sort and proc sql. I turned off threads in base SAS using options nothreads and now no util files are being created.

 

It's a little crazy you can be working with a 350GB table and can tripple the disk space usage to sort a table. 

SASKiwi
PROC Star

You may want to consider the COMPRESS SAS option as well if your table has a large number of columns. Try:

 

options COMPRESS = BINARY;

 

and you might be surprised with how much space you save including WORK datasets. You may find it speeds up your processing if your IO is a bottleneck.  

econ
Quartz | Level 8

Compress is slow. Work is all SSD for my setup. I have tons of SAN space but it is slow past a certain point.

 

The thing I learned today is if you don't specify options nothreads, be prepared to use up some disk space.

SASKiwi
PROC Star

In many SAS server environments COMPRESS is often set to be the default, including ours. It actually speeds up our processing because IO is the usual bottleneck not CPU.

 

As always you need to test if it works in your own environment. If it slows down your jobs that suggests CPU is a bottleneck for you.

LinusH
Tourmaline | Level 20

Sorry for crashing in late.

Perhaps this solution fits your needs, but I think I would like to add another solution, for those who still wish to make use of SAS threading capabilities.

Use the UTILLOC option, where you can specify a separate location for your utility files:

http://support.sas.com/documentation/cdl/en/lesysoptsref/68023/HTML/default/p1texr4rxo0ipyn1ovajj11r...

 

Data never sleeps
econ
Quartz | Level 8

This is what i was trying to avoid using. You have to modify the sasv9.cfg file. This is managed by my admin, so I would need to bug him when I needed it changed rather than doing it myself.

sas-innovate-2024.png

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.

 

Register now!

How to connect to databases in SAS Viya

Need to connect to databases in SAS Viya? SAS’ David Ghan shows you two methods – via SAS/ACCESS LIBNAME and SAS Data Connector SASLIBS – in this video.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 7 replies
  • 2386 views
  • 0 likes
  • 4 in conversation