BookmarkSubscribeRSS Feed
Shakti_Sourav
Quartz | Level 8

Hy,

I have one job which is name is Kalia_Job. and its taking 4 hours to execute, in kalia Job 3 datasets are used, per dataset have 50 lakhs record. and also i have used INSERTBUFF=32767 in table loader transformation. 

 

Kindly suggest me how to incraeses the performance ?

2 REPLIES 2
ErikLund_Jensen
Rhodochrosite | Level 12

Hi @Shakti_Sourav 

 

In order to help with this question, a lot more information is needed. You have 3 input data sets which are processed in a series of transformations to end with a table loader. The log contains information about the time used for every step in the job, and it is impossible to answer your question without knowing which transformation is slow running. It could be in the data preparation steps or the table loader. So please provide a full log and and a screenshot of the loader settings + information about the output libname. Is it SAS or a DBMS?

LinusH
Tourmaline | Level 20

Beside the tips from @ErikLund_Jensen , some bottlenecks can be hidden because the extensive and default use of views in DI Studio.

If you find a branch in your job that takes long time, temporarily change outputs to be tables instead. By doing so, you will in more detail see what step consumes the most resources.

A part from this, you can consider this as optimizing normal SAS code. Use options to further track whats's going on (FULLSTIMER, MSGLEVEL=i, _method PROC SQL option, and SASTRACE options for SAS/ACCESS processing).

Data never sleeps

SAS Innovate 2025: Register Now

Registration is now open for SAS Innovate 2025 , our biggest and most exciting global event of the year! Join us in Orlando, FL, May 6-9.
Sign up by Dec. 31 to get the 2024 rate of just $495.
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
  • 2 replies
  • 727 views
  • 0 likes
  • 3 in conversation