Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

Dealing with big database

Reply
New Contributor
Posts: 2

Dealing with big database

I am connecting SAS EM with some large tables (>1 million records), and the speed of running each node become extremely slow. Even the data node or the graphic explore node become unacceptably slow. I am running some analysis on the whole datasets, and I can't only run a sample. Could anyone give me some suggestion about how to accelerate the job?

Thank you in advance!

Frequent Contributor
Posts: 95

Re: Dealing with big database

Posted in reply to martinmuller

How many variables are there in your data set? You may need to work with less number of variables than what is in your data set to make it work.

Try creating another data set with let's say 1000 variables.We had a similar issue with over 5000 variables and 12 milliions observations and we ended up spliting the data set into 7 sets.

Ask a Question
Discussion stats
  • 1 reply
  • 183 views
  • 0 likes
  • 2 in conversation