Baunsgaard commented on PR #1843: URL: https://github.com/apache/systemds/pull/1843#issuecomment-1714176241
> <img alt="peformance_eval" width="604" src="https://user-images.githubusercontent.com/68165368/267070022-2f5105d5-36dc-4ce4-bbc4-eab517940105.png";> > > here is a result from the performance evaluation on my local machine. I am little bit unsure about the intel mkl, i used this by setting MKL_NUM_THREADS = 4 as I found that it works best for my machine. It was little bit surprising that although I have intel chip, OpenBlas significantly performed better. This already looks promising, please run the tests with repeats such that you get the average run time of ~100 runs. also run it both with a sparse input and a dense input. Thanks! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
![https://user-images.githubusercontent.com/68165368/267070022-2f5105d5-36dc-4ce4-bbc4-eab517940105.png"](https://user-images.githubusercontent.com/68165368/267070022-2f5105d5-36dc-4ce4-bbc4-eab517940105.png"){kind=link}