pytorch / FBGEMM

FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
Other
1.2k stars 494 forks source link

move set_cache and set_async to background thread #3033

Closed duduyi2013 closed 1 month ago

duduyi2013 commented 2 months ago

Summary: set cache is inserting values into L2 cache set_async it inserting values into SSD(rocksdb)

this 2 steps could be done asynchronously in the background thread wihtout blocking the read path. Actually, it could potentially block the next read if it take close to each train iteration time.

Differential Revision: D61418016

netlify[bot] commented 2 months ago

Deploy Preview for pytorch-fbgemm-docs ready!

Name Link
Latest commit a471f7d8eec9b46980a2e75341c0a93437fbb4e2
Latest deploy log https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/66d4902b45d8b70008e7b6b7
Deploy Preview https://deploy-preview-3033--pytorch-fbgemm-docs.netlify.app
Preview on mobile
Toggle QR Code...

QR Code

Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify site configuration.

facebook-github-bot commented 2 months ago

This pull request was exported from Phabricator. Differential Revision: D61418016

facebook-github-bot commented 2 months ago

This pull request was exported from Phabricator. Differential Revision: D61418016

facebook-github-bot commented 2 months ago

This pull request was exported from Phabricator. Differential Revision: D61418016

facebook-github-bot commented 1 month ago

This pull request was exported from Phabricator. Differential Revision: D61418016

facebook-github-bot commented 1 month ago

This pull request has been merged in pytorch/FBGEMM@886e5db1851f75e2297a0484bb3b6c6d1155a8f3.