hkvision commented on issue #17822: [Question] Distributed training performance for one worker and one server on the same node URL: https://github.com/apache/incubator-mxnet/issues/17822#issuecomment-598521256 > You probably want to check out https://medium.com/apache-mxnet/intel-mlsl-makes-distributed-training-with-mxnet-faster-7186ad245e81 Hi Eric @eric-haibin-lin Thanks for your reply. This post is about using mlsl with horovod for distributed training. It will be a good approach to try later. But currently I'm just using the traditional ps-lite approach with dist kvstore. Want to ask is there any related performance stats or practice for this?
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
