[GitHub] [incubator-mxnet] hkvision commented on issue #17822: [Question] Distributed training performance for one worker and one server on the same node

[GitBox <gi...@apache.org>]

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:
users@infra.apache.org


With regards,
Apache Git Services