MiMatrix: A Massively Distributed Deep Learning Framework on a Petascale High-density Heterogeneous Cluster

Explore this paper's citation graph

Summary

This paper develops and implements a novel job server parallel software framework, named by "MiMatrix", for distributed deep learning training, and proposes a novel GPUDirect Remote direct memory access~(RDMA)-aware parallel algorithm of AllReucde executed by computing servers.

Type
article
Published
2018-02-07
Cited by
0
References
42
Access
Open access

Keywords

Computer science, Petascale computing, Bottleneck, Distributed computing, Server

References

Cited by

No citing papers recorded for this paper.

Related papers