Ammar Ahmad Awan et al.: Optimized Broadcast for Deep Learning Workloads on Dense-GPU InfiniBand Clusters: MPI or NCCL? (2017)journals/corr/AwanCSP171707.09414Optimized Broadcast for Deep Learning Workloads on Dense-GPU InfiniBand Clusters: MPI or NCCL?4Ammar Ahmad Awan1Ching-Hsiang Chu2Hari Subramoni3Dhabaleswar K. Panda 00014CoRRCoRRabs/1707.094142017provenance information for RDF data of dblp record 'journals/corr/AwanCSP17'2021-06-10T11:42:04+0200