COMPSs GPU Cache PyTorch CNN Distributed Training

**Name:** PyTorch CNN - Imagenet **Contact Person**: cristian.tatu@bsc.es **Access Level**: public **License Agreement**: Apache2 **Platform**: COMPSs **Machine**: Minotauro-MN4 PyTorch distributed training of CNN on GPU and leveraging COMPSs GPU Cache for deserialization speedup. Launched using 32 GPUs (16 nodes). Version dislib-0.9 Version PyTorch 1.7.1+cu101

Publisher

[<#ROCrate::Organization https://ror.org/05sd8tv96 @properties={"@id"=>"https://ror.org/05sd8tv96", "@type"=>"Organization", "name"=>"Barcelona Supercomputing Center"}>]

License

Apache-2.0

Contents

• main_pytorch_sync_5_nodes.py
  Size: 2107
  Format: text/plain
• complete_graph.svg
  Size: 167040
  Format: [["image/svg+xml", <#ROCrate::ContextualEntity https://www.nationalarchives.gov.uk/PRONOM/fmt/92 @properties={"@id"=>"https://www.nationalarchives.gov.uk/PRONOM/fmt/92", "@type"=>"WebSite", "name"=>"Scalable Vector Graphics"}>]]
• App_Profile.json
  Size: 4515
  Format: ["application/json", <#ROCrate::ContextualEntity https://www.nationalarchives.gov.uk/PRONOM/fmt/817 @properties={"@id"=>"https://www.nationalarchives.gov.uk/PRONOM/fmt/817", "@type"=>"WebSite", "name"=>"JSON Data Interchange Format"}>]
• compss_submission_command_line.txt
  Size: 769
  Format: text/plain
• ro-crate-info.yaml
  Size: 928
  Format: ["YAML", <#ROCrate::ContextualEntity https://www.nationalarchives.gov.uk/PRONOM/fmt/818 @properties={"@id"=>"https://www.nationalarchives.gov.uk/PRONOM/fmt/818", "@type"=>"WebSite", "name"=>"YAML"}>]
• dnn_cnn_5epochs.csv
  Size: 3277
• compss-1930568.out
  Size: 3012
  Format: text/plain
• compss-1930568.err
  Size: 64
  Format: text/plain