COMPSs GPU Cache DNN Distributed Training Version 1
Download RO CrateName: Dislib Distributed Training - Cache ON
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).
Dataset: Imagenet
Version dislib-0.9
Version PyTorch 1.7.1+cu101
Average task execution time: 36 seconds
Click and drag the diagram to pan, double click or use the controls to zoom.
Version History
Version 1 (earliest) Created 25th Mar 2024 at 11:27 by Cristian Tatu
Frozen
Version-1887f42c
Creators and SubmitterCreator
Additional credit
The Workflows and Distributed Computing Team (https://www.bsc.es/discover-bsc/organisation/scientific-structure/workflows-and-distributed-computing/)
Submitter
Views: 296
Created: 25th Mar 2024 at 11:27
TagsThis item has not yet been tagged.
AttributionsNone
Related items
https://orcid.org/0009-0003-8848-9436
eFlows4HPC project aims at providing workflow software stack and an additional set of services to enable the integration of HPC simulations and modelling with big data analytics and machine learning in scientific and industrial applications. The project is also developing the HPC Workflows as a Service (HPCWaaS) methodology that aims at providing tools to simplify the development, deployment, execution and reuse of workflows. The project demonstrates its advances through three application Pillars ...
Teams: Cluster Emergent del Cervell Humà, Workflows and Distributed Computing, Pillar I: Manufacturing, Pillar II: Climate, Pillar III: Urgent computing for natural hazards, eFlows4HPC general, COMPSs Tutorials
Web page: https://eflows4hpc.eu
Distributed computing aims to offer tools and mechanisms that enable the sharing, selection, and aggregation of a wide variety of geographically distributed computational resources in a transparent way. The research done in this team is based on the past expertise of the group, and on extending it towards the aspects of distributed computing that can benefit from this expertise. The team at BSC has a strong focus on programming models and resource management and scheduling in distributed computing ...
