We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.DC

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Distributed, Parallel, and Cluster Computing

Title: A Hybrid Parallelization Approach for Distributed and Scalable Deep Learning

Abstract: Recently, Deep Neural Networks (DNNs) have recorded great success in handling medical and other complex classification tasks. However, as the sizes of a DNN model and the available dataset increase, the training process becomes more complex and computationally intensive, which usually takes a longer time to complete. In this work, we have proposed a generic full end-to-end hybrid parallelization approach combining both model and data parallelism for efficiently distributed and scalable training of DNN models. We have also proposed a Genetic Algorithm based heuristic resources allocation mechanism (GABRA) for optimal distribution of partitions on the available GPUs for computing performance optimization. We have applied our proposed approach to a real use case based on 3D Residual Attention Deep Neural Network (3D-ResAttNet) for efficient Alzheimer Disease (AD) diagnosis on multiple GPUs. The experimental evaluation shows that the proposed approach is efficient and scalable, which achieves almost linear speedup with little or no differences in accuracy performance when compared with the existing non-parallel DNN models.
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Cite as: arXiv:2104.05035 [cs.DC]
  (or arXiv:2104.05035v2 [cs.DC] for this version)

Submission history

From: Samson Akintoye Dr. [view email]
[v1] Sun, 11 Apr 2021 15:42:38 GMT (1432kb,D)
[v2] Sun, 6 Feb 2022 13:34:21 GMT (1160kb,D)

Link back to: arXiv, form interface, contact.