Document Type
Conference Proceeding
Publication Date
11-1-2020
Abstract
Although recent scaling up approaches to train deep neural networks have proven to be effective, the computational intensity of large and complex models, as well as the availability of large-scale datasets require deep learning frameworks to utilize scaling out techniques. Parallelization approaches and distribution requirements are not considered in the primary designs of most available distributed deep learning frameworks and most of them still are not able to perform effective and efficient fine-grained inter-node communication. We present Phylanx that has the potential to alleviate these shortcomings. Phylanx presents a productivity-oriented frontend where user Python code is translated to a futurized execution tree that can be executed efficiently on multiple nodes using the C++ standard library for parallelism and concurrency (HPX), leveraging fine-grained threading and an active messaging task-based runtime system.
Publication Source (Journal or Book title)
Proceedings of DLS 2020: Deep Learning on Supercomputers, Held in conjunction with SC 2020: The International Conference for High Performance Computing, Networking, Storage and Analysis
First Page
20
Last Page
30
Recommended Citation
Hasheminezhad, B., Shirzad, S., Wu, N., Diehl, P., Schulz, H., & Kaiser, H. (2020). Towards a Scalable and Distributed Infrastructure for Deep Learning Applications. Proceedings of DLS 2020: Deep Learning on Supercomputers, Held in conjunction with SC 2020: The International Conference for High Performance Computing, Networking, Storage and Analysis, 20-30. https://doi.org/10.1109/DLS51937.2020.00008