Demystifying Parallel and Distributed Deep Learning

Ben-Nun, T., & Hoefler, T. (2019). Demystifying Parallel and Distributed Deep Learning. ACM.

Type

Journal article

Intellectual base

Primary research

Year

2019

DOI

10.1145/3320060

Link

https://doi.org/10.1145/3320060

Cited by

1 citation across 1 artifact.

Each citation links to the sentence it supports in the citing article.

Domain-specific

• AI Infrastructure
  • Capacity cannot be inferred from accelerator count alone

    This sourceA systematic concurrency analysis of parallel and distributed deep learning, showing that achieved training throughput is governed by memory, communication and synchronisation structure rather than by accelerator count.

Verification

This reference passed the adversarial substantiation pipeline: it was checked to exist and to support the claim it is attached to. See how references were verified.

Registry ID ref:55cea51b9021 · see in the full table