AIM Banners_978 x 90

Automating model parallelism with just one line of code

The difference between these two approaches maps naturally to the heterogeneity of a typical compute cluster.
Researchers from Google, Amazon Web Services, UC Berkeley, Shanghai Jiao Tong University, Duke University and Carnegie Mellon University have published a paper titled “Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learning” at OSDI 2022. The paper introduces a new method for automating the complex process of parallelising a model with only one line of code. So how does Alpa work? Model parallelism Data parallelism is a technique where model weights are duplicated across accelerators while only partitioning and distributing the training data. The dataset is split into ‘N’ parts in data parallelism with ‘N’ being the quantity of GPUs. The parts are assigned to parallel computational machines after which gradients are calculated for each copy of
Subscribe or log in to Continue Reading

Uncompromising innovation. Timeless influence. Your support powers the future of independent tech journalism.

Already have an account? Sign In.

📣 Want to advertise in AIM? Book here

Picture of Avi Gopani
Avi Gopani
Avi Gopani is a technology journalist that seeks to analyse industry trends and developments from an interdisciplinary perspective at Analytics India Magazine. Her articles chronicle cultural, political and social stories that are curated with a focus on the evolving technologies of artificial intelligence and data analytics.
Related Posts
AIM Print and TV
Don’t Miss the Next Big Shift in AI.
Get one year subscription for ₹5999
Download the easiest way to
stay informed