Local SGD with periodic averaging: Tighter analysis and adaptive synchronization

Farzin Haddadpour, Mohammad Mahdi Kamani, Mehrdad Mahdavi, Viveck R. Cadambe

Research output: Contribution to journalConference articlepeer-review

5 Scopus citations

Abstract

Communication overhead is one of the key challenges that hinders the scalability of distributed optimization algorithms. In this paper, we study local distributed SGD, where data is partitioned among computation nodes, and the computation nodes perform local updates with periodically exchanging the model among the workers to perform averaging. While local SGD is empirically shown to provide promising results, a theoretical understanding of its performance remains open. We strengthen convergence analysis for local SGD, and show that local SGD can be far less expensive and applied far more generally than current theory suggests. Specifically, we show that for loss functions that satisfy the Polyak-Lojasiewicz condition, O((pT)1/3) rounds of communication suffice to achieve a linear speed up, that is, an error of O(1/pT), where T is the total number of model updates at each worker. This is in contrast with previous work which required higher number of communication rounds, as well as was limited to strongly convex loss functions, for a similar asymptotic performance. We also develop an adaptive synchronization scheme that provides a general condition for linear speed up. Finally, we validate the theory with experimental results, running over AWS EC2 clouds and an internal GPU cluster.

Original languageEnglish (US)
JournalAdvances in Neural Information Processing Systems
Volume32
StatePublished - 2019
Event33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019 - Vancouver, Canada
Duration: Dec 8 2019Dec 14 2019

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Fingerprint Dive into the research topics of 'Local SGD with periodic averaging: Tighter analysis and adaptive synchronization'. Together they form a unique fingerprint.

Cite this