Cloudless-Training: A Framework to Improve Efficiency of Geo-Distributed ML Training

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tan, Wenting, Shi1, Xiao, Lv, Cunchi, Zhao, Xiaofang
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913063345586176
author Tan, Wenting
Shi1, Xiao
Lv, Cunchi
Zhao, Xiaofang
author_facet Tan, Wenting
Shi1, Xiao
Lv, Cunchi
Zhao, Xiaofang
contents Geo-distributed ML training can benefit many emerging ML scenarios (e.g., large model training, federated learning) with multi-regional cloud resources and wide area network. However, its efficiency is limited due to 2 challenges. First, efficient elastic scheduling of multi-regional cloud resources is usually missing, affecting resource utilization and performance of training. Second, training communication on WAN is still the main overhead, easily subjected to low bandwidth and high fluctuations of WAN. In this paper, we propose a framework, Cloudless-Training, to realize efficient PS-based geo-distributed ML training in 3 aspects. First, it uses a two-layer architecture with control and physical training planes to support elastic scheduling and communication for multi-regional clouds in a serverless maner.Second, it provides an elastic scheduling strategy that can deploy training workflows adaptively according to the heterogeneity of available cloud resources and distribution of pre-existing training datasets. Third, it provides 2 new synchronization strategies for training partitions among clouds, including asynchronous SGD with gradient accumulation (ASGD-GA) and inter-PS model averaging (MA). It is implemented with OpenFaaS and evaluated on Tencent Cloud. Experiments show that Cloudless-Training can support general ML training in a geo-distributed way, greatly improve resource utilization (e.g., 9.2%-24.0% training cost reduction) and synchronization efficiency (e.g., 1.7x training speedup over baseline at most) with model correctness guarantees.
format Preprint
id arxiv_https___arxiv_org_abs_2303_05330
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Cloudless-Training: A Framework to Improve Efficiency of Geo-Distributed ML Training
Tan, Wenting
Shi1, Xiao
Lv, Cunchi
Zhao, Xiaofang
Distributed, Parallel, and Cluster Computing
Artificial Intelligence
Geo-distributed ML training can benefit many emerging ML scenarios (e.g., large model training, federated learning) with multi-regional cloud resources and wide area network. However, its efficiency is limited due to 2 challenges. First, efficient elastic scheduling of multi-regional cloud resources is usually missing, affecting resource utilization and performance of training. Second, training communication on WAN is still the main overhead, easily subjected to low bandwidth and high fluctuations of WAN. In this paper, we propose a framework, Cloudless-Training, to realize efficient PS-based geo-distributed ML training in 3 aspects. First, it uses a two-layer architecture with control and physical training planes to support elastic scheduling and communication for multi-regional clouds in a serverless maner.Second, it provides an elastic scheduling strategy that can deploy training workflows adaptively according to the heterogeneity of available cloud resources and distribution of pre-existing training datasets. Third, it provides 2 new synchronization strategies for training partitions among clouds, including asynchronous SGD with gradient accumulation (ASGD-GA) and inter-PS model averaging (MA). It is implemented with OpenFaaS and evaluated on Tencent Cloud. Experiments show that Cloudless-Training can support general ML training in a geo-distributed way, greatly improve resource utilization (e.g., 9.2%-24.0% training cost reduction) and synchronization efficiency (e.g., 1.7x training speedup over baseline at most) with model correctness guarantees.
title Cloudless-Training: A Framework to Improve Efficiency of Geo-Distributed ML Training
topic Distributed, Parallel, and Cluster Computing
Artificial Intelligence
url https://arxiv.org/abs/2303.05330