Distributed GraphLab: A Framework for Machine Learning in the Cloud

TL;DR

This paper presents Distributed GraphLab, a framework for scalable, efficient machine learning in the cloud, extending the original GraphLab to distributed systems with data consistency, fault tolerance, and significant performance improvements.

Contribution

It introduces a distributed extension of the GraphLab framework that maintains data consistency, reduces network congestion, and incorporates fault tolerance for large-scale machine learning.

Findings

Achieves 10-100x performance gains over Hadoop-based systems.

Successfully implements fault tolerance using Chandy-Lamport snapshots.

Demonstrates scalability on Amazon EC2 with large datasets.

Abstract

While high-level data parallel frameworks, like MapReduce, simplify the design and implementation of large-scale data processing systems, they do not naturally or efficiently support many important data mining and machine learning algorithms and can lead to inefficient learning systems. To help fill this critical void, we introduced the GraphLab abstraction which naturally expresses asynchronous, dynamic, graph-parallel computation while ensuring data consistency and achieving a high degree of parallel performance in the shared-memory setting. In this paper, we extend the GraphLab framework to the substantially more challenging distributed setting while preserving strong data consistency guarantees. We develop graph based extensions to pipelined locking and data versioning to reduce network congestion and mitigate the effect of network latency. We also introduce fault tolerance to the GraphLab abstraction using the classic Chandy-Lamport snapshot algorithm and demonstrate how it can be easily implemented by exploiting the GraphLab abstraction itself. Finally, we evaluate our distributed implementation of the GraphLab abstraction on a large Amazon EC2 deployment and show 1-2 orders of magnitude performance gains over Hadoop-based implementations.