Decoupled Strategy for Imbalanced Workloads in MapReduce Frameworks

TL;DR

This paper proposes a decoupled strategy integrating MPI one-sided communication and non-blocking I/O to improve performance in imbalanced workloads within MapReduce frameworks, demonstrating up to 23% speedup.

Contribution

It introduces a novel decoupled approach that overlaps Map and Reduce phases using one-sided operations, enhancing efficiency in unbalanced workload scenarios.

Findings

Achieved up to 23% performance improvement.

Validated approach with Word-Count and PUMA dataset.

Effective in handling workload imbalance.

Abstract

In this work, we consider the integration of MPI one-sided communication and non-blocking I/O in HPC-centric MapReduce frameworks. Using a decoupled strategy, we aim to overlap the Map and Reduce phases of the algorithm by allowing processes to communicate and synchronize using solely one-sided operations. Hence, we effectively increase the performance in situations where the workload per process is unexpectedly unbalanced. Using a Word-Count implementation and a large dataset from the Purdue MapReduce Benchmarks Suite (PUMA), we demonstrate that our approach can provide up to 23% performance improvement on average compared to a reference MapReduce implementation that uses state-of-the-art MPI collective communication and I/O.