Experience with benchmarking dependability and performance of MapReduce systems

MapReduce provides a convenient means for distributed data processing and automatic parallel execution on clusters of machines. It has various applications and is used by several services featuring fault tolerance and scalability. Many studies investigated the dependability and performance of MapRed...

Full description

Saved in:
Bibliographic Details
Published in:Performance evaluation 2016-07, Vol.101, p.1-19
Main Authors: Sangroya, Amit, Bouchenak, Sara, Serrano, Damián
Format: Article
Language:English
Subjects:
Automation
Benchmarking
Clouds
Clusters
Computer Science
Dependability
Distributed, Parallel, and Cluster Computing
Emerging Technologies
Fault tolerance
Hadoop
MapReduce
Mathematical models
Modulus of rupture in bending
Statistics
Systems and Control
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:MapReduce provides a convenient means for distributed data processing and automatic parallel execution on clusters of machines. It has various applications and is used by several services featuring fault tolerance and scalability. Many studies investigated the dependability and performance of MapReduce, ranging from job scheduling to data placement and replication, adaptive and on-demand fault tolerance to new fault tolerance models. However, the ad-hoc and overly simplified setting used to evaluate most MapReduce fault tolerance and performance improvement solutions poses significant challenges to the analysis and comparison of the effectiveness of these solutions. The paper precisely addresses this issue and presents MRBS, a comprehensive benchmark suite for evaluating the dependability and performance of MapReduce systems. MRBS includes five benchmarks covering several application domains and a wide range of execution scenarios such as data-intensive vs. compute-intensive applications, or batch applications vs. online interactive applications. MRBS allows to inject various workloads, dataloads and faultloads, and produces extensive reliability, availability and performance statistics. We implemented the MRBS benchmark suite for Hadoop MapReduce, and we illustrate its use with various case studies running on Amazon EC2 and on a private cloud.
ISSN:0166-5316
1872-745X
DOI:10.1016/j.peva.2016.04.001