Improving performance of MapReduce framework on InterCloud by avoiding transmission of unnecessary data

Authors:
Seungmi Choi;Shingyu Kim;Hyuck Han;Heon Y. Yeom
Affiliations:
Seoul National University, Seoul, South Korea;Seoul National University, Seoul, South Korea;Seoul National University, Seoul, South Korea;Seoul National University, Seoul, South Korea
Venue:
ACM SIGMETRICS Performance Evaluation Review
Year:
2011

Citing 5
Cited 0

Space/time trade-offs in hash coding with allowable errors

Communications of the ACM
MapReduce: simplified data processing on large clusters

Communications of the ACM - 50th anniversary issue: 1958 - 2008
What's inside the Cloud? An architectural map of the Cloud landscape

CLOUD '09 Proceedings of the 2009 ICSE Workshop on Software Engineering Challenges of Cloud Computing
Blueprint for the Intercloud - Protocols and Formats for Cloud Computing Interoperability

ICIW '09 Proceedings of the 2009 Fourth International Conference on Internet and Web Applications and Services
Scatter-Gather-Merge: An efficient star-join query processing algorithm for data-parallel frameworks

Cluster Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recently, cloud computing and MapReduce have become attractive computing environments as the need for data processing rapidly increases. This inflation in the demand for resources led to the introduction of InterCloud, which provisions resources from two or more cloud service providers. However, when MapReduce applications run on InterCloud environments, they underperform contrary to our expectations. In this paper, we suggest a technique that mitigates the performance problem of MapReduce applications on InterCloud environments. As shown in our preliminary results, our technique leads to up to five times better performance in InterCloud environments.