2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum (2011)
Anchorage, Alaska USA
May 16, 2011 to May 20, 2011
An integral part of many data-intensive applications is the need to collect and analyze enormous datasets efficiently. Concurrent with such application needs is the increasing adoption of MapReduce as a programming model for processing large datasets using a cluster of machines. Current MapReduce systems, however, require the data set to be loaded into the cluster before running analytical queries, and thereby incur high delays to start query processing. Furthermore, existing systems are geared towards batch processing. In this paper, we seek to answer a fundamental question: what architectural changes are necessary to bring the benefits of the MapReduce computation model to incremental, one-pass analytics, i.e., to support stream processing and online aggregation? To answer this question, we first conduct a detailed empirical performance study of current MapReduce implementations including Hadoop and MapReduce Online using a variety of workloads. By doing so, we identify several drawbacks of existing systems for one-pass analytics. Based on the insights from our study, we conclude by listing key design requirements and arguing for architectural changes of MapReduce systems to overcome their current limitations and fully embrace incremental one-pass analytics and showing promising preliminary results.
Y. Diao, E. Mazur, B. Li and P. Shenoy, "Towards Scalable One-Pass Analytics Using MapReduce," 2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum(IPDPSW), Anchorage, Alaska USA, 2011, pp. 1102-1111.