Workload Characterization and Optimization of TPC-H Queries on Apache Spark

Besides being an in-memory oriented computing framework, Spark runs on top of a Java Virtual Machine (JVM), so JVM parameters must be tuned to improve Spark application performance. Misconfigured parameters and settings degrade performance. For example, using Java heaps that are too large often causes long garbage collection pause time, which accounts for over 10-20\% of application execution time. Moreover, recent modern computing nodes have many cores and support running multiple threads simultaneously with SMT technology. Thus, optimization in full stack is also important. Not only JVM parameters but also OS parameters, Spark configuration, and application code itself based on CPU characteristics need to be optimized to take full advantage of underlying computing resource. In this paper, we use TPC-H benchmark as our optimization case study and gather many perspective logs such as application log, JVM log such as GC and JIT, system utilization, and hardware events from PMU. We investigate the existing problems and then introduce several optimization approaches for accelerating Spark performance. As a result, our optimization achieves 30 - 40\% speed up on average, and up to 5x faster than the naive configuration.

By: Tatsuhiro Chiba and Tamiya Onodera

Published in: Proceedings of 2016 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) Piscataway, NJ,, IEEE, p.112-21 - 10.1109/ISPASS.2016.7482079 in 2016

LIMITED DISTRIBUTION NOTICE:

This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.

RT0968.pdf

Questions about this service can be mailed to reports@us.ibm.com .