Dealing with JVM Limitations in Apache Cassandra

The Cassandra distributed database runs on the JVM, which is a huge productivity boost in many ways. However, the more you care about performance, the less you can afford to let abstractions such as I/O and memory management exist as black boxes.

This session covers how Cassandra has gone beyond the interfaces provided by the JVM to achieve higher performance in three areas: (1) how Cassandra uses platform-specific features such as posix_fadvise, munmap, and mlock; (2) how memory management is tuned to avoid pain points in real-world Java garbage collection; and (3) how Cassandra uses the JDK's instrumentation hooks to measure the real size of on-heap structures and make better decisions about what to keep in memory.

Copyright © 2013 Oracle and/or its affiliates. Oracle® is a registered trademark of Oracle and/or its affiliates. All rights reserved. Oracle disclaims any warranties or representations as to the accuracy or completeness of this recording, demonstration, and/or written materials (the "Materials"). The Materials are provided "as is" without any warranty of any kind, either express or implied, including without limitation warranties of merchantability, fitness for a particular purpose, and non-infringement.