Big data technologies allow for the processing of very large data sets and deliver the promise of new, innovative solutions.
However, consistent, low latency performance isn’t by definition a guarantee for every Java-based big data application. In-depth attention needs to be given to product choices, deployment topologies and runtime components in order to maximize the value of these new big data solutions, including Apache Cassandra and supporting components such as Spark and Solr.
This benchmark study compares the response time performance of two JVMs, Azul Zing and Oracle HotSpot, while running Apache Cassandra at different throughput levels. Benchmark results were obtained using the Cassandra-stress framework, a utility for load testing and benchmarking a Cassandra cluster, and the jHiccup JVM measurement tool.
The benchmark was configured with one server and three Cassandra data nodes. Testing methodology design aimed to measure JVM response time consistency based on percentiles and Cassandra application max outliers at a variety of throughput rates. The jHiccup measurement tool was used to obtain and graph the response time distribution of the two JVMs for all the benchmark runs.
Download the benchmark study to learn about the performance results.