Evaluating the Impact of Java Virtual Machines on Energy Consumption

International audienceBackground. The Java Virtual Machine (JVM) platforms have known multiple evolutions along the last decades to enhance both the performance they exhibit and the features they offer. With regards to energy consumption, few studies have investigated the energy consumption of code and data structures. Yet, we keep missing an evaluation of the energy efficiency of existing JVM platforms and an identification of the configurations that minimize the energy consumption of software hosted on the JVM. Aims. The purpose of this paper is to investigate the variations in energy consumption between different JVM distributions and parameters to help developers configuring the least consuming environment for their Java application. Method. We thus assess the energy consumption of some of the most popular and supported JVM platforms using 12 Java benchmarks that explore different performance objectives. Moreover, we investigate the impact of the different JVM parameters and configurations on the energy consumption of software. Results. Our results show that some JVM platforms can exhibit up to 100% more energy consumption. JVM configurations can also play a substantial role to reduce the energy consumption during the software execution. Interestingly, the default configuration of the garbage collector was energy efficient in only 50% of our experiments. Conclusion. Finally, we provide an OSS tool, named J-Referral that recommends an energy-efficient JVM distribution and configuration for any Java application