Cloud-Scale Java Profiling at Alibaba

Fangxi Yin, Denghui Dong, Chuansheng Lu, Tongbao Zhang, Sanhong Li, Jianmei Guo, Kingsum Chow

Abstract: On the 2017 Double 11 Global Shopping Festival, Alibaba's cloud platform achieved total sales of more than 25 billion dollars and supported peak volumes of 325,000 transactions and 256,000 payments per second. Most of the cloud-based e-commerce transactions were processed by hundreds of thousands of Java applications with above a billion lines of code. It is challenging to achieve comprehensive and efficient performance profiling for large-scale, cloud-based Java applications in production. We developed ZProfiler, a fine-grained, low-overhead Java performance profiler. ZProfiler allows developers to load a profiling agent on the fly without restarting Java virtual machines, and its profiling information also facilitates code warmup. ZProfiler is developed based on Alibaba JDK (AJDK), a customized version of OpenJDK, and it has been rolled out to Alibaba's cloud platform to support large-scale performance tuning for online critical business.