DJXPerf: Identifying Memory Inefficiencies via Object-centric Profiling for Java

TL;DR

DJXPerf is a lightweight Java memory profiler that links hardware performance metrics to Java objects, helping optimize memory locality and improve application performance with minimal overhead.

Contribution

It introduces a novel object-centric profiling technique that associates hardware memory metrics with Java objects without requiring source code or hardware modifications.

Findings

Achieves ~8% runtime overhead and ~5% memory overhead.

Guides optimization of object allocation, layout, and access patterns.

Demonstrates significant speedups on benchmarks and real-world applications.

Abstract

Java is the "go-to" programming language choice for developing scalable enterprise cloud applications. In such systems, even a few percent CPU time savings can offer a significant competitive advantage and cost saving. Although performance tools abound in Java, those that focus on the data locality in the memory hierarchy are rare. In this paper, we present DJXPerf, a lightweight, object-centric memory profiler for Java, which associates memory-hierarchy performance metrics (e.g., cache/TLB misses) with Java objects. DJXPerf uses statistical sampling of hardware performance monitoring counters to attribute metrics to not only source code locations but also Java objects. DJXPerf presents Java object allocation contexts combined with their usage contexts and presents them ordered by the poor locality behaviors. DJXPerf's performance measurement, object attribution, and presentation techniques guide optimizing object allocation, layout, and access patterns. DJXPerf incurs only ~8% runtime overhead and ~5% memory overhead on average, requiring no modifications to hardware, OS, Java virtual machine, or application source code, which makes it attractive to use in production. Guided by DJXPerf, we study and optimize a number of Java and Scala programs, including well-known benchmarks and real-world applications, and demonstrate significant speedups.