Interpreting Performance Profiles with Deep Learning

TL;DR

This paper introduces a deep learning-based system that combines performance profiling data with code semantics to help software engineers interpret program inefficiencies more effectively, especially in Java applications.

Contribution

The work integrates code summarization with performance profiles using a fine-tuned CodeBERT model to improve understanding of program inefficiencies.

Findings

Effective assistance in analyzing Java benchmark performance.

Improved interpretation of profiling data through semantic integration.

Enhanced user interface for visualizing code summaries and profiles.

Abstract

Profiling tools (also known as profilers) play an important role in understanding program performance at runtime, such as hotspots, bottlenecks, and inefficiencies. While profilers have been proven to be useful, they give extra burden to software engineers. Software engineers, as the users, are responsible to interpret the complex performance data and identify actionable optimization in program source code. However, it can be challenging for users to associate inefficiencies with the program semantics, especially if the users are not the authors of the code, which limits the applicability of profilers. In this thesis, we explore a new direction to combine performance profiles and program semantics with a deep learning approach. The key idea is to glean code summary for semantic information (at a certain level) and integrate it into a profiler, which can better understand program inefficiencies for actionable optimization. To be concrete, we combine profiles generated by Async Profiler (the state-of-the-art Java profiler) with code summarization from a fine-tuned CodeBERT-based model. We demonstrate the code summaries of any selected call path in a graphic user interface. Our system can effectively assist analysis on many Java benchmarks.