Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CodeGreen: Towards Improving Precision and Portability in Software Energy Measurement

Saurabhsingh Rajput, Tushar Sharma

Abstract

Accurate software energy measurement is critical for optimizing energy, yet existing profilers force a trade-off between measurement accuracy and overhead due to tight coupling with supported specific hardware or languages. We present CodeGreen, a modular energy measurement platform that decouples instrumentation from measurement via an asynchronous producer-consumer architecture. We implement a Native Energy Measurement Backend (NEMB) that polls hardware sensors (Intel RAPL, NVIDIA NVML, AMD ROCm) independently, while lightweight timestamp markers enable tunable granularity. CodeGreen leverages Tree-sitter AST queries for automated instrumentation across Python, C++, C, and Java, with straightforward extension to any Tree-sitter-supported grammar, enabling developers to target specific scopes (loops, methods, classes) without manual intervention. Validation against "Computer Language Benchmarks Game" demonstrates $R^2 = 0.9934$ correlation with RAPL ground truth and $R^2 = 0.9997$ energy-workload linearity. By bridging fine-grained measurement precision with cross-platform portability, CodeGreen enables practical algorithmic energy optimization across heterogeneous environments. Source code, video demonstration, and documentation for the tool are publicly available at: https://smart-dal.github.io/codegreen/.

CodeGreen: Towards Improving Precision and Portability in Software Energy Measurement

Abstract

Accurate software energy measurement is critical for optimizing energy, yet existing profilers force a trade-off between measurement accuracy and overhead due to tight coupling with supported specific hardware or languages. We present CodeGreen, a modular energy measurement platform that decouples instrumentation from measurement via an asynchronous producer-consumer architecture. We implement a Native Energy Measurement Backend (NEMB) that polls hardware sensors (Intel RAPL, NVIDIA NVML, AMD ROCm) independently, while lightweight timestamp markers enable tunable granularity. CodeGreen leverages Tree-sitter AST queries for automated instrumentation across Python, C++, C, and Java, with straightforward extension to any Tree-sitter-supported grammar, enabling developers to target specific scopes (loops, methods, classes) without manual intervention. Validation against "Computer Language Benchmarks Game" demonstrates correlation with RAPL ground truth and energy-workload linearity. By bridging fine-grained measurement precision with cross-platform portability, CodeGreen enables practical algorithmic energy optimization across heterogeneous environments. Source code, video demonstration, and documentation for the tool are publicly available at: https://smart-dal.github.io/codegreen/.
Paper Structure (20 sections, 12 figures, 1 table)

This paper contains 20 sections, 12 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Comparative Tool Analysis
  4. Design and Implementation
  5. Validation and Benchmarking
  6. Implications
  7. Threats to Validity
  8. Conclusions
  9. Appendix: Tool Demonstration Walkthrough
  10. Availability
  11. Quick Start Guide
  12. Step 1: Installation
  13. Available Commands:
  14. Step 2: Initialize Hardware Sensors
  15. Step 3: Validate with Built-in Benchmark
  16. ...and 5 more sections

Figures (12)

  • Figure 1: CodeGreen architecture
  • Figure 2: CodeGreen Website: The official website at https://smart-dal.github.io/codegreen/ provides installation guides, API references, usage examples, and demo videos.
  • Figure 3: Installation: Running ./install.sh automatically detects the host architecture, compiles NEMB, and installs Python bindings and CLI tools.
  • Figure 4: Command Overview: The codegreen --help command displays the complete CLI interface, categorizing subcommands into measurement (measure, benchmark), analysis (analyze), and system management (init-sensors, config, doctor, info) tasks.
  • Figure 5: Sensor Initialization: The codegreen init-sensors command detects Intel RAPL, NVIDIA NVML, and AMD ROCm interfaces, configuring non-root access permissions.
  • ...and 7 more figures