Worldwide Energy-Savings through Optimization of Widely Deployed Open-Source Software

Open source software that is deployed worldwide by the 100s of million instances bear a great energy efficiency potential. In this work, a suited open source software project shall be selected and an energy efficiency measure in software shall be developed for this project. The energy savings shall be measured and the potential worldwide savings shall be estimated.
Ideally, the changes to the code shall be returned into the open source software project for worldwide deployment.
A metric for the estimation of the worldwide deployment of specific open source projects shall be defined.

  • Tasks

    1. Select at least one interesting/suitable open source library.

    2. Research tools for optimizing code using source code analysis (e.g. Intel® VTune™ Performance Analyzer, etc.).

      • If necessary, develop a small analysis script that checks for a known inefficiency (bad loop, inefficient data structure, etc.).

    3. Find/define metrics and tools to quantify and repeatedly measure energy consumption.

    4. Create a test setup, e.g. a simple program that uses the selected library and runs through one or more reproducible processes.

    5. Use optimization tools and/or other code analysis methods to find areas in the library source where optimizations can be made.

    6. Test the optimizations using the test setup.

    7. Document the results.

  • Further Information

    • 40% theory, 60% implementation
    • Term or master’s thesis, 1-2 students
    • Prior knowledge recommended in
      • Software-Engineering
      • Java, Python, C and/or C++

Haben wir dein Interesse geweckt?

Ich bin an der Studienarbeit Worldwide Energy-Savings through Optimization of Widely Deployed Open-Source Software interessiert und möchte mehr erfahren.