Open Source Program Office

Taskflow

Project leader:

Tsung-Wei (TW) Huang; (tsung-wei.huang@wisc.edu)

Project description:

This project will enhance the unit testing facility of Taskflow, a general-purpose task-parallel programming system in modern C++. You will work with the Taskflow team (led by Dr. Tsung-Wei Huang in ECE) to design and implement comprehensive testing frameworks for verifying correctness, performance, and scalability of Taskflow’s core components. The project will involve creating automated test suites, integrating continuous integration (CI) tools, and improving test coverage for advanced features such as condition tasks, dynamic tasking, and GPU offloading.

To sustain the growth and reliability of Taskflow, this proposal aims to strengthen its continuous integration (CI) and testing infrastructure. Taskflow is a general-purpose task-parallel programming system using modern C++. Started in 2018 as a DARPA-sponsored research project, it has been evolving to a popular programming system in the C++ community. As Taskflow continues to attract users and contributors from academia and industry, ensuring robust, scalable, and comprehensive testing becomes essential. The proposed effort will focus on expanding the test coverage across heterogeneous environments, incorporating performance regression tracking, and automating validation workflows to detect subtle concurrency and portability issues. By advancing Taskflow’s CI ecosystem, this project will not only improve code quality and developer productivity but also ensure that Taskflow remains a dependable foundation for high-performance parallel programming in modern C++. Focused CI tasks are outlined below:

1. Performance Regression Tracking: Integrate automated benchmarking into CI to detect and report performance regressions on representative workloads, ensuring each commit maintains or improves execution efficiency.

2. Thread-Safety and Concurrency Stress Testing: Develop CI modules that run randomized stress tests to detect data races, deadlocks, or nondeterministic failures, using tools such as ThreadSanitizer and Valgrind’s Helgrind.

3. Static and Dynamic Analysis Integration: Incorporate static analyzers (e.g., clang-tidy, cppcheck) and dynamic tools (e.g., AddressSanitizer, UndefinedBehaviorSanitizer) into the CI workflow to proactively identify subtle bugs and undefined behavior.

4. Documentation and Example Validation: Automate the compilation and execution of Taskflow tutorials, examples, and documentation snippets to guarantee correctness and prevent divergence between code and documentation.

Intern needs:

C++, GitHub

Application Requirements:

1. Review the available projects by visiting the various project pages.
2. Interns should apply through the UW Student Jobs portal. Applicants who are not currently admitted or enrolled as a UW-Madison Student or without a UW NetID can login as well as create an account. Please note, you must apply to each project individually that you want to be considered for.
3. Application materials submitted through the UW Student Jobs portal should include:
• A one-page cover letter that highlights your qualifications based on skills identified in the project listing and your interest in open source broadly.
• A resume that includes your name, school email address, phone number, field(s) of study (major, minor, degree, certificate), relevant coursework, extracurricular activities, expected graduation date, relevant sample work (ex: GitHub link, personal website, etc.) and any relevant work or research experience.
• The names and contact information of three references.
4. Submit a resume, cover letter, and three references as part of your application.
5. Interviews will be arranged for selected candidates on a rolling basis after applications close.