Early Exposure to Parallel Computing Through Curriculum Integration and Applied Projects
Wednesday, June 24, 2026 3:45 PM to 5:15 PM · 1 hr. 30 min. (Europe/Berlin)
Foyer D-G - 2nd Floor
Women in HPC Poster
Development of HPC SkillsEducation and Training
Information
Poster is on display and will be presented at the poster pitch session.
The growing demand for high-performance computing (HPC) skills across academia and industry continues to expose a gap in computing education, where parallel and performance-aware thinking is often introduced late in the curriculum or limited to specialized courses. This work presents an education-focused approach for introducing parallel and distributed computing (PDC) concepts early through intentional integration within core computing courses and reinforcement through applied, mentored student projects.
Shared-memory parallelism using OpenMP is embedded in algorithm-focused courses such as Data Structures and Algorithms and Design and Analysis of Algorithms. Rather than treating parallelism as a standalone topic, parallel concepts are introduced alongside familiar algorithmic material, allowing students to reason about performance, scalability, and trade-offs as part of regular problem solving. The approach emphasizes low-barrier tools and conceptual understanding, with students comparing sequential and parallel implementations instead of focusing on hardware-specific optimization.
To extend learning beyond the classroom, this curricular integration is supported by mentored projects involving students from data structures, object-oriented programming, and machine learning courses. Across multiple course offerings, students have applied OpenMP to projects in diverse application areas, including smart energy management and peak load optimization, as well as graph-based money laundering cycle detection. Mentoring focuses on identifying performance bottlenecks, incrementally introducing parallel constructs, and evaluating performance impact through empirical analysis.
Project outcomes and student reflections suggest increased confidence in parallel programming concepts, stronger engagement with performance-oriented thinking, and improved preparedness for internships and research-oriented roles. Complementary outreach and training activities in HPC further broaden exposure beyond formal coursework. Overall, this work highlights how early, integrated exposure to parallel computing, reinforced through applied projects and mentoring, can help make HPC concepts more accessible and support a more inclusive education-to-workforce pipeline.
The growing demand for high-performance computing (HPC) skills across academia and industry continues to expose a gap in computing education, where parallel and performance-aware thinking is often introduced late in the curriculum or limited to specialized courses. This work presents an education-focused approach for introducing parallel and distributed computing (PDC) concepts early through intentional integration within core computing courses and reinforcement through applied, mentored student projects.
Shared-memory parallelism using OpenMP is embedded in algorithm-focused courses such as Data Structures and Algorithms and Design and Analysis of Algorithms. Rather than treating parallelism as a standalone topic, parallel concepts are introduced alongside familiar algorithmic material, allowing students to reason about performance, scalability, and trade-offs as part of regular problem solving. The approach emphasizes low-barrier tools and conceptual understanding, with students comparing sequential and parallel implementations instead of focusing on hardware-specific optimization.
To extend learning beyond the classroom, this curricular integration is supported by mentored projects involving students from data structures, object-oriented programming, and machine learning courses. Across multiple course offerings, students have applied OpenMP to projects in diverse application areas, including smart energy management and peak load optimization, as well as graph-based money laundering cycle detection. Mentoring focuses on identifying performance bottlenecks, incrementally introducing parallel constructs, and evaluating performance impact through empirical analysis.
Project outcomes and student reflections suggest increased confidence in parallel programming concepts, stronger engagement with performance-oriented thinking, and improved preparedness for internships and research-oriented roles. Complementary outreach and training activities in HPC further broaden exposure beyond formal coursework. Overall, this work highlights how early, integrated exposure to parallel computing, reinforced through applied projects and mentoring, can help make HPC concepts more accessible and support a more inclusive education-to-workforce pipeline.
Format
on-demandon-site
