Abstract:

This research article explores the significance of integrating high-performance computing (HPC) into lower-level computer science curriculum and the implementation of a web-enhanced tool called the Virtual Data Science Lab (VDSL) to facilitate this educational endeavor. With the growing demand for computational skills in various industries, it is essential to equip students with the knowledge and practical experience required to leverage HPC resources effectively. This article discusses the benefits of introducing HPC early in the curriculum and presents a framework and the early effort for establishing virtual data science labs to foster hands-on learning experiences and supportive interactions between instructors and students. Additionally, potential challenges and considerations for implementing this approach are addressed, alongside recommendations for instructors and educational institutions aiming to enhance computer science education.

The rest of this article is organized as follows:

Section I will introduce the increasing reliance on computational techniques across diverse domains highlighting the need for incorporating high-performance computing (HPC) in computer science education. This section also talks about the importance of integrating HPC into lower-level computer science curriculum and the benefit of a virtual data science lab as a pedagogical tool. It emphasizes the development of critical computational skills, fostering problem-solving abilities, and enhancing students' understanding of parallel processing and performance optimization.

Section II delves into our early effort in developing virtual data science labs using a High-Performance Computing facility and the exertion in providing practical experiences to students. The proposed VDSL is a web-enhanced tool that assists students’ learning in data science. The tool intends to offer rich learning experiences for students in lower-level computer science programs, therefore helping them to build confidence and prepare them for future needs in data science, machine learning, and other computational tasks. It is not just another IDE. It is a free tool for all students enrolled in the designated courses and provides automatic and interactive feedback that allows students to learn by doing, with step-by-step guidance on analyzing and visualizing data. Interactive feedback will allow students to track their progress and receive targeted feedback on their work, helping them to improve their skills and knowledge over time.

Section III presents a framework for establishing VDSL within the computer science curriculum. It outlines the necessary infrastructure, software tools, and resources required to create a conducive virtual learning environment. Additionally, considerations for curriculum integration and assessment methodologies are addressed. This section will also consider potential challenges and considerations in implementing virtual data science labs and integrating HPC in lower-level computer science curricula. It addresses factors such as resource constraints, technical requirements, and instructor training.

Finally, Section IV concludes by providing practical recommendations for instructors and educational institutions aiming to introduce HPC and VDSL. We also present a survey of case studies and success stories from educational institutions that have successfully implemented HPC and virtual data science labs in their lower-level computer science curriculum.

Keywords:

High-Performance Computing, HPC education, Virtual Data Science Lab, Students Engagement.