A Curriculum Initiative on Parallel and Distributed Computing - Toward Core Topics for Undergraduates

Georgia State University Research Foundation, Inc., Atlanta GA

Investigators

Abstract

Parallel and Distributed Computing (PDC) now permeates most computing activities. The penetration of this technology in the daily lives has resulted in common users relying on its effectiveness and reliability. The mass marketing of multicores and general-purpose graphics processing units in home and office PCs and laptops has a potential for empowering even common users to become technology contributors. Certainly, it is no longer sufficient for even basic programmers to acquire only the conventional programming skills. All these phenomena point to the need for imparting a broad-based skill set in parallel and distributed computing at various levels, impacting Computer Science (CS) and Computer Engineering (CE) programs and related computational disciplines. However, the rapid change in computing hardware platforms and devices, languages, and supporting programming environments, and the research advances, more than ever challenges the educators in deciding what to teach in any given semester. Students and their future employers face similar challenges on what constitutes basic expertise. Our vision is bringing all stakeholder experts working together and periodically providing guidance on restructuring standard curriculum across various courses and modules related to parallel and distributed computing. Immediate benefit would be providing CS/CE students and their instructors with periodic guidelines on which aspects to cover in which courses. New programs at colleges (nationally and internationally) will receive guidance in setting up courses and/or integrating parallelism in the Computer Science or Engineering or Computational Science curriculum. Employers would have well defined expectations from students on PDC skills. This will similarly help in retraining and certifications of existing professionals. As background preparation, a planning workshop was held in Feb 2010 to explore the state of curriculum in parallel and distributed education, assess the needs, provide action plans and recommend mechanisms for how best to address the curricular needs in both the short and long terms. The planning workshop and its related set of activities have drawn experts from various stakeholders. The primary task is to propose a set of core topics in parallel and distributed computing for undergraduate curriculum for CS/CE students. For various topics in the three sub-areas of Programming, Algorithms, and Architecture, the expected deliverables are (i) learning outcomes, (ii) level of coverage (Bloom's classification), and (iii) examples of how to teach each topic. In summer 2010, we aim to carry out a public release of preliminary curriculum, collect feedback from various stakeholders, and identify and recruit colleagues who will try out aspects the proposed curriculum during fall 2010 and spring 2011. Following these, during these two semesters, the activities will include monitoring and supporting the courses and feedback collection from instructors and students. Our larger goal is to enable students to be fully prepared for their future careers in light of the technological shifts and mass marketing of multicores and GPUs and to make real impact on all the stakeholders including employers, authors, and educators. This curricular guidance and its trajectory, along with periodic feedback and other evaluation data on adoption and use, will also help steer companies hiring students and interns, hardware and software vendors, and, of course, the authors and researchers.