CISE/CNS/CSR PEC 7354 Org Code 05050000 $25,000 There is an urgent need for curricular guidance on parallel and distributed computing (PDC) due to rapid technological changes and mass marketing of multicores and general-purpose graphics processing units. Educators struggle to decide the right balance between theory and practice, and to choose from the diverse set of models of computation, languages, software and hardware platforms, and tools.

The working group that has participants from various institutions has deliberated upon various topics and subtopics, and released a preliminary version of the curriculum in December 2010. A competition for granting Early Adopters status was held in Spring-11. The proposing instructors will be employing and evaluating the proposed curriculum in their courses. The primary task ahead is to organize a follow-up curriculum workshop (EduPar-11 at IPDPS) to bring together early adopters and other experts, collect feedback from the early adopters and the community, revise the preliminary version of the curriculum, and hold a second round of competition for Fall-11 for early adopter status.

Project Report

Outcomes: A working group composed of researchers from academia, government, and industry has formulated the first proposed core curriculum on parallel and distributed computing (PDC). The goal of this effort is to ensure that all students graduating with a bachelors' degree in computer science/computer engineering receive an education that fully prepares them for careers in a field where technological advances occur at a daunting pace. By means of the curriculum, students and their instructors will receive periodically updated guidelines that identify aspects of PDC that are important to cover, and that suggest specific core courses in which their coverage might find an appropriate context. Currently, 55 early-adopter institutions worldwide are trying out this curriculum. The early adopters have been awarded stipends through three rounds of competitions (Spring and Fall 2011, and Spring 2012) with support from NSF, Intel, and NVIDIA. Additional competitions will be held in Fall 2012 and annually thereafter. This fund supported (i) the travel of the Spring-11 early adopters to the EduPar-11 workshop where they came together to provide their feedback, and (ii) the stipends for the Fall-11 early adopter competition. Curriculum Initiative Website: www.cs.gsu.edu/~tcpp/curriculum/index.php Impact/benefits: Through the curriculum, students and their instructors will receive periodic guidelines that indentify aspects of PDC that are important to cover, and suggest specific core courses in which their coverage might find an appropriate context. New programs at colleges (nationally and internationally) will receive guidance in setting up courses or integrating parallelism into the institutions’ current curriculums. Employers will also benefit by using the new curriculum model to gain a better sense of what they can expect from students in the area of parallel and distributed computing skills. Curriculum guidelines will similarly provide guidance to companies retraining or seeking certifications for existing professionals. The CS2013 ACM/IEEE Computer Science Curriculum Joint Task Force has recognized PDC (along with security) as a main thrust area. We are closely interacting with the Task Force, providing expert feedback on the PDC portion of their initial draft on PDC in Oct, 2011. Detailed Activities: The first step of our work on PDC was a planning workshop sponsored by NSF and IEEE/TCPP in Feb., 2010. Based on the outcome of that workshop, a working group has taken up the challenge of developing (and justifying) PDC curricular guidelines for CS and CE undergraduates, with particular emphasis on identifying a core curriculum that represents what every graduate should know about PDC. Throughout 2010, the working group has deliberated upon various topics and subtopics, specifying both the expected minimum level of coverage and the desired learning outcomes; we employed Bloom's classification as the medium for expressing the desired level of expertise on a topic. The group also developed suggestions on how to teach each topic, and guidance on where each topic could potentially be incorporated into a core course. Recognizing the differing needs of different types of institutions in different cultural and geographical settings, the group committed itself not to be prescriptive in its recommendations, but rather to provide alternatives with rationales for each. A preliminary version of the core curriculum was released in Dec, 2010. The Early Adopter Competitions: A major mechanism for evaluating a snapshot of our guidelines are the Early Adopter competitions that we have thus far run three times: in Spring 2011, using funds from the original grants from NSF (for US institutions) and Intel (for international institutions), and in Fall 2011 (this grant) and Spring 2012, supplemented by GPU card donations from NVIDIA. Aspiring Early Adopters submit a proposal that is evaluated by a committee from the taskforce. We selected 16, 18, and 21 institutions, respectively, in the Spring 2011, Fall 2011 and Spring 2012 competitions; we were able to award an average of $1.5K/institution. Faculty associated with the selected proposals are employing our initial curriculum guidelines in one or more courses at their respective institutions. The EduPar Workshop: In order to allow the Early Adopters, the public, and the working group to benefit from everyone's experiences and evaluations, we organized the first EduPar Workshop, collocated with IPDPS in Anchorage, Alaska in May 2011, to bring together the Early Adopters (this grant) and others interested in PDC education - primarily to receive the feedback from the Adopters, but also to stimulate discussion of curricular and other educational issues. The inaugural EduPar workshop - the first workshop devoted exclusively to educational matters at IPDPS - was a great success, with attendance in the range of 40-80 throughout the single day of the event. EduPar’12 was just held at Shanghai in May as a regular IPDPS’12 satellite workshop, with similar attendance. The proceedings of the EduPar-11 workshop is available at www.cs.gsu.edu/~tcpp/curriculum/?q=node/16950 and the video coverage of all presentations by Intel is available at http://techtalks.tv/events/53/.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Communication Foundations (CCF)
Type
Standard Grant (Standard)
Application #
1135124
Program Officer
Balasubramanian Kalyanasundaram
Project Start
Project End
Budget Start
2011-05-01
Budget End
2012-04-30
Support Year
Fiscal Year
2011
Total Cost
$49,998
Indirect Cost
Name
Georgia State University Research Foundation, Inc.
Department
Type
DUNS #
City
Atlanta
State
GA
Country
United States
Zip Code
30303