NCSI: Parallel and Cluster Computing

Registration Closed
Parallel and Cluster Computing
Second Look — 2010
NCSI's Parallel and Cluster Computing workshops focus on teaching faculty how to move from traditional desktop computing to modern high-performance hardware. Offerings range from end-user focused applications ready-designed for HPC to the details of creating applications designed to run on clusters, many-core machines, shared-memory machines, and graphics processing units. The material is designed for undergraduate faculty from a variety of disciplines who would like to add parallel computing to their undergraduate teaching and research. In addition, undergraduate and graduate students are encouraged to attend alongside a sponsoring faculty member. The workshop is hands-on, with exercises in both computing and curriculum development.

Institution	Kean University Union, NJ
Dates	Jun 27 - Jul 2
Local Coordinators	David Joiner
Lead Instructors	David Joiner and Russ Manson
Notify By	May 31
Details	This workshop will focus on helping faculty with the process of moving their research from the Desktop to High Performance Computing systems. With a focus on applications, participants will gain experience using a variety of HPC tools on a 130 node, 1040 core cluster at Kean and advanced visualization hardware. Much of the content will be cross-disciplinary, focusing on the details of running problems on HPC hardware (technical issues related to logging on and executing programs, schedulers, and realistic expectations of speedup,) as well as visualization tools. Participants are encouraged to bring their own problems to the workshop. A pre-workshop poll will be conducted with participants ahead of time to identify specific programs which should be demonstrated. Topics at the workshop will include an overview of current HPC technology and protocols, realistic expectations of speedup for common algorithms on multi-core and many-core hardware, standard protocols used for scheduling and running jobs on professional HPC systems, cross-disciplinary mathematical tools, and visualization tools. Additionally, specialized content will be selected based on applicant interest in a variety of parallel sessions, using domain specific applications in computational chemistry, physics, biology, and engineering.