CIS 888.08R: Research Seminar, Cluster Architecture


This research seminar is offered every quarter.

Instructor: Prof. M. Lauria

Presentation Schedule

The exact list of reading topics will be decided at the beginning of class; suggestions are welcome. Here is the presentation schedule (please email me the dates on which you want to present a paper). Here the suggested list of papers to present.

Objectives

Cluster technology leverages microprocessors with GFlop/s processing rates, Gigabit/sec speed networks (Myrinet, Giganet, Gigabit Ethernet), multi-GB RAM memories, and commodity operating systems to build scalable, low cost parallel computers. Cluster at different institutions can be pooled together to execute larger, more demanding applications, following the Computational Grid model.

Novel classes of applications are becoming popular in the high performance computing (HPC) arena, such as Computational Biology problems involving large amounts of data (genome assemblies, whole genome comparisons, protein folding and other proteomics applications), Geophysics applications (processing of satellite and interplanetary probe data), High Energy Physics (handling od data from high energy physics experiments).  These new applications pose new challenges to the designer of systems for high end computing, particularly in the areas of  network and I/O susbsystems performance, the handling (transfer, storage, processing) of large data sets, and the effective scheduling of data intensive jobs.

By participating in this seminar, the student can participate to our interesting research projects on the above issues. The topic of the specific project to be carried out by the student is chosen by me together with the student based on his/her background and preferences. Participation of students from other disciplines (biology, physics, engineering, etc) and having computationally challenging problems to suggest is strongly encouraged. The projects will be carried out on our Datacluster and/or on clusters at other institutions if appropriate.
 

Prerequisites

Permission of instructor
 

Materials

Papers from the literature + hands-on research projects in the Datacluster Lab.
 

Student with disabilities

Any student who feels he or she may need an accomodation based on the impact of a disability should contact me privately to discuss his or her specific needs. Please contact the Office for Disability Services at (614) 292-3307, or visit 150 Pomerene Hall, to coordinate reasonable accomodations for students with documented disabilities.


Last Updated: Sept 9, 2002