CSE 920:  Selected Topics in High-Performance Computing

Spring 2007 Topic:   BIOLOGICALLY-INSPIRED DISTRIBUTED COMPUTING

Instructor: Dr. Philip McKinley

Time: TuTh 2:40-4:00pm
Room: 1145 EB
         
MOTIVATIONS

The rapidly increasing complexity of computing technology, and its increasing interaction with the physical environment, has produced a flurry of exciting research that addresses how we can use knowledge of natural organisms as a guide to building robust and self-managing computational systems.  These "biologically-inspired" approaches include neural networks, genetic algorithms, biomimetics, and digital evolution.  In recent years, these methods and others have been applied to a wide variety of problems in distributed systems: protecting computers from cyber-attack through artificial immune systems; mimicking emergent behaviors in species such as ants and bees (biomimetics) to accomplish collective tasks in swarms of micro-robots; modeling the autonomic nervous system in the design of self-managing server farms; designing robust and adaptable logical topologies for overlay networks; and many others.

This course will study key research contributions in the area of biologically-inspired distributed computing, focusing primarily on techniques and technologies applied in the subareas of computer security, communication protocols, and robotics.  The course provides a forum for group discussion and analysis of important papers from each subarea.  The meeting times will be devoted primarily to presentation and discussion of papers from the literature.  Students will be required to give class presentations of selected research papers.  Each student will also conduct a term project or write a term survey paper related to topics discussed in the course.

COURSE STRUCTURE

 The course is intended to provide a forum for group discussion of research results. Hence, the meeting times for this course will be devoted primarily to presentation and discussion of papers from the literature.  Students will be required to give class presentations and to write summaries and/or critiques of research papers.  Each student will also conduct a term project or write a term survey paper related to topics discussed in the course.
GRADING FORMULA
Presentations
 45%
Critiques/Writeups 20%
Class Participation 10%
Term Project/Paper 25%


PREREQUISITES
This course is intended for graduate students interested in autonomic computing.  An old prerequisite, CSE 822, does not apply.  Students who want to enroll should send email to mckinley@cse.msu.edu.