DATE: Thursday, February 16, 2006
TIME: Noon - 1 pm
PLACE: CIC 2101

SPEAKER:
Adam Wierman
CMU

TITLE:
Open vs. Closed: A Cautionary Tale

ABSTRACT:
Every systems researcher is well aware of the importance of setting up one's experiments so that the system being modeled is "accurately represented." Fundamental to this is using a realistic workload generator that accurately models the service request demands, popularity distribution, etc. One particularly important aspect of a workload generator that is not given the attention it deserves is whether the workload generator uses an open or closed system model. Using a combination of implementation and simulation experiments, we will illustrate that there is a vast difference in behavior between the open and closed system models in realistic settings, even when all other workload parameters are held fixed. These results illustrate that understanding the appropriate system model for a given application is essential to evaluating the impact of design decisions, e.g. proposed changes in the scheduling policy used. Thus, we will also provide guidelines for determining when open and closed system models are appropriate.

BIO:
Adam Wierman is currently a doctoral student at Carnegie Mellon University. He received a BS with University Honors in Computer science and Mathematics with minors in Psychology and Statistics from Carnegie Mellon University in 2001. He is a recipient of the NSF Graduate Research Fellowship, the best student paper award at the ACM Sigmetrics conference, and multiple teaching awards, including the Alan J. Perlis Student Teaching Award. He currently works on the analysis of scheduling policies for queuing systems. His main focus is on understanding the impact of scheduling techniques and heuristics on efficiency and fairness metrics.

For Further Seminar Info:
, or visit http://www.pdl.cmu.edu/SDI/