Performance Analysis for High Performance Computing Systems

The exponential disparity between the speeds of CPUs and storage systems continues to widen. Data-intensive computations may require hundreds of disks per CPU to utilize modern processors. Even previously CPU-bound workloads are now becoming I/O-bound. Our efforts are focused on finding and fixing I/O performance problems on high performance computing systems by creating a suite of tools to benchmark, trace, profile, analyze, and visualize file and storage systems.

This project is a collaboration with Klaus Mueller and Ethan L. Miller.

Conference and Workshop Papers:

#	Title (click for html version)	Formats	Published In	Date	Comments
1	DARC: Dynamic Analysis of Root Causes of Latency Distributions	PS PDF BibTeX	International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS 2008)	Jun 2008	Source code and benchmark information.

Current Students:

#	Name (click for home page)	Program	Member Since
1	Vasily Tarasov	PhD	Jan 2008

Past Students:

#	Name (click for home page)	Program	Period	Current Location
1	Nikolai Joukov	PhD	Jan 2004 - Dec 2006	Research Staff Member, Storage and Data Services Research group, IBM T. J. Watson Research Center (Hawthorne, NY)
2	Avishay Traeger	PhD	Sep 2003 - Aug 2008	Research Staff Member, Storage Systems and Performance Management group, IBM Tel Aviv Research Lab (Tel-Aviv, Israel)
3	Ivan Deras Tabora	MS	Jan 2007 - Dec 2007	Teacher, Computer Science, Universidad Tecnologica Centroamericana (San Pedro Sula, Cortes, Honduras)

Sponsors:

#	Sponsor	Amount	Period	Type	Title (click for award abstract)
1	NSF HECURA	$760,253	2006-2009	PI	File System Tracing, Replaying, Profiling, and Analysis on HEC Systems

(Last updated: Tue Mar 2 22:47:59 EST 2010)