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 Extracting Flexible, Replayable Models from Large Block Traces BibTeX Tenth USENIX Conference on File and Storage Technologies (FAST 2012) Feb 2012 To appear
2 Don't Thrash: How to Cache your Hash on Flash PS PDF BibTeX 3rd USENIX Workshop in Hot Topics in Storage and File Systems (HotStorage 2011) Jun 2011  
3 Benchmarking File System Benchmarking: It *IS* Rocket Science PS PDF BibTeX 13th USENIX Workshop in Hot Topics in Operating Systems (HotOS XIII) May 2011  
4 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
2 Amar Mudrankit MS Jan 2011

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 Koundinya Santhosh Kumar MS Sep 2010 - Dec 2011 Software Engineer, Fusion-io (Alviso, CA)
4 Gyumin Sim MS Jan 2010 - Dec 2010 Software Engineer, Data Center Power Team Google (Mountain View, CA)
5 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: Sat Jan 21 16:54:01 EST 2012)