Versatile, Portable, and Efficient OS Profiling via Latency Analysis
Operating systems are complex and their behavior depends on many factors.
Source code, if available, does not directly help understand the OS's
behavior, as the behavior depends on actual workloads and external inputs.
Runtime profiling is a key technique to prove new concepts, debug problems,
and optimize performance. Unfortunately, existing profiling methods are
lacking in important areas-they do not provide enough necessary information
about the OS's behavior, they require OS modification and therefore are not
portable, or they exact high overheads thus perturbing the profiled OS.
OSprof is a versatile, portable, and efficient OS profiling
method based on latency distributions analysis. OSprof automatically
selects important profiles for subsequent visual analysis. We have
demonstrated that a suitable workload can be used to profile virtually
any OS component. OSprof is portable because it can intercept operations
and measure OS behavior from user-level or from inside the kernel without
requiring source code. OSprof has typical CPU time overheads below 4%.
||Title (click for html version)
||Versatile, Portable, and Efficient File System Profiling
||PS PDF BibTeX ||Stony Brook U. CS TechReport FSL-06-05
||Ph.D. Thesis: an expanded version of our OSDI 2006 paper.
||Name (click for home page)
||Jan 2004 - Dec 2006
||Research Staff Member, Storage and Data Services Research group, IBM T. J. Watson Research Center (Hawthorne, NY)
||Sep 2003 - Aug 2008
||Research Staff Member, Storage Systems and Performance Management group, IBM Tel Aviv Research Lab (Tel-Aviv, Israel)
||Charles P. Wright
||May 2003 - May 2006
||Software Developer, Eladian Partners, LLC (New York, NY)
||Rakesh Narayan Iyer
||Sep 2004 - Dec 2005
||Storage Software engineer, Software Storage, NVidia (Santa Clara, CA)
||Jan 2007 - Dec 2007
||Member of Technical Staff, Engineering Development, Riverbed Technology (Sunnyvale, CA)
(Last updated: Tue May 21 11:08:00 EDT 2013)