INDUSTRIAL PANEL:  PERFORMANCE  ANALYSIS IN THE REAL WORLD

Panel description

What issues are on the minds of industrial performance analysts?  Four representatives of world-class product organizations will describe their work at the front lines of measurement, modeling, and performance tuning.  Topics will include performance engineering of middleware at IBM, tools for detecting false sharing in large-scale multiprocessors at Hewlett-Packard, kernel thread-scheduling performance in multiprocessors at Microsoft, and low-overhead instrumentation for profiling large-scale services at Google.  Plenty of time will be available to ask questions about how to direct our research to have the greatest impact on industrial practice.

Panelists

Curtis Hrischuk is a Lead Performance Engineer for IBM in the WebSphere performance team.  The products he has optimized range in size from embedded devices to J2EE application server clusters, with an emphasis on the optimization of new emerging technologies.  He has worked as the Lead Performance Engineer for SIP for the last three years and is also working with soft real-time Java.  He has published in several international, refereed journals and conferences.  His won a Best Paper Award at the Computer Measurement Groups International conference in  2006    He completed the M.Eng. (1995) and Ph.D. (1998) degrees from Carleton University, Ottawa, Canada in Systems and Computer Engineering.  He is also an Apologist for the Reasons to Believe organization.
 
Chris Ruemmler is a Master level software engineer responsible for HP-UX performance in the Enterprise Servers and Storage division at Hewlett-Packard.  He has worked at HP for over fifteen years measuring and improving the performance of HP-UX and applications that run under HP-UX.  He specializes in MP synchronization, disk I/O performance, database performance, and fully utilizing the features of the Intel Itanium processor.  He has a B.S. degree in Electrical Engineering and Computer Science and an M.S. degree in Computer Science from the University of California, Berkeley.  His current interests include the TPC-E benchmark, solid state disk technology, and server blade technology.
 
Ben Sigelman has been a software engineer and technical lead at Google since 2003.  His recent work focuses on the transparent monitoring of large-scale distributed systems and on building new performance tools for applications which utilize tens of thousands of host machines.  In his earlier days at Google, Ben built significant pieces of distributed infrastructure for the AdWords system.  He has an Sc.B. degree in Mathematics and Computer Science from Brown University.  When he's not accidentally taking down major Google services, Ben passes his time playing the cello, banjo, and fiddle semi-professionally.  
 
Brad Waters is a Development Manager at Microsoft in the Windows Core Fundamentals Development Team where he focuses on Windows Server kernel and multi-core performance and scalability.  Brad joined Microsoft in 1995 with the primary goals of improving Windows NT scalability, performance and enabling Windows to be used efficiently in all major server scenarios.  Brad currently leads the Windows Server performance kernel and multi-core team working with key architects, researchers and developers at Microsoft to prepare for future multi-core systems and continually improving Windows Server scalability and performance.  Brad has a Masters degree in Computer Science from New York University and a Bachelor degree in Computer Science from North Carolina State University.  Brad has worked in software engineering field for over 25 years at Microsoft, Digital Equipment Corp, A Large Commercial Bank in NYC, The City of New York, IBM and ITT Telecommunications.