Wenhao Jia
Electrical Engineering, Princeton University
Princeton, NJ 08544
201-***-**** / ****@*********.***
EDUCATION Doctor or Philosophy, Electrical Engineering
Princeton University, Princeton, NJ, expected August 2013
Concentration: Computer Engineering
GPA: 3.7/4.0
Bachelor of Engineering, Electrical Engineering
Tsinghua University, Beijing, China, 2008
EXPERIENCE Research Assistant Since 2009
Thesis Advisor: Margaret Martonosi, Princeton University, NJ
Improving microarchitectural features of the GPU memory system.
Developed an automated GPU design space exploration tool based on regression
analysis.
Worked on System-level ISA (SISA) project, in which multi-threaded LLVM
programs are annotated with control and data ow information for better run-
time core assignment and scratchpad allocation.
Software Engineering Intern Summer 2011
Google Inc., Mountain View, CA
Utilized hardware performance counters and GCC related tools to study x86
assembly code level performance variations across Intel and AMD CPU plat-
forms.
Searched for function-level code multi-versioning performance optimization op-
portunities and heuristics resulted from platform-speci c performance varia-
tions.
Research Intern Summer 2010
IBM T.J. Watson Research Center, Hawthorne, NY
Parallelized a business analytics application with X10 and optimized its perfor-
mance in an OpenMP based distributed computing environment.
Designed a parallel and distributed Cluster-Feature Tree building algorithm.
Teaching Assistant Fall 2009
Course: Computer Architecture, Princeton University, Princeton, NJ
PUBLICATIONS Stargazer: Automated Regression-Based GPU Design Space Exploration
Wenhao Jia, Kelly A. Shaw, Margaret Martonosi
In Proceedings of the 2012 IEEE International Symposium on Performance Analysis
of Systems and Software (ISPASS 2012)
Characterizing and Improving the Use of Demand-Fetched Caches in GPUs
Wenhao Jia, Kelly A. Shaw, Margaret Martonosi
In Proceedings of the 26th International Conference on Supercomputing (ICS 2012)