Hadi Esmaeilzadeh

PhD Student

Computer Science and Engineering

University of Washington

PO Box 352350

Seattle, WA 98195

Email: hadianeh [AT] cs [DOT] washington [DOT] edu

Curriculum Vitae

Research Interests:

Computer Architecture, Approximate Computing, Mixed-signal Analog Architectures, Machine Learning, Compilers, ...

I am currently working with my advisors Doug Burger and Luis Ceze in the Sampa group on a new class of accelerators for general-purpose approximate programs.

Recent Publications:

Architecture Support for Disciplined Approximate Programming, Hadi Esmaeilzadeh, Adrian Sampson, Luis Ceze, Doug Burger, International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '12), London, UK, March 2012.
Dark Silicon and the End of Multicore Scaling, Hadi Esmaeilzadeh, Emily Blem, Renee St. Amant, Karthikeyan Sankaralingam, and Doug Burger, The 38th International Symposium on Computer Architecture (ISCA '12), San Jose, CA, June 2011.
Looking Back on the Language and Hardware Revolution: Measured Power, Performance, and Scaling, Hadi Esmaeilzadeh, Ting Cao, Yang Xi, Stephen M. Blackburn, and Kathryn S. McKinley, International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '12), Newport Beach, CA, March 2011. (Measurement Data)

News:

Our Power-Performance Meausrment paper presented in March 2011 at the International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '11) is selected for CACM Research Highlights.
Our Dark Silicon paper presented in June 2011 at the International Symposium on Computer Architecture (ISCA '11) is profiled in the New York Times:
``Even today, the most advanced microprocessor chips have so many transistors that it is impractical to supply power to all of them at the same time. So some of the transistors are left unpowered ... The phenomenon is known as dark silicon. As early as next year, these advanced chips will need 21 percent of their transistors to go dark at any one time ... And in just three more chip generations - a little more than a half-decade - ... as many as half of them will have to be turned off to avoid overheating."
Our work on energy-aware discplined approximate computing has been covered in ACM TechNews, GeekWire, Engadget.