Hadi Esmaeilzadeh

PhD Candidate

University of Washington

Computer Science and Engineering

AC101 Paul G. Allen Center, Box 352350

Seattle, WA 98195-2350

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

Curriculum Vitae Research Statement Teaching Statement

I will be joining the School of Computer Science at the Georgia Institute of Technology as an assistant professor in the fall of 2013. I am interested in developing new technologies and cross-stack solutions to improve the performance and energy efficiency of computer systems for emerging applications. For more details, read my research statement. I am eagerly looking for students that ambitiously want to make a difference!

Recent Publications:

• Neural Acceleration for General-Purpose Approximate Programs, Hadi Esmaeilzadeh, Adrian Sampson, Luis Ceze, and Doug Burger, IEEE Micro Top Picks from the 2012 Computer Architecture Conferences, vol. 33, no. 3, May/June 2013. (original at MICRO '12)
• Power Challenges May End the Multicore Era , Hadi Esmaeilzadeh, Emily Blem, Renee St. Amant, Karthikeyan Sankaralingam, and Doug Burger, Communication of ACM (CACM), Research Highlights, 2013. (original at ISCA '11). Pradip Bose's perspective on our paper.
• How to Implement Effective Prediction and Forwarding for Fusable Dynamic Multicore Architectures, Behnam Robatmil, Dong Li, Hadi Esmaeilzadeh, Sibi Govindan, Aaron Smith, Andrew Putnam, Doug Burger, Stephen W. Keckler, International Symposium on High Performance Computer Architecture (HPCA), Shenzhen, China, February 2013.
• Neural Acceleration for General-Purpose Approximate Programs, Hadi Esmaeilzadeh, Adrian Sampson, Luis Ceze, Doug Burger, International Symposium on Microarchitecture (MICRO), Vancouver, Canada, December 2012. (selected for IEEE Micro Top Picks)
• Multicore Model from Abstract Single Core Inputs, Emily Blem, Hadi Esmaeilzadeh, Renee St. Amant, Karthikeyan Sankaralingam, and Doug Burger, Computer Architecture Letters (CAL), vol. PP, no. 99, August 2012.
• Power Limitations and Dark Silicon Challenge the Future of Multicore, Hadi Esmaeilzadeh, Emily Blem, Renee St. Amant, Karthikeyan Sankaralingam, and Doug Burger, ACM Transaction on Computer Systems (TOCS), vol. 30, no. 3, August 2012. (invited, original at ISCA '11)
• Looking Back and Looking Forward: Power, Performance, and Upheaval , Hadi Esmaeilzadeh, Ting Cao, Yang Xi, Stephen M. Blackburn, and Kathryn S. McKinley, Communications of the ACM (CACM), Research Highlights vol. 55, no. 7, July 2012. (original at ASPLOS '11). David Patterson's perspective on our paper.
• Addressing Dark Silicon Challenges with Disciplined Approximate Computing, Hadi Esmaeilzadeh, Adrian Sampson, Michael Ringenburg, Luis Ceze, Dan Grossman, Doug Burger, The Forth Workshop on Energy-Efficient Design (DaSi) held in conjunction with ISCA, Portland, OR, June 2012.
• Toward Neural Acceleration for General-Purpose Approximate Programs, Hadi Esmaeilzadeh, Adrian Sampson, Luis Ceze, Doug Burger, The Forth Workshop on Energy-Efficient Design (WEED) held in conjunction with ISCA, Portland, OR, June 2012.
• Dark Silicon and the End of Multicore Scaling, Hadi Esmaeilzadeh, Emily Blem, Renee St. Amant, Karthikeyan Sankaralingam, and Doug Burger, IEEE Micro Top Picks from the 2011 Computer Architecture Conferences, vol. 32, no. 3, May/June 2012. (original at ISCA '11)
• What Is Happening to Power, Performance, and Software?, Hadi Esmaeilzadeh, Ting Cao, Yang Xi, Stephen M. Blackburn, and Kathryn S. McKinley, IEEE Micro Top Picks from the 2011 Computer Architecture Conferences, vol. 32, no. 3, May/June 2012. (original at ASPLOS '11)
• 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), 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), San Jose, CA, June 2011. (selected for Communications of ACM Research Highlights and IEEE Micro Top Picks)
• 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), Newport Beach, CA, March 2011. (Measurement Data) (selected for Communications of ACM Research Highlights and IEEE Micro Top Picks)

For a full list, please check out my Google Scholar profile!

Music:

Guest Performer. Concert in Dashti. Bereket UT-Austin Middle Eastern Ensemble, Butler School of Music, The University of Texas at Austin, Bates Concert Hall, April 2010

• Tonbak (Persian goblet drum)
• Daf (chained Persian frame drum)

News:

• The NPU paper presented in December 2012 at the International Symposium on Microarchitecture (MICRO '12) is selected for IEEE Micro top picks from the 2012 computer architecture conferences.
• The Dark Silicon paper presented in June 2011 at the International Symposium on Computer Architecture (ISCA '11) is selected for CACM Research Highlights.

  ``This paper is not just a doomsday predictor. It raises our awareness of the problem through scientific quantification; but it should also serve as a springboard for innovative research, especially for computer architects. However, the architect cannot hope to invent in a vacuum; the needed innovations will surely come, but only by adopting a holistic, cross-layer view of the full system—from devices, through circuits, microarchitecture, system architecture, and the software stack." Pardip Bose

• Both the ISCA Dark Silicon and the ASPLOS Power-Performance Measurement papers are selected for the IEEE Micro top picks from the 2011 computer architecture conferences.

  ``The last theme presents interesting perspective on computer architecture trends by looking back and looking forward. In ``What Is Happening to Power, Performance, and Software?" Hadi Esmaeilzadeh et al. analyze measured power and performance of a large variety of processors and workloads to better understand the combined effects and interactions of the rise of parallel processors and managed programming languages in the past decade. In ``Dark Silicon and the End of Multicore Scaling," Hadi Esmaeilzadeh et al. highlight a crucial impending problem where future multicore chips will be power-limited to the point that an increasing fraction of cores will have to be kept powered off ("dark") at every new technology generation." Co-Chairs of the selection committee

• John Hennessey and David Patterson used the power-performance measurement data from our ASPLOS '11 paper in the most recent edition of their canonical computer architecture textbook, Computer Architecture: A Quantitative Approach.

  ``With the availability of recent extensive and insightful measurements done by Esmaeilzadeh et al. [2011], we can look at the performance and energy benefits of using SMT in single i7 core using a set of multithreaded applications.'' Chapter three

• The Dark Silicon paper presented in June 2011 at the International Symposium on Computer Architecture (ISCA '11) is invited for publication in ACM Transaction on Computer Systems (TOCS).
• The Power-Performance Measurement 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.

  ``This treasure chest of data-recorded in large tables in the ACM Digital Library in addition to this paper-allows the authors (and the rest of us) to ask and answer many questions based on real hardware. This opportunity is a refreshing change from research results based on simulation, which has dominated the literature for the last decade." David Patterson

• The 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 disciplined approximate computing has been covered in ACM TechNews, GeekWire, Engadget.