EECS500 Spring 2015 Department Colloquium

Fengbo Ren
VLSI Design to The Rescue: Building Future Application Drivers with Energy Efficiency
Arizona State University
White 411
March 3, 2015

Technology scaling has changed over the last decade. It is no longer providing energy efficiency gain as in the past due to the end of VDD scaling and increasing leakage currents. Applications have also changed. Mobile devices and server farms in data center are becoming the major application drivers. Yet, computing energy becomes an increasingly severe problem for both platforms due to the limitations of battery capacity and heat removal capability. While we cannot count on device scaling any more, what can we do in building the future application drivers?

In this talk, I will explain how smart customization and memory innovations in VLSI design can come to rescue. I will first present a scalable VLSI architecture design that can efficiently maps the sparse approximation algorithms to demonstrate how programmable hardware units can be designed to accelerate complex algorithms (domain-specific) with orders of magnitude better energy efficiency than general purpose processors. Then, I will talk about our research progress on developing the spin-torque transfer random access memory (STT-RAM) to project how memory innovations can further improve the energy efficiency of memory bounded systems. In the end, I will share my visions on how these technologies can be leveraged to revolutionize the underlying hardware systems for future applications.


Fengbo Ren joined the School of CIDSE at Arizona State University as Assistant Professor in January 2015. He received the B.S. degree in Electrical Engineering from Zhejiang University in 2008, and the M.S. and Ph.D. degree in Electrical Engineering from the University of California, Los Angeles (UCLA) in 2010 and 2014, respectively. He is interested in understanding the interplay between CMOS circuits and nano-devices, as well as the interplay between VLSI architectures and complex algorithms. Specifically, his research has been focused on designing energy-efficient VLSI systems, accelerating sparse signal processing, and developing emerging memory technology.  He is the recipient of 2012-2013 Broadcom Fellowship.