Building the next ubiquitous computing platform drawing on condensed matter physics

Monday, February 22, 2021

Computing is at a momentous point today. AI, big data and decentralized work are driving a surging demand for computing power. At the same time, an ending of Moore’s Law and Dennard’s scaling are making it increasingly difficult (and expensive) to improve processor performance. The energy consumed by computing is therefore growing exponentially, doubling every 3 years, and could go on to consume as much as 25% of the world’s primary energy production in the next few decades if action is not taken to significantly improve computing energy efficiency [1]. The past decades improvements in computing, drawing from improvements in systems engineering, are not poised to deliver the demands of next decade.

A Unified Computing framework  translated to physical, complexity and thermodynamic axes :  In this talk, I will first outline a unified framework for reducing energy consumption whilst increasing compute performing, combining energy/dimension scaling (Moore’s law) with computer error rates (Shannon computing) and complexity (architectures), drawing from our Nature Physics perspective on Beyond CMOS computing with spin and polarization. Furthermore, we identify the key limiting factors for near-term computing: the utilization of logic inside a computing system (activity factor); the Turing wall to access stored data (memory bandwidth); and thermal extraction.

Next-generation computing with quantum materials : Building on this framework, I will describe a quantum and memory-materials-centric approach to enable beyond-CMOS computing, outline a number of pathways [3,4] for computing devices that utilize quantum materials.

Artificial General Intelligence and condensed matter physics: I will end by describing the potential ways to build the hardware for artificial general intelligence, overcoming the key interconnect, memory and compute bottlenecks, and the role that condensed matter physics needs to play.

The aim of this talk is to give an overview of the potential for using condensed matter to scale computing beyond-CMOS. I will generalize the search for the next ubiquitous computing device with a comprehensive list of quantum materials classes, and will highlight the top-10 outstanding problems that a condensed matter physicist can solve to address this generational goal [4].      

[1] https://www.src.org/about/decadal-plan/     
[2] Manipatruni, Nature Physics 14, no. 4 (2018): 338.       
[3] Manipatruni, Nature, 2019                                 
[4] Manipatruni, Nature, 2021 (invited)

Speaker Bio: Dr. Sasikanth Manipatruni is the Chief Technology Officer at Kepler Computing. Prior to this he is the founding research director of Intel-FEINMAN center (Functional Electronics Integration and Manufacturing), to build the next room temperature transistor with quantum materials. He received PhD from Cornell in silicon photonics and quantum optics where he demonstrated ultra-fast silicon electro-optic switches, opto-mechanical non-reciprocity and synchronization of opto- mechanical systems. At Intel, he developed materials & devices for beyond CMOS memory/logic and built 1st industrially adopted spintronic/quantum SPICE tool. He was awarded the US-National Academy of Engineering recognition for young engineers 2019, IEEE/ACM under 40 innovator award at DAC'17, Mahboob Khan outstanding liaison award '16, CSPIN outstanding industry liaison award '16, and serves on several industry panels for US-wide research selections.

https://berkeley.zoom.us/s/98262274858
Meeting ID: 982 6227 4858
Passcode: 279538
SIP: 98262274858@zoomcrc.com
Passcode: 279538

Location: 
virtual (zoom)
Speaker: 
Sasikanth Manipatruni
Affiliation: 
Kepler Computing: Condensed matter for ultralow power computing