Mimicking Natural Ways of Computing N. Shukla 1 , A. Parihar 2 , A. - - PowerPoint PPT Presentation

mimicking natural ways of computing
SMART_READER_LITE
LIVE PREVIEW

Mimicking Natural Ways of Computing N. Shukla 1 , A. Parihar 2 , A. - - PowerPoint PPT Presentation

Mimicking Natural Ways of Computing N. Shukla 1 , A. Parihar 2 , A. Raychowdhury 2 , and Suman Datta 1 1 University of Notre Dame, Notre Dame, IN, USA 2 Georgia Tech, Atlanta, GA, USA Dagstuhl Seminar, Feb 2017 1 Natural Ways of Computing


slide-1
SLIDE 1

Mimicking Natural Ways of Computing

  • N. Shukla1, A. Parihar2,
  • A. Raychowdhury2, and Suman Datta1

1University of Notre Dame, Notre Dame, IN, USA 2Georgia Tech, Atlanta, GA, USA

1

Dagstuhl Seminar, Feb 2017

slide-2
SLIDE 2

Natural Ways of Computing

slide-3
SLIDE 3

Natural Ways of Computing

slide-4
SLIDE 4

Prediction

slide-5
SLIDE 5

EXCEL (sponsored by NSF/NRI)

Notre Dame, Georgia Tech, Penn St, U Chicago, UC San Diego

slide-6
SLIDE 6

Computational Paradigms

Many body analog computing with both feedback and feedforward dynamics Potential super-Turing capabilities (debatable) Compatible with CMOS Room temperature

  • peration

Coupled Dynamical Systems

slide-7
SLIDE 7

Vertex Coloring of Graphs

time evolution of the dynamical system created by the coupled oscillator circuit correspond to an approximate vertex color sorting algorithm

Parihar, Shukla et a “Vertex coloring of graphs via phase dynamics of coupled oscillatory networks“ (submitted to Nature Scientific Reports)

slide-8
SLIDE 8

Coupled Oscillatory Systems

Synchronization of Metronomes Spin Torque Nano Oscillators CMOS PLLs

Zhang, Mian, et al. "Synchronization of micromechanical oscillators using light." Physical review letters 109.23 (2012): 233906.

Opto-mechanical Oscillators

8

https://www.youtube.com/watch?v=JWToUATLGzs; (Ikelguchi Lab)

Kaka, Shehzaad, Matthew R. Pufall, William H. Rippard, Thomas J. Silva, Stephen E. Russek, and Jordan A. Katine. "Mutual phase-locking of microwave spin torque nano-oscillators." Nature 437, no. 7057 (2005): 389-392. Shibata, Tadashi, et al. "CMOS supporting circuitries for nano-oscillator-based associative memories.“ CNNA, 2012 13th International Workshop on. IEEE, 2012.

Fixed layer Free layer

slide-9
SLIDE 9

9

Thermally driven Insulator-metal phase transition in VO2

Abrupt change in VO2 resistivity through electron correlation dynamics in ultra-thin VO2 films.

240 270 300 330 10

  • 4

10

  • 3

10

  • 2

10

  • 1

10

Temperature (K)

Resistivity (Ω.cm)

insulator metal

  • M. Huefner, R. Ghosh, E. Freeman, N. Shulka, H. Paik, D. G. Schlom, and S. Datta "Hubbard Gap Modulation in Vanadium Dioxide Nanoscale Tunnel Junctions", Nano

Letters, October 2014.

monoclinic

Insulating VO2 Metallic VO2

rutile

slide-10
SLIDE 10

Abrupt, hysteretic phase transition can be electrically triggered

Electrically induced “abrupt”, “reversible” and “hysteretic” phase transition 20 40 0.0 0.5 1.0 1.5 2.0

Current (mA)

Electric Field (kV/cm)

insulator

metal

10

VDC

slide-11
SLIDE 11

VO2 based Relaxation Oscillators

Negative feedback by series resistor (RS) enable oscillations

V1 VDC RS

V1 t

VDC t

V1

IM

VO2 C

VDC

V1

MI

VO2 C

VDC

RS RS

V1 V1 t t

11

50 100 150 6 8 10 12

V1 (V)

Time (µs)

experiment

slide-12
SLIDE 12

VGS D S V1 VDC

Si substrate

0.0 0.2 0.4 0.6 0.8 20 40 60 80 100

Frequency (kHz) VGS (V)

Hybrid VO2-MOSFET (HVFET) oscillator

50 100 150 5 10 15

Time (µs)

V1 (V)

Voltage controlled VO2 oscillator realized by replacing RS with a MOSFET (HVFET Oscillator)

VDC t

V1 t

12

slide-13
SLIDE 13

VGS,1 VGS,2

Si substrate

VDC VDC

V2 (output) V1 (output)

S D D S

CC

Pairwise Coupled HVFET Oscillators

Capacitively coupled oscillators show frequency synchronization

10 20 30

  • 50
  • 40
  • 30
  • 20
  • 10

Frequency (kHz) Power (dB)

Resonant frequency

V1 V2

CC= 2.2nF

VDC

t

V2

t

V1

t

13

slide-14
SLIDE 14

Equivalent Circuit for coupled HVFET Oscillators

Equivalent circuit to analyze synchronization dynamics of pairwise coupled HVFET oscillators

14

MOSFET

VMOS RVO

g0

gmVgs

C1

+

g0

gmVgs

C2

+

MOSFET

VMOS

VGS,1 VGS,2

CC

2 1

  • RVO2

2

slide-15
SLIDE 15

Locked Locked Unlocked Unlocked

Phase Space Diagram (locking / unlocking)

VO2

VDC

Vgs1

VO2

VDC

Vgs2

CC Gate Input voltage difference decides synchronization

ΔVGS=VGS,2-VGS,1

ΔVGS=-0.2 V ΔVGS=-0.02 V ΔVGS=0.2 V ΔVGS=0.02 V

15

slide-16
SLIDE 16

Experiments vs Simulation

16

slide-17
SLIDE 17

Degree of Locking (Locked Case)

Part of steady state periodic orbit in the XOR=0 region

  • f the phase space depends on ΔVGS = VGS,2 - VGS,1

0.6 0.7 0.8 0.6 0.7 0.8

P

0.6 0.7 0.8 0.6 0.7 0.8 0.6 0.7 0.8 0.6 0.7 0.8

V1 (V) V2 (V)

VGS,1 = 0.3V VGS,2 = 0.3V VGS,1 = 0.3V VGS,2 = 0.32V VGS,1 = 0.3V VGS,2 = 0.35V

XOR=0

17

slide-18
SLIDE 18

Phase Difference measurement

Coupled Oscillator Output Output Threshold

XOR

Time Average

V1 V2

18

System implementation to measure steady state periodic orbit of the HVFET oscillators

slide-19
SLIDE 19

Graph Coloring Experiments

Net phase of any oscillator is the aggregate of the ‘repelling effect’ of all the adjacent oscillators

19

slide-20
SLIDE 20

Connectivity of Graphs

Tradeoff between settling time and # of colors detected (as a function of average connectivity)

20

slide-21
SLIDE 21

Heuristics vs Dynamical System

Dynamical systems can outperform heuristic algorithms in some cases

21 DIMACS implementation challenge

slide-22
SLIDE 22

22

Connection between Phase Dynamics of coupled oscillator system and spectral techniques of solving graph coloring problems provides a new way to think about next generation computing systems Perhaps we can finally realize the true potential of beyond-CMOS devices by matching the “physics” of emerging nanodevices to the “mathematics” of the computing”