TY - JOUR
T1 - On Memory System Design for Stochastic Computing
AU - Karen Khatamifard, S.
AU - Hassan Najafi, M.
AU - Ghoreyshi, Ali
AU - Karpuzcu, Ulya
AU - Lilja, David J
N1 - Publisher Copyright:
© 2002-2011 IEEE.
PY - 2018/7/1
Y1 - 2018/7/1
N2 - Growing uncertainty in design parameters (and therefore, in design functionality) renders stochastic computing particularly promising, which represents and processes data as quantized probabilities. However, due to the difference in data representation, integrating conventional memory (designed and optimized for non-stochastic computing) in stochastic computing systems inevitably incurs a significant data conversion overhead. Barely any stochastic computing proposal to-date covers the memory impact. In this paper, as the first study of its kind to the best of our knowledge, we rethink the memory system design for stochastic computing. The result is a seamless stochastic system, StochMem, which features analog memory to trade the energy and area overhead of data conversion for computation accuracy. In this manner StochMem can reduce the energy (area) overhead by up-to 52.8% (93.7%) at the cost of at most 0.7% loss in computation accuracy.
AB - Growing uncertainty in design parameters (and therefore, in design functionality) renders stochastic computing particularly promising, which represents and processes data as quantized probabilities. However, due to the difference in data representation, integrating conventional memory (designed and optimized for non-stochastic computing) in stochastic computing systems inevitably incurs a significant data conversion overhead. Barely any stochastic computing proposal to-date covers the memory impact. In this paper, as the first study of its kind to the best of our knowledge, we rethink the memory system design for stochastic computing. The result is a seamless stochastic system, StochMem, which features analog memory to trade the energy and area overhead of data conversion for computation accuracy. In this manner StochMem can reduce the energy (area) overhead by up-to 52.8% (93.7%) at the cost of at most 0.7% loss in computation accuracy.
KW - Stochastic computing
KW - analog memory
KW - energy-efficient design
KW - memory system design
KW - near-sensor processing
UR - http://www.scopus.com/inward/record.url?scp=85041835780&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85041835780&partnerID=8YFLogxK
U2 - 10.1109/LCA.2018.2804926
DO - 10.1109/LCA.2018.2804926
M3 - Article
AN - SCOPUS:85041835780
SN - 1556-6056
VL - 17
SP - 117
EP - 121
JO - IEEE Computer Architecture Letters
JF - IEEE Computer Architecture Letters
IS - 2
ER -