TY - GEN
T1 - Efficient GPGPU computing with cross-core resource sharing and core reconfiguration
AU - Dhar, Ashutosh
AU - Chen, Deming
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/6/30
Y1 - 2017/6/30
N2 - GPUs are capable of running a variety of applications, however their generic parallel-architecture can lead to inefficient use of resources and reduced power efficiency, due to algorithmic or architectural constraints. In this work, taking inspiration from CGRAs (coarse-grained reconfigurable architectures), we demonstrate resource sharing and re-distribution as a solution that can be leveraged by reconfiguring the GPU on a kernel-by-kernel basis. We explore four different schemes that trade the number of active SMs (streaming multiprocessor) for increased occupancy and local memory resources per SM and demonstrate improved power and energy with limited impact to performance. Our most aggressive scheme, BigSM, is capable of saving energy by up to 54%, and 26% on an average.
AB - GPUs are capable of running a variety of applications, however their generic parallel-architecture can lead to inefficient use of resources and reduced power efficiency, due to algorithmic or architectural constraints. In this work, taking inspiration from CGRAs (coarse-grained reconfigurable architectures), we demonstrate resource sharing and re-distribution as a solution that can be leveraged by reconfiguring the GPU on a kernel-by-kernel basis. We explore four different schemes that trade the number of active SMs (streaming multiprocessor) for increased occupancy and local memory resources per SM and demonstrate improved power and energy with limited impact to performance. Our most aggressive scheme, BigSM, is capable of saving energy by up to 54%, and 26% on an average.
KW - CGRA
KW - GPGPU
KW - Reconfigurable architecture
UR - http://www.scopus.com/inward/record.url?scp=85027721598&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85027721598&partnerID=8YFLogxK
U2 - 10.1109/FCCM.2017.59
DO - 10.1109/FCCM.2017.59
M3 - Conference contribution
AN - SCOPUS:85027721598
T3 - Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017
SP - 48
EP - 55
BT - Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 25th Annual IEEE International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017
Y2 - 30 April 2017 through 2 May 2017
ER -