TY - GEN
T1 - The Future of High-Performance Computing
AU - Zahran, Mohamed
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - We are witnessing several factors in computing that offers as much opportunities as challenges. We are witnessing the end of Moore's law, almost two decades after the death of Dennard's scaling. Exascale computing is within reach but the last few steps are the most difficult. For instance, what is the best programming model in the exascale era? How can we make progress toward higher performance without Moore's law? In this article, we discuss the future of high-performance computing in the post-Moore era from hardware and software perspectives. This article is a glimpse to the near future and raises many research questions.We are witnessing several factors in computing that offers as much opportunities as challenges. We are witnessing the end of Moore's law, almost two decades after the death of Dennard's scaling. Exascale computing is within reach but the last few steps are the most difficult. For instance, what is the best programming model in the exascale era? How can we make progress toward higher performance without Moore's law? In this article, we discuss the future of high-performance computing in the post-Moore era from hardware and software perspectives. This article is a glimpse to the near future and raises many research questions.
AB - We are witnessing several factors in computing that offers as much opportunities as challenges. We are witnessing the end of Moore's law, almost two decades after the death of Dennard's scaling. Exascale computing is within reach but the last few steps are the most difficult. For instance, what is the best programming model in the exascale era? How can we make progress toward higher performance without Moore's law? In this article, we discuss the future of high-performance computing in the post-Moore era from hardware and software perspectives. This article is a glimpse to the near future and raises many research questions.We are witnessing several factors in computing that offers as much opportunities as challenges. We are witnessing the end of Moore's law, almost two decades after the death of Dennard's scaling. Exascale computing is within reach but the last few steps are the most difficult. For instance, what is the best programming model in the exascale era? How can we make progress toward higher performance without Moore's law? In this article, we discuss the future of high-performance computing in the post-Moore era from hardware and software perspectives. This article is a glimpse to the near future and raises many research questions.
KW - heterogeneous computing
KW - high-performance computing
KW - parallel programming
KW - specialized chips
UR - http://www.scopus.com/inward/record.url?scp=85126536372&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85126536372&partnerID=8YFLogxK
U2 - 10.1109/ICENCO49852.2021.9698918
DO - 10.1109/ICENCO49852.2021.9698918
M3 - Conference contribution
AN - SCOPUS:85126536372
T3 - 17th International Computer Engineering Conference, ICENCO 2021
SP - 129
EP - 134
BT - 17th International Computer Engineering Conference, ICENCO 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 17th International Computer Engineering Conference, ICENCO 2021
Y2 - 29 December 2021 through 30 December 2021
ER -