TY - GEN
T1 - Exploring hardware overprovisioning in power-constrained, high performance computing
AU - Patki, Tapasya
AU - Lowenthal, David K.
AU - Rountree, Barry
AU - Schulz, Martin
AU - De Supinski, Bronis R.
PY - 2013
Y1 - 2013
N2 - Most recent research in power-aware supercomputing has focused on making individual nodes more efficient and measuring the results in terms of flops per watt. While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level. Traditional approaches in supercomputer design use worst-case power provisioning: the total power allocated to the system is determined by the maximum power draw possible per node. In a world where power is plentiful and nodes are scarce, this solution is optimal. However, as power becomes the limiting factor in supercomputer design, worst-case provisioning becomes a drag on performance. In this paper we demonstrate how a policy of overprovisioning hardware with respect to power combined with intelligent, hardware-enforced power bounds consistently leads to greater performance across a range of standard benchmarks. In particular, leveraging overprovisioning requires that applications use effective configurations; the best configuration depends on application scalability and memory contention. We show that using overprovisioning leads to an average speedup of more than 50% over worst-case provisioning.
AB - Most recent research in power-aware supercomputing has focused on making individual nodes more efficient and measuring the results in terms of flops per watt. While this work is vital in order to reach exascale computing at 20 megawatts, there has been a dearth of work that explores efficiency at the whole system level. Traditional approaches in supercomputer design use worst-case power provisioning: the total power allocated to the system is determined by the maximum power draw possible per node. In a world where power is plentiful and nodes are scarce, this solution is optimal. However, as power becomes the limiting factor in supercomputer design, worst-case provisioning becomes a drag on performance. In this paper we demonstrate how a policy of overprovisioning hardware with respect to power combined with intelligent, hardware-enforced power bounds consistently leads to greater performance across a range of standard benchmarks. In particular, leveraging overprovisioning requires that applications use effective configurations; the best configuration depends on application scalability and memory contention. We show that using overprovisioning leads to an average speedup of more than 50% over worst-case provisioning.
KW - high-performance computing
KW - overprovisioned
KW - power
KW - rapl
UR - http://www.scopus.com/inward/record.url?scp=84879812114&partnerID=8YFLogxK
U2 - 10.1145/2464996.2465009
DO - 10.1145/2464996.2465009
M3 - Conference contribution
AN - SCOPUS:84879812114
SN - 9781450321303
T3 - Proceedings of the International Conference on Supercomputing
SP - 173
EP - 182
BT - ICS 2013 - Proceedings of the 2013 ACM International Conference on Supercomputing
T2 - 27th ACM International Conference on Supercomputing, ICS 2013
Y2 - 10 June 2013 through 14 June 2013
ER -