Paper Info
Title
Automatic runtime frequency-scaling system for energy savings in parallel applications
Abstract
Although high-performance computing has always been about efficient application execution, both energy and power consumption have become critical concerns owing to their effect on operating costs and failure rates of large-scale computing platforms. Modern processors provide techniques, such as dynamic voltage and frequency scaling (DVFS) and CPU clock modulation (called throttling), to improve energy efficiency on-the-fly. Without careful application, however, DVFS and throttling may cause a significant performance loss due to system overhead. This paper proposes a novel runtime system that maximizes energy saving by selecting appropriate values for DVFS and throttling in parallel applications. Specifically, the system automatically predicts communication phases in parallel applications and applies frequency scaling considering both the CPU offload, provided by the network-interface card, and the architectural stalls during computation. Experiments, performed on NAS parallel benchmarks as well as on real-world applications in molecular dynamics and linear system solution, demonstrate that the proposed runtime system obtaining energy savings of as much as 14 % with a low performance loss of about 2 %.
Year
DOI
Venue
2014
10.1007/s11227-013-1062-0
The Journal of Supercomputing
Keywords
Field
DocType
MPI,Dynamic voltage and frequency scaling,Throttling,Infiniband,Parallel computing
Central processing unit,Linear system,InfiniBand,Computer science,Efficient energy use,Parallel computing,Frequency scaling,Bandwidth throttling,Clock rate,Runtime system,Distributed computing
Journal
Volume
Issue
ISSN
68
2
0920-8542
Citations 
PageRank 
References 
9
0.66
24
Authors
3
Name
Order
Citations
PageRank
Vaibhav Sundriyal1494.99
Masha Sosonkina227245.62
Zhao Zhang3797.53