Title :
Design and Analysis of a 32-bit Embedded High-Performance Cluster Optimized for Energy and Performance
Author :
Cloutier, Michael F. ; Paradis, Chad ; Weaver, Vincent M.
Author_Institution :
Electr. & Comput. Eng., Univ. of Maine, Orono, ME, USA
Abstract :
A growing number of supercomputers are being built using processors with low-power embedded ancestry, rather than traditional high-performance cores. In order to evaluate this approach we investigate the energy and performance tradeoffs found with ten different 32-bit ARM development boards while running the HPL Linpack and STREAM benchmarks.Based on these results (and other practical concerns) we chose the Raspberry Pi as a basis for a power-aware embedded cluster computing testbed. Each node of the cluster is instrumented with power measurement circuitry so that detailed cluster-wide power measurements can be obtained, enabling power / performance co-design experiments.While our cluster lags recent x86 machines in performance, the power, visualization, and thermal features make it an excellent low-cost platform for education and experimentation.
Keywords :
benchmark testing; embedded systems; microcontrollers; parallel processing; power aware computing; ARM development boards; HPL Linpack benchmark; Raspberry Pi; STREAM benchmark; cluster-wide power measurements; embedded high-performance cluster analysis; embedded high-performance cluster design; energy optimization; instrumented cluster node; low-power embedded system; performance feature; performance optimization; power feature; power measurement circuitry; power-aware embedded cluster computing testbed; power-performance co-design experiments; supercomputers; thermal feature; visualization feature; Benchmark testing; Power measurement; Program processors; Random access memory; Servers; Supercomputers; Universal Serial Bus;
Conference_Titel :
Hardware-Software Co-Design for High Performance Computing (Co-HPC), 2014
Conference_Location :
New Orleans, LA
DOI :
10.1109/Co-HPC.2014.7
