Abstract | ||
---|---|---|
DARPA's Ubiquitous High-Performance Computing (UHPC) program asked researchers to develop computing systems capable of achieving energy efficiencies of 50 GOPS/Watt, assuming 2018-era fabrication technologies. This paper describes Runnemede, the research architecture developed by the Intel-led UHPC team. Runnemede is being developed through a co-design process that considers the hardware, the runtime/OS, and applications simultaneously. Near-threshold voltage operation, fine-grained power and clock management, and separate execution units for runtime and application code are used to reduce energy consumption. Memory energy is minimized through application-managed on-chip memory and direct physical addressing. A hierarchical on-chip network reduces communication energy, and a codelet-based execution model supports extreme parallelism and fine-grained tasks. We present an initial evaluation of Runnemede that shows the design process for our on-chip network, demonstrates 2–4x improvements in memory energy from explicit control of on-chip memory, and illustrates the impact of hardware-software co-design on the energy consumption of a synthetic aperture radar algorithm on our architecture. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/HPCA.2013.6522319 | High Performance Computer Architecture |
Keywords | Field | DocType |
memory energy,ubiquitous high-performance computing,hierarchical on-chip network,energy consumption,communication energy,co-design process,on-chip memory,intel-led uhpc team,on-chip network,application-managed on-chip memory,energy efficiency,computer architecture,ubiquitous computing | Supercomputer,Computer science,Parallel computing,Computing with Memory,Real-time computing,Conventional memory,Engineering design process,Execution model,Ubiquitous computing,Memory address,Energy consumption,Embedded system | Conference |
ISSN | ISBN | Citations |
1530-0897 | 978-1-4673-5585-8 | 38 |
PageRank | References | Authors |
1.17 | 19 | 19 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nicholas P. Carter | 1 | 349 | 33.84 |
Aditya Agrawal | 2 | 546 | 34.80 |
Shekhar Borkar | 3 | 4236 | 494.95 |
Romain Cledat | 4 | 67 | 4.61 |
Howard David | 5 | 399 | 12.83 |
Dave Dunning | 6 | 189 | 15.79 |
Joshua Bruce Fryman | 7 | 80 | 7.19 |
Ivan Ganev | 8 | 43 | 1.61 |
Roger A. Golliver | 9 | 79 | 10.73 |
Rob Knauerhase | 10 | 173 | 7.38 |
Richard Lethin | 11 | 118 | 17.17 |
Benoît Meister | 12 | 138 | 12.84 |
Asit K. Mishra | 13 | 1216 | 46.21 |
Wilfred R. Pinfold | 14 | 38 | 1.17 |
Justin Teller | 15 | 49 | 3.15 |
Josep Torrellas | 16 | 3838 | 262.89 |
Nicolas Vasilache | 17 | 354 | 19.45 |
Ganesh Venkatesh | 18 | 274 | 17.97 |
Jianping Xu | 19 | 53 | 1.76 |