Name
Abstract | ||
|---|---|---|
The success of the exascale supercomputer is largely debated to remain dependent on novel breakthroughs in technology that effectively reduce the power consumption and thermal dissipation requirements. In this work, we consider the integration of co-processors in high-performance computing (HPC) to enable low-power, seamless computation offloading of certain operations. In particular, we explore the so-called Vision Processing Unit (VPU), a highly-parallel vector processor with a power envelope of less than 1W. We evaluate this chip during inference using a pre-trained GoogLeNet convolutional network model and a large image dataset from the ImageNet ILSVRC challenge. Preliminary results indicate that a multi-VPU configuration provides similar performance compared to reference CPU and GPU implementations, while reducing the thermal-design power (TDP) up to 8x in comparison. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/IPDPSW.2018.00098 | 2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) |
Keywords | DocType | Volume |
Vision Processing Unit,High-Performance Computing,Machine Learning | Journal | abs/1810.04150 |
ISSN | ISBN | Citations |
2164-7062 | 978-1-5386-5556-6 | 0 |
PageRank | References | Authors |
0.34 | 13 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sergio Rivas-Gomez | 1 | 7 | 2.89 |
| Antonio J. Peña | 2 | 248 | 25.41 |
| David Moloney | 3 | 12 | 7.69 |
| Erwin Laure | 4 | 369 | 44.71 |
| Stefano Markidis | 5 | 207 | 28.78 |