Minimizing Thermal Variation in Heterogeneous HPC Systems with FPGA Nodes
| Autor: | Gokhan Memik, Xiaoyang Wang, Yingyi Luo, Kazutomo Yoshii, Pete Beckman, Seda Ogrenci-Memik |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Process variation
020203 distributed computing Task (computing) Computer science Distributed computing Thermal Hardware_INTEGRATEDCIRCUITS 0202 electrical engineering electronic engineering information engineering 02 engineering and technology Thermal management of electronic devices and systems Field-programmable gate array Thermal variation 020202 computer hardware & architecture |
| Zdroj: | ICCD |
| DOI: | 10.1109/iccd.2018.00086 |
| Popis: | The presence of FPGAs in data centers has been growing due to their superior performance as accelerators. Thermal management, particularly battling the cooling cost in these high performance systems, is a primary concern. Introduction of new heterogeneous components only adds further complexities to thermal modeling and management. The thermal behavior of multi-FPGA systems deployed within large compute clusters is little explored. In this paper, we first show that the thermal behaviors of different FPGAs of the same generation can vary due to their physical locations in a rack and process variation, even though they are running the same tasks. We present a machine learning based model to capture the thermal behavior of a multi-node FPGA cluster. We then propose to mitigate thermal variation and hotspots across the cluster by proactive task placement guided by our thermal model. Our experiments show that through proper placement of tasks on the multi-FPGA system, we can reduce the peak temperature by up to 11.50°C with no impact on performance. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|