A hybrid HPC/cloud distributed infrastructure: Coupling EC2 cloud resources with HPC clusters to run large tightly coupled multiscale applications |
| |
Affiliation: | 1. College of Applied Sciences & Beijing Engineering Research Center of Precision Measurement Technology and Instrument, Beijing University of Technology, Beijing 100124, China;2. The Pilot College of Beijing University of Technology, Beijing 101101, China;1. Department of Radiology, The First Affiliated Hospital of Xinxiang Medical University, Xinxiang, China;2. Department of Medical Imaging, Xi''an No.3 Hospital, Xi''an, China;3. School of Life Science and Technology, Xi''an Jiaotong University, Xi''an, China;1. RADICAL Laboratory, Electric and Computer Engineering, Rutgers University, New Brunswick, NJ, USA;2. Brookhaven National Laboratory, Upton, NY 11793, United States |
| |
Abstract: | In this paper, we report on the experimental results of running a large, tightly coupled, distributed multiscale computation over a hybrid High Performance Computing (HPC) infrastructures. We connected EC2 based cloud clusters located in USA to university clusters located in Switzerland. We ran a concurrent multiscale MPI based application on this infrastructure and measured the overhead induced by extending our HPC clusters with EC2 resources. Our results indicate that accommodating some parts of the multiscale computation on cloud resources can lead to low performance without a proper adjustment of CPUs power and workload. However, by enforcing a load-balancing strategy one can benefit from the extra Cloud resources. |
| |
Keywords: | Distributed computation HPC infrastructure Cloud resources Multiscale applications Performance measures MAPPER project |
本文献已被 ScienceDirect 等数据库收录! |
|