期刊文献+

基于高性能计算的开源云平台性能评估 被引量:19

Performance evaluation on open source cloud platform for high performance computing
在线阅读 下载PDF
导出
摘要 云计算是一种提供各种IT服务的互联网资源利用的新模式,已经广泛地应用在包括高性能计算的各种领域。然而,虚拟化带来了一些性能开销;同时,不同的云平台实施虚拟化技术的不同,使得在这些云平台上应用高性能计算服务的性能也千差万别。通过HPC Challenge(HPCC)Benchmark和NAS Parallel Benchmark(NPB)分别对CPU、内存、网络、扩展性和高性能计算真实负载进行评估,比较并分析了诸如Nimbus、OpenNebula和OpenStack实施高性能计算的性能,实验显示OpenStack对计算密集型的高性能应用负载表现出较好的性能,因此,OpenStack是实施高性能计算的开源云平台的一个好的选择。 Cloud computing is a new model of the Internet resource utilization to provide a variety of IT services. It has been widely used in various fields, including High Performance Computing ( HPC). However, its virtualization has caused some performance overhead. Meanwhile, there are some differences in virtualization technology of different cloud platforms, so the performance in implementing HPC is different among them. The performance and real workload evaluation of HPC of open source clouds platforms, including Nimbus, OpenNebula and OpenStack, were compared and analyzed by HPC Challenge (HPCC) benchmark suite and NAS Parallel Benchmark (NPB) from CPU, memory, communication, scalability and HPC, respectively. The experimental results show that OpenStack has better performance for computation-intensive high performance applications, thus it is a good selection for implementing HPC applications in open source cloud platform.
出处 《计算机应用》 CSCD 北大核心 2013年第12期3580-3585,共6页 journal of Computer Applications
基金 国家自然科学基金资助项目(61170222) 云南大学第五届研究生科研课题一般资助项目(ynuy201258)
关键词 高性能计算 云计算 基准测试 HPCC NAS并行基准测试程序 开源 High Performance Computing (HPC) cloud computing benchmark High Performance ComputingChallenge (HPCC) NAS Parallel Benchmark (NPB) open source
  • 相关文献

参考文献21

  • 1VOUK M A. Cloud computing - issues, research and implementa?tions[C] / / ITI 2008: Proceedings of the 30th International Confer?ence on Information Technology Interfaces. Piscataway: IEEE, 2008: 31 -40.
  • 2YOUNGE AJ, HENSCHEL R, BROWNJ T, et al. Analysis of vir?tualization technologies for high performance computing environments[C] / / CLOUD '11: Proceedings of the 2011 IEEE 4th International Conference on Cloud Computing. Washington, DC: IEEE Computer Society, 2011: 9 - 16.
  • 3DIAZJ, von LASZEWSKI G, WANG F, et al. Towards cloud de?ployments using FutureGrid[R]. Bloomington: Indiana University, 2011.
  • 4KEAHEY K, FOSTER I, FREEMAN T, et al. Virtual workspaces in the grid[C] / / Proceedings of the 11 th International Euro-Par Conference, LNCS 3648. Berlin: Springer-Verlag, 2005: 421 - 431.
  • 5ZHANG X, KEAHEY K, FOSTER I, et al. Virtual cluster work?spaces for grid applications, ANUMCS-PI246- 0405[RlOL].[2013 -03 -16]. http://www. nimbusproject. orgifiles/VWClus?ter_TR_ANL_MCS-PI246-D405. pdf.
  • 6ADABALA S, CHADHA V, CHAWLA P, et al. From virtualized resources to virtual computing grids: the In- VIGO system[J]. Fu?ture Generation Computer Systems, 2005, 21(6): 896 -909.
  • 7YOUSEFF L, WOLSKI R, GORDA B, et al. Evaluating the per- formance impact of Xen on MPI and process execution for HPC sys?tems[C] / / VTDC 2006: Proceedings of the 2nd International Workshop on Virtualization Technology in Distributed Computing. Washington, DC: IEEE Computer Society, 2006: 1.
  • 8MERGEN M F, UHLIG V, KRIEGER 0, et al. Virtualization for high-performance computing[J]. ACM SIGOPS Operating System Review, 2006, 40( 2): 8 -11.
  • 9EVANGELINOS C, HILL C N. Cloud computing for parallel scien?tific HPC applications: feasibility of running coupled atmosphere-o?cean climate models on Amazon's EC2[C]I/ CCA 2008: The 1st Workshop on Cloud Computing and its Applications. Chicago:[s. n.], 2008: 1 -6.
  • 10GAVRILOVSKA A, KUMAR S, RAJ K, et al. High-performance hypervisor architectures: virtualization in HPC systems[C/OL] / / Proceedings of the 1 st Workshop on System-level Virtualization for High Performance Computing[2013 - 05 - 06]. http://www. csm. om!. gov/srtlhpcvirt07/papersipaperl0. pdf.

同被引文献157

引证文献19

二级引证文献49

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部