Please wait a minute...
FITEE
Front. Inform. Technol. Electron. Eng.  2016, Vol. 17 Issue (11): 1154-1175    DOI: 10.1631/FITEE.1601336
    
Storage wall for exascale supercomputing
Wei Hu, Guang-ming Liu, Qiong Li, Yan-huang Jiang, Gui-lin Cai
College of Computer, National University of Defense Technology, Changsha 410073, China; National Supercomputer Center in Tianjin, Tianjin 300457, China
Download:   PDF(0KB)
Export: BibTeX | EndNote (RIS)      

Abstract  The mismatch between compute performance and I/O performance has long been a stumbling block as supercomputers evolve from petaflops to exaflops. Currently, many parallel applications are I/O intensive, and their overall running times are typically limited by I/O performance. To quantify the I/O performance bottleneck and highlight the significance of achieving scalable performance in peta/exascale supercomputing, in this paper, we introduce for the first time a formal definition of the ‘storage wall’ from the perspective of parallel application scalability. We quantify the effects of the storage bottleneck by providing a storage-bounded speedup, defining the storage wall quantitatively, presenting existence theorems for the storage wall, and classifying the system architectures depending on I/O performance variation. We analyze and extrapolate the existence of the storage wall by experiments on Tianhe-1A and case studies on Jaguar. These results provide insights on how to alleviate the storage wall bottleneck in system design and achieve hardware/software optimizations in peta/exascale supercomputing.

Key words： Storage-bounded speedup      Storage wall      High performance computing      Exascale computing     
Received: 15 June 2016      Published: 07 November 2016
CLC:  TP338.6  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Wei Hu
Guang-ming Liu
Qiong Li
Yan-huang Jiang
Gui-lin Cai
Cite this article:

Wei Hu, Guang-ming Liu, Qiong Li, Yan-huang Jiang, Gui-lin Cai. Storage wall for exascale supercomputing. Front. Inform. Technol. Electron. Eng., 2016, 17(11): 1154-1175.

URL:

http://www.zjujournals.com/xueshu/fitee/10.1631/FITEE.1601336     OR     http://www.zjujournals.com/xueshu/fitee/Y2016/V17/I11/1154


面向E级高性能计算存储墙问题研究

概要：I/O性能与计算性能的不匹配是超级计算机从P级向E级发展的主要障碍之一。当前，I/O密集型应用迅速增加，其运行效率受到了I/O性能的制约。为进一步量化I/O性能瓶颈，分析其对大规模并行应用可扩展性的重要影响，本文在定义和分析存储受限加速比的基础上，第一次从并行应用可扩展性的角度提出\"存储墙\"的定义，量化和讨论了\"存储墙\"的存在性，并基于存储墙特性对现有高性能计算系统结构提出了新的分类方法和可扩展性分析方法。通过基于\"天河1A\"超级计算机的实验和基于\"美洲虎\"超级计算机的案例分析，本文验证了存储墙的存在性并分析了其变化特性。这些实验和分析结果对缓解和消除未来高性能计算机的存储墙瓶颈，实现平衡的软硬件设计具有重要指导意义。

关键词： 存储受限加速比,  存储墙,  高性能计算,  E级计算 
No related articles found!