基于高性能计算的海量存储系统新型访问策略研究  

Research on the New Access Policy of Mass Storage System Based on HPC

作　　者：顾锦旗[1] 廖伟[1] 

机构地区：[1]江南计算技术研究所,无锡214000

出　　处：《高性能计算技术》2012年第4期1-7,共7页

摘　　要：为了解决海量信息处理中实时访问中的“I／O墙”问题，提高海量信息分布式存储系统的性能，提出了一种基于HPC的存储部件新型访问策略。首先，分析传统访问模型存在的问题；其次，研究存储部件直通路模式的工作机理，将存储系统分解为多层次、分布式的模型，根据不同的层次和映射策略实现存储空间物理地址、缓存地址、存储系统逻辑空间地址的连续映射；第三，分析直通路访问模式下的存储路径时间开销；第四，在模拟环境下进行存储部件访问的性能测试，在实际采用该策略的应用系统中进行验证。通过验证测试结果表明，该方法能有效地提高了存储系统的性能，能够不断满足海量信息处理实时需要。To solve the ＂I/O wall＂ problem in the case of real-time accessing about mass information processing, to improve performance of distributed mass storage systems, an access policy based on storage unit pass-through is proposed, and analyses the problem of traditional access models. Then study the mechanism of pass-through pattern, and build the model which divide the storage system into multi-level, distributed storage system. Next, the continuous mapping of physical address, cache address of storage space and logical space address of storage system is realized depended on the different level and mapping strategies. Analyses the time consuming of pass-through storage path in pass-through pattern. Last tests the performance of the storage unit in the simulated environment, and validate in the application system which employ the strategies. The validated results show that the method can improve the performance of storage system effectively, and can meet the needs of real-time accessing about massive information processing in the ubiquitous network architecture continuously.

关 键 词：高性能计算 海量存储系统 存储部件直通路 存储层次映射 

分 类 号：TP333[自动化与计算机技术—计算机系统结构]