一种存储系统性能模型的提取方法及模型应用

Profile-based Model Extraction of Storage Systems and Model Applications

存储系统性能模型的获取对于存储系统设计、系统管理以及系统性能评价等相关领域的理论研究和工程实践均具有极为重要的意义.相对于常规的分析建模方法,本文提出了一种基于运行时特征的存储系统性能模型提取方法,该方法依据I/O负载的运行时特征和系统性能,借助多元回归分析理论,在线提取系统的性能模型.实验结果表明文中方法所获取性能模型具有较高的精确度,同时还实验证实了所提取性能模型在存储管理领域性能预测和what-if分析两个重要问题场景下的实用性和有效性.

Modeling storage systems is vital for the design,management,and performance evaluation of storage systems.As opposed to conventional first-principles modeling techniques,this paper,leveraging the multivariable regression theory,proposes a profile-based model extraction approach for storage systems.Experimental results show that the performance models derived from our approach can achieve rather high accuracy.Furthermore,our experiments validate the practicability and effectiveness of the applications of the performance models to the problems of performance prediction and what-if analysis,both of which are the core tasks in the domain of storage systems management.