Profile-based Model Extraction of Storage Systems and Model Applications
Juan Wang · Journal of Chinese Computer Systems · 2010
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.