An approximate performance model of a Unitree mass storage system

Mass storage systems are finding greater use in scientific computing research environments for retrieving and archiving the large volumes of data generated and manipulated by scientific computation. The paper presents a queuing network model that can be used to carry out capacity planning studies of the Unitree mass storage system. Measurements taken on an existing system and a detailed workload characterization provided the workload intensity and resource demand parameters for the various types of read and write requests. The performance model developed here is based on approximations to multi-class mean value analysis of queuing networks. The approximations were validated through the use of discrete event simulation. The resulting baseline model was used to predict the performance of the system as the workload intensity increases