R&D: Approach for RAID Scaling Based on STAR-Code

IEEE Xplore has published, in 2019 IEEE 2nd International Conference on Computer and Communication Engineering Technology (CCET) proceedings, an article written by Xuejun Zhong, Zhu Yuan, Yu Hu, and Ping Xie, Computer College of Qinghai Normal University, Xining, China.

Abstract: “Today, the information data of the world presents explosive growth and the existing storage systems can no longer adapt the requirements of storing massive data. Disk array is composed of a large number of independent disks, which are combined into a huge disk group and provide more I/O parallelism. Therefore, we often adopt the way, adding new disks to disk array for resolving the insufficient storage capacity, which is also called as RAID scaling. Since RAID was proposed, different levels of RAID storage systems have been put forward for different data storage requirements. Compared with other RAID storage systems, RAID storage system based on STAR-Code has better data fault tolerance and higher data recovery ability, which can tolerate three disks fault simultaneously. In this paper, we put forward RS3 approach based on STAR-Code for RAID scaling. Compared with Round-Robin and Semi-RR approach, this approach reduces the data migration by 64.3%～89.5% and decreases the migration time by 66.1%~90.8%.“