CSR: Small: Turbo Button: A Semantically-Smart SSD-based RAID System for Internet-Scale Applications
It is timely and opportunistic to rethink the designs of flash-memory SSD-integrated RAID storage systems for Internet-scale applications. Therefore, how to best utilize the high sequential-I/O performance and cost-efficient characteristics of HDD to assist the SSD-based RAID systems becomes an important research issue. This project seeks to develop a Semantically-Smart SSD (S4D) framework to explore and exploit the file system and application semantic information to boost the performance and improve the reliability of flash-memory SSDs.
In particular, this project qualitatively and quantitatively identifies the critical issues for existing flash-memory SSDs, and conveys the file-system block liveness and correlation information to the underlying S4D with the standard or modified block interface. Secondly, S4D exploits the block liveness information to efficiently supplement the log-block pool with free blocks, and reduce the FTL block mapping table size. Finally, based on S4D, Turbo Button, an SSD-HDD-Hybrid RAID storage system, will be designed and constructed, in order to leverage the advantages of HDD judiciously to address the problems of straightforwardly applying RAID algorithms to SSDs. The broader impact of the project lies in its (1) research development that provides significant performance, reliability, and energy-efficiency improvement for existing and future flash-based and flash-integrated storage systems (specially RAID-structured storage systems) widely deployed in data centers that serve the Internet-scale application community; (2) infrastructure development that enhances research and education at UNL and, through accessibility via public domain, the high performance and data-intensive computing community.
