Improving Hybrid Flash-Disk Storage by Incorporating Sequential Pattern Mining into Fully-Associative Sector Translation
Un-Keun Yoon, Hanjoon Kim · 2009
In this paper, we present a new way of improving hybrid flash-disk storage systems by intelligently prefetching file objects with sequential pattern mining technique. Our ultimate goal is to minimize overall file I/O processing time of hybrid storage systems with the fully associated sector translation (FAST) technique that is known to be the best mapping method in managing flash memory. The FAST assumes that the flash memory space consists of log blocks and data blocks. Since the FAST technique can effectively handle sequential accesses while highly utilizing log block space, it is very significant to further enhance the system performance of the hybrid storage when applying FAST to it. In our work, the hybrid storage uses the flash memory as a cache space. With this memory architecture, we focus on prefetching file objects onto sequential log blocks to be accessed in the near future through sequential pattern mining. Through intensive experiments using UCC data and synthetic data, we show that the proposed prefetching method contributes to improving the performance of the hybrid storage.