Optimizing Emerging Storage Primitives with Virtualization for Flash Memory Storage Systems

摘要

NAND flash memory has become the mainstream storage medium for both enterprise high performance computers and embedded systems. However, over the past several decades, the storage primitives that access secondary storage have remained unchanged, forcing NAND flash memory to serve merely as a block device like hard disk drive. Recently, several emerging storage primitives have been presented to explore the potential value of non-volatile memory devices. They provide high-level virtual address to logical address mapping and allow the applications to directly manipulate I/O operations. Although these primitives can significantly boost the access performance by providing virtual to logical address mappings, they still need to handle the logical to physical address translations in current flash-based storage systems and require further support for update operations. This paper presents ESP to optimize Emerging Storage Primitives with virtualization for flash memory storage systems. We propose an optimization strategy, namely virtual duplication, to solve the critical issues in existing emerging storage primitives. The objective is to reduce unnecessary flash memory accesses and improve the address translation for emerging storage primitives. We have evaluated ESP on an embedded development platform. Experimental result show that ESP can improve the write/read performance by 56.71% and reduce over 30% of garbage collection operations.