Real-time garbage collection for flash-memory storage systems of real-time embedded systems
ACM Transactions on Embedded Computing Systems (TECS)
CFLRU: a replacement algorithm for flash memory
CASES '06 Proceedings of the 2006 international conference on Compilers, architecture and synthesis for embedded systems
A flash-memory based file system
TCON'95 Proceedings of the USENIX 1995 Technical Conference Proceedings
Hybrid solid-state disks: combining heterogeneous NAND flash in large SSDs
Proceedings of the 2008 Asia and South Pacific Design Automation Conference
BPLRU: a buffer management scheme for improving random writes in flash storage
FAST'08 Proceedings of the 6th USENIX Conference on File and Storage Technologies
Proceedings of the eleventh international joint conference on Measurement and modeling of computer systems
CFDC: a flash-aware replacement policy for database buffer management
Proceedings of the Fifth International Workshop on Data Management on New Hardware
System software for flash memory: a survey
EUC'06 Proceedings of the 2006 international conference on Embedded and Ubiquitous Computing
A space-efficient flash translation layer for CompactFlash systems
IEEE Transactions on Consumer Electronics
FAB: flash-aware buffer management policy for portable media players
IEEE Transactions on Consumer Electronics
Hi-index | 0.00 |
This research is to design a new buffer cache for flash memory storage systems. Specifically, flash memory based storage systems are compact and fast, compared with the traditional hard disks. Eventually, flash memory based solid state disks (SSDs) became another attractive storage unit, to be applied to general computing systems. However, flash memory based SSDs show critical drawbacks that stem from the unique characteristics of the electronic storage units, incurring huge overhead. Especially these drawbacks cause serious performance problem when the access pattern of overwrite operations is irregular. This is mainly because the irregular access pattern increases the frequency of overwrite operations and merge operations in the flash memory, which may cause serious overhead. In order to reduce such overhead, this paper proposes a new buffer cache, called the ML-ELFRU (Multi-Level Elaborate Least Frequently/Recently Used Buffer), for a flash translation layer (FTL) in flash memory storages or SSDs. Additionally, experiments are carried out to show that the ML-ELFRU can reduce the number of replacement operations in the buffer cache and the number of merge operations in the electronic storage unit compared with other existing buffer cache policies.