Designing an Efficient Flash Translation
Flash memory is a linear device consisting of an array of blocks. The flash translation layer (FTL) is a software/firmware layer implemented inside a flash-based storage device to make the linear device act like a block-based device. This translation layer helps existing applications use flash-based storage devices without any modifications. The FTL receives logical read and write commands from the application and converts them to internal flash commands. The FTL also controls the logical-to-physical page mapping inside the storage device, which directly affects the read and write performance and cell wear-out problem. This mapping can be maintained at the page-level, block-level, or a hybrid combination of both block-level and page-level. Additional research is needed to design an efficient FTL that provides high performance with efficient use of space. A related problem is designing a wear-leveling algorithm that considers the limited number of block erasures permitted in each flash block. We have found one of the critical issues for designing efficient FTL and wear-leveling algorithms is how to identify Hot and Cold Data, which is data that will be updated in the near future and data that will not be updated for a long period of time, respectively.
Status
Ongoing
Publications
Date | Publication | |
---|---|---|
Jan 10, 2022 |
Devashish Purandare,
Pete Wilcox,
Heiner Litz,
Shel Finkelstein,
Append is Near: Log-based Data Management on ZNS SSDs,Conference on Innovative Data Systems Research 2022 (CIDR '22), January 2022. [Archival Storage] [Designing an Efficient Flash Translation] [Designing systems for QLC flash] [Shingled Disk] [Computational Storage] |
|
Sep 7, 2015 |
Jehan-François Pâris,
Darrell D. E. Long,
Reducing the Energy Footprint of a Distributed Consensus Algorithm,Proceedings of the European Dependable Computing Conference, September 2015. [Designing an Efficient Flash Translation] |
|
Sep 9, 2014 |
Yan Li,
Darrell D. E. Long,
Which Storage Device is Greenest? Modeling the Energy Cost of I/O Workloads,Proceedings of the Twenty-second International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems , September 2014. [Designing an Efficient Flash Translation] |
Last modified 23 May 2019