Paper Details
Reference:
Ruibin Li, Xiang Ren, Xu Zhao, Siwei He, Michael Stumm, and Ding Yuan,
ctFS: Converting File Index Traversals to Hardware Memory Translation through Contiguous File Allocation for Persistent Memory",
;login:, February, 2022, pp. online.
Download:
Abstract:
Persistent byte-addressable memory (PM) is poised to become prevalent in future computer systems. PMs are significantly faster than disk storage, and accesses to PMs are governed by the Memory Management Unit (MMU) just as accesses with volatile RAM. These unique characteristics shift the bottleneck from I/O to operations such as block address lookup---for example, in write workloads, up to 45% of the overhead in ext4-DAX is due to building and searching extent trees to translate file offsets to addresses on persistent memory.We propose a novel contiguous file system, ctFS, that eliminates most of the overhead associated with indexing structures such as extent trees in the file system. ctFS represents each file as a contiguous region of virtual memory, hence a lookup from the file offset to the address is simply an offset operation, which can be efficiently performed by the hardware MMU at a fraction of the cost of software-maintained indexes. Evaluating ctFS on real-world workloads such as LevelDB shows it outperforms ext4-DAX and SplitFS by 3.6× and 1.8×, respectively.
Keywords:
File systems management; allocation/deallocation strategies; persistent memory, file system, page table, memory allocation, data center
BibTeX:
@article(Li-Login22,
author = {Ruibin Li and Xiang Ren and Xu Zhao and Siwei He and Michael Stumm and Ding Yuan},
title = {ctFS: Converting File Index Traversals to Hardware Memory Translation through Contiguous File Allocation for Persistent Memory},
month = {February},
year = {2022},
pages = {online},
keywords = {File systems management; allocation/deallocation strategies; persistent memory, file system, page table, memory allocation, data center}
)