The Composite-file File System: Decoupling One-to-one Mapping of Files and Metadata for Better Performance

Grants: NSF grants CNS-144387 and FSU Research Foundation


PIs:
Shuanglong Zhang, Google
Robert Roy, Florida State Univerity
Leah Rumancik, Florida State University
Helen Catanese, Florida State University
An-I Andy Wang, Florida State University


The design and implementation of traditional file systems typically use the one-to-one mapping of logical files to their physical metadata representations. File system optimizations generally follow this rigid mapping and miss opportunities for an entire class of optimizations.

We designed, implemented, and evaluated a composite-file file system, which allows many-to-one mappings of files to metadata. Through exploring different mapping strategies, our empirical evaluation shows up to a 27% performance improvement under webserver and software development workloads, for both disks and SSDs. This result demonstrates that our approach of relaxing file-to-metadata mapping is promising.

Figure 1: Creation of the internal composite file (bottom) from the two original files (top).


Publications


Zhang20
Shuanglong Zhang, Robert Roy, Leah Rumancik, and An-I Andy Wang. The Composite-file File System: Decoupling One-to-one Mapping of Files and Metadata for Better Performance. ACM Transactions on Storage (TOS), 16(1), 2020. [20% acceptance rate for the original USENIX FAST paper]

Zhang16
Shuanglong Zhang, Helen Catanese, and An-I Andy Wang. The Composite-file File System: Decoupling the One-to-one Mapping of Files and Metadata for Better Performance. Proceedings of the 14th USENIX Conference on File and Storage Technologies (FAST), February 2016 [8 pages, 20% acceptance rate] Supersedes Technical Report TR-150921, Department of Computer Science, Florida State University, September 2015.


Presentations



Shuanglong Zhang. The Composite-file File System: Decoupling the One-to-one Mapping of Files and Metadata for Better Performance, slides. Presented at the 14th USENIX Conference on File and Storage Technologies, Santa Clara, CA, February 23, 2016.


Press Coverage



Kate Mueller. Neuroscience Student Wins Timed Thesis Competition. Florida State 24/7, November 19, 2014.
 


This material is based upon work supported by the National Science Foundation under grant no. CNS-1065127. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation, US Department of Education, the Philanthropic Education Organization, or the FSU Research Foundation.


mitchell@cs.fsu.edu

Last modified:  June 23, 2015