Beating the I/O bottleneck: a case for log-structured file systems
ACM SIGOPS Operating Systems Review
The design and implementation of a log-structured file system
ACM Transactions on Computer Systems (TOCS)
LinLogFS: a log-structured filesystem for Linux
ATEC '00 Proceedings of the annual conference on USENIX Annual Technical Conference
Implementation of a Linux log-structured file system with a garbage collector
ACM SIGOPS Operating Systems Review
DejaView: a personal virtual computer recorder
Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
Evaluating Non-In-Place Update Techniques for Flash-Based Transaction Processing Systems
DEXA '09 Proceedings of the 20th International Conference on Database and Expert Systems Applications
Proceedings of the 2009 International Conference on Hybrid Information Technology
FlashVM: revisiting the virtual memory hierarchy
HotOS'09 Proceedings of the 12th conference on Hot topics in operating systems
FIOS: a fair, efficient flash I/O scheduler
FAST'12 Proceedings of the 10th USENIX conference on File and Storage Technologies
Revisiting storage for smartphones
FAST'12 Proceedings of the 10th USENIX conference on File and Storage Technologies
BURRITO: wrapping your lab notebook in computational infrastructure
TaPP'12 Proceedings of the 4th USENIX conference on Theory and Practice of Provenance
Revisiting storage for smartphones
ACM Transactions on Storage (TOS)
FlashFQ: a fair queueing I/O scheduler for flash-based SSDs
USENIX ATC'13 Proceedings of the 2013 USENIX conference on Annual Technical Conference
I/O stack optimization for smartphones
USENIX ATC'13 Proceedings of the 2013 USENIX conference on Annual Technical Conference
Journaling of journal is (almost) free
FAST'14 Proceedings of the 12th USENIX conference on File and Storage Technologies
Hi-index | 0.00 |
Toward enhancing the reliability of the Linux file system, we are developing a new log-structured file system (NILFS) for the Linux operating system. Instead of overwriting existing blocks, NILFS appends consistent sets of modified or newly created blocks continuously into segmented disk regions. This writing method allows NILFS to achieve faster recovery time and higher write performance. The address of the block that is written to changes for each write, which makes it difficult to apply modern file system technologies such as B-tree structures. To permit such writing on the Linux kernel basis, NILFS has its own write mechanism that handles data and meta data as one unit and allows them to be relocated. This paper presents the design and implementation of NILFS focussing on the write mechanism.