Apologizing versus asking permission: optimistic concurrency control for abstract data types
ACM Transactions on Database Systems (TODS)
SIGMOD '90 Proceedings of the 1990 ACM SIGMOD international conference on Management of data
Scheduling algorithms for modern disk drives
SIGMETRICS '94 Proceedings of the 1994 ACM SIGMETRICS conference on Measurement and modeling of computer systems
On optimistic methods for concurrency control
ACM Transactions on Database Systems (TODS)
Inside Windows NT
Unifying File System Protection
Proceedings of the General Track: 2002 USENIX Annual Technical Conference
FS: An In-Kernel Integrity Checker and Intrusion Detection File System
LISA '04 Proceedings of the 18th USENIX conference on System administration
Proceedings of the twentieth ACM symposium on Operating systems principles
Analysis and evolution of journaling file systems
ATEC '05 Proceedings of the annual conference on USENIX Annual Technical Conference
Design and evaluation of a continuous consistency model for replicated services
OSDI'00 Proceedings of the 4th conference on Symposium on Operating System Design & Implementation - Volume 4
OSDI '06 Proceedings of the 7th USENIX Symposium on Operating Systems Design and Implementation - Volume 7
Metadata update performance in file systems
OSDI '94 Proceedings of the 1st USENIX conference on Operating Systems Design and Implementation
Journaling versus soft updates: asynchronous meta-data protection in file systems
ATEC '00 Proceedings of the annual conference on USENIX Annual Technical Conference
Scalability in the XFS file system
ATEC '96 Proceedings of the 1996 annual conference on USENIX Annual Technical Conference
Generalized file system dependencies
Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
LazyBase: trading freshness for performance in a scalable database
Proceedings of the 7th ACM european conference on Computer Systems
FAST'12 Proceedings of the 10th USENIX conference on File and Storage Technologies
Probabilistically bounded staleness for practical partial quorums
Proceedings of the VLDB Endowment
Disks from the perspective of a file system
Communications of the ACM
Robustness in the Salus scalable block store
nsdi'13 Proceedings of the 10th USENIX conference on Networked Systems Design and Implementation
Towards efficient, portable application-level consistency
Proceedings of the 9th Workshop on Hot Topics in Dependable Systems
ViewBox: integrating local file systems with cloud storage services
FAST'14 Proceedings of the 12th USENIX conference on File and Storage Technologies
Journaling of journal is (almost) free
FAST'14 Proceedings of the 12th USENIX conference on File and Storage Technologies
Blizzard: fast, cloud-scale block storage for cloud-oblivious applications
NSDI'14 Proceedings of the 11th USENIX Conference on Networked Systems Design and Implementation
| Hi-index | 0.00 |
We introduce optimistic crash consistency, a new approach to crash consistency in journaling file systems. Using an array of novel techniques, we demonstrate how to build an optimistic commit protocol that correctly recovers from crashes and delivers high performance. We implement this optimistic approach within a Linux ext4 variant which we call OptFS. We introduce two new file-system primitives, osync() and dsync(), that decouple ordering of writes from their durability. We show through experiments that OptFS improves performance for many workloads, sometimes by an order of magnitude; we confirm its correctness through a series of robustness tests, showing it recovers to a consistent state after crashes. Finally, we show that osync() and dsync() are useful in atomic file system and database update scenarios, both improving performance and meeting application-level consistency demands.