The design and implementation of a log-structured file system
ACM Transactions on Computer Systems (TOCS)
Analysis and evolution of journaling file systems
ATEC '05 Proceedings of the annual conference on USENIX Annual Technical Conference
File system design for an NFS file server appliance
WTEC'94 Proceedings of the USENIX Winter 1994 Technical Conference on USENIX Winter 1994 Technical Conference
OSDI '06 Proceedings of the 7th symposium on Operating systems design and implementation
Bigtable: a distributed storage system for structured data
OSDI '06 Proceedings of the 7th symposium on Operating systems design and implementation
Minimizing Latency in Fault-Tolerant Distributed Stream Processing Systems
ICDCS '09 Proceedings of the 2009 29th IEEE International Conference on Distributed Computing Systems
PLFS: a checkpoint filesystem for parallel applications
Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis
Journaling of journal is (almost) free
FAST'14 Proceedings of the 12th USENIX conference on File and Storage Technologies
Hi-index | 0.00 |
Synchronous small writes play a critical role in the reliability and availability of file systems and applications that use them to safely log recent state modifications and quickly recover from failures. However, storage stacks usually enforce page-sized granularity in their data transfers from memory to disk. We experimentally show that subpage writes may lead to storage bandwidth waste and high disk latencies. To address the issue in a journaled file system, we propose wasteless journaling as a mount mode that coalesces synchronous concurrent small writes of data into full page-sized blocks before transferring them to the journal. Additionally, we propose selective journaling that automatically applies wasteless journaling on data writes whose size lies below a fixed preconfigured threshold. In the Okeanos prototype implementation that we developed, we use microbenchmarks and application-levelworkloads to showsubstantial improvements in write latency, transaction throughput and storage bandwidth requirements.