TY - GEN
T1 - Evolving EXT4 for shingled disks
AU - Aghayev, Abutalib
AU - Ts'O, Theodore
AU - Gibson, Garth
AU - Desnoyers, Peter
N1 - Funding Information:
We thank anonymous reviewers, Sage Weil (our shepherd), Phil Gibbons, and Greg Ganger for their feedback; Tim Feldman and Andy Kowles for the SMR disks and for their help with understanding the SMR disk behavior; Lin Ma, Prashanth Menon, and Saurabh Kadekodi for their help with experiments; Jan Kara for reviewing our code and for explaining the details of Linux virtual memory. We thank the member companies of the PDL Consortium (Broadcom, Citadel, Dell EMC, Facebook, Google, HewlettPackard Labs, Hitachi, Intel, Microsoft Research, MongoDB, NetApp, Oracle, Samsung, Seagate Technology, Tintri, Two Sigma, Uber, Veritas, Western Digital) for their interest, insights, feedback, and support. This research is supported in part by National Science Foundation under award CNS-1149232.
Publisher Copyright:
© Proceedings of the 15th USENIX Conference on File and Storage Technologies, FAST 2017. All rights reserved.
PY - 2017
Y1 - 2017
N2 - Drive-Managed SMR (Shingled Magnetic Recording) disks offer a plug-compatible higher-capacity replacement for conventional disks. For non-sequential workloads, these disks show bimodal behavior: After a short period of high throughput they enter a continuous period of low throughput. We introduce ext4-lazy1, a small change to the Linux ext4 file system that significantly improves the throughput in both modes. We present benchmarks on four different drive-managed SMR disks from two vendors, showing that ext4-lazy achieves 1.7-5.4× improvement over ext4 on a metadata-light file server benchmark. On metadata-heavy benchmarks it achieves 2-13× improvement over ext4 on drive-managed SMR disks as well as on conventional disks.
AB - Drive-Managed SMR (Shingled Magnetic Recording) disks offer a plug-compatible higher-capacity replacement for conventional disks. For non-sequential workloads, these disks show bimodal behavior: After a short period of high throughput they enter a continuous period of low throughput. We introduce ext4-lazy1, a small change to the Linux ext4 file system that significantly improves the throughput in both modes. We present benchmarks on four different drive-managed SMR disks from two vendors, showing that ext4-lazy achieves 1.7-5.4× improvement over ext4 on a metadata-light file server benchmark. On metadata-heavy benchmarks it achieves 2-13× improvement over ext4 on drive-managed SMR disks as well as on conventional disks.
UR - http://www.scopus.com/inward/record.url?scp=85077179709&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85077179709&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85077179709
T3 - Proceedings of the 15th USENIX Conference on File and Storage Technologies, FAST 2017
SP - 105
EP - 119
BT - Proceedings of the 15th USENIX Conference on File and Storage Technologies, FAST 2017
PB - USENIX Association
T2 - 15th USENIX Conference on File and Storage Technologies, FAST 2017
Y2 - 27 February 2017 through 2 March 2017
ER -