TY - GEN
T1 - Modernizing file system through in-storage indexing
AU - Koo, Jinhyung
AU - Im, Junsu
AU - Song, Jooyoung
AU - Park, Juhyung
AU - Lee, Eunji
AU - Kim, Bryan S.
AU - Lee, Sungjin
N1 - Funding Information:
We would like to thank our shepherd, Dr. Donald E. Porter, and five anonymous reviewers for all their helpful comments. This work was supported by Samsung Electronics Co., Ltd. and the National Research Foundation (NRF) of Korea (NRF-2018R1A5A1060031 and NRF-2019R1A2C1090337). We thank Samsung Electronics for providing KV-SSD prototypes. (Corresponding author: Sungjin Lee)
Publisher Copyright:
© 2021 by The USENIX Association. All rights reserved.
PY - 2021
Y1 - 2021
N2 - We argue that a key-value interface between a file system and an SSD is superior to the legacy block interface by presenting KEVIN. KEVIN combines a fast, lightweight, and POSIX-compliant file system with a key-value storage device that performs in-storage indexing. We implement a variant of a log-structured merge tree in the storage device that not only indexes file objects, but also supports transactions and manages physical storage space. As a result, the design of a file system with respect to space management and crash consistency is simplified, requiring only 10.8K LOC for full functionality. We demonstrate that KEVIN reduces the amount of I/O traffic between the host and the device, and remains particularly robust as the system ages and the data become fragmented. Our approach outperforms existing file systems on a block SSD by a wide margin – 6.2× on average – for metadata-intensive benchmarks. For realistic workloads, KEVIN improves throughput by 68% on average.
AB - We argue that a key-value interface between a file system and an SSD is superior to the legacy block interface by presenting KEVIN. KEVIN combines a fast, lightweight, and POSIX-compliant file system with a key-value storage device that performs in-storage indexing. We implement a variant of a log-structured merge tree in the storage device that not only indexes file objects, but also supports transactions and manages physical storage space. As a result, the design of a file system with respect to space management and crash consistency is simplified, requiring only 10.8K LOC for full functionality. We demonstrate that KEVIN reduces the amount of I/O traffic between the host and the device, and remains particularly robust as the system ages and the data become fragmented. Our approach outperforms existing file systems on a block SSD by a wide margin – 6.2× on average – for metadata-intensive benchmarks. For realistic workloads, KEVIN improves throughput by 68% on average.
UR - http://www.scopus.com/inward/record.url?scp=85113881282&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85113881282&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85113881282
T3 - Proceedings of the 15th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2021
SP - 75
EP - 92
BT - Proceedings of the 15th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2021
PB - USENIX Association
T2 - 15th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2021
Y2 - 14 July 2021 through 16 July 2021
ER -