RAID has been widely adopted to enhance the performance, capacity, and reliability of the existing storage systems. However, we observe that the Linux software RAID (mdraid) suffers from its poor implementation of the lock mechanism. To address this, we propose ScalaRAID, which refines the role domain of locks and designs a new data structure to prevent different threads from preempting the RAID resources. By doing so, ScalaRAID can maximize the thread-level parallelism and reduce the time consumption of I/O request handling. Our evaluation results reveal that ScalaRAID can improve throughput by 89.4% while decreasing 99.99th percentile latency by 85.4% compared to mdraid.
|Title of host publication||HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems|
|Publisher||Association for Computing Machinery, Inc|
|Number of pages||7|
|Publication status||Published - 2022 Jun 27|
|Event||14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022 - Virtual, Online, United States|
Duration: 2022 Jun 27 → 2022 Jun 28
|Name||HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems|
|Conference||14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022|
|Period||22/6/27 → 22/6/28|
Bibliographical noteFunding Information:
The authors thank Prof. Guangyan Zhang of Tsinghua University for shepherding their paper. This research is mainly supported by Peking University start-up package (7100603645). Dr. Jung is in part supported by NRF 2021R1AC4001773 and IITP 2021-0-00524 & 2022-0-00117, KAIST IDEC & Start-up (G01190015), Samsung HiPHER, and Samsung Research Grant (G01200447). Jie Zhang is the corresponding author.
© 2022 ACM.
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications
- Hardware and Architecture
- Information Systems