Maximizing I/O Throughput and Minimizing Performance Variation via Reinforcement Learning Based I/O Merging for SSDs

Chao Wu, Cheng Ji*, Qiao Li, Congming Gao, Riwei Pan, Chenchen Fu, Liang Shi, Chun Jason Xue

*Corresponding author for this work

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

24 Citations (Scopus)

Abstract

Merging technique is widely adopted by I/O schedulers to maximize system I/O throughput. However, I/O merging could increase the latency of individual I/O, thus incurring prolonged I/O latencies and enlarged performance variations. Even with better system throughput, higher worst-case latency experienced by some requests could block the SSD storage system, which violates the QoS (Quality of Service) requirement. In order to improve QoS performance while providing higher I/O throughput, this paper proposes a reinforcement learning based I/O merging approach. Through learning the characteristic of various I/O patterns, the proposed approach makes merging decisions adaptively based on different I/O workloads. Evaluation results show that the proposed scheme is capable of reducing the standard deviation of I/O latency by 19.1 percent on average, worst-case latency by 7.3-60.9 percent at the 99.9th percentile compared with the latest I/O merging scheme, while maximizing system throughput.
Original languageEnglish
Article number8823009
Pages (from-to)72-86
JournalIEEE Transactions on Computers
Volume69
Issue number1
Online published3 Sept 2019
DOIs
Publication statusPublished - Jan 2020

Research Keywords

  • I/O scheduler
  • Merging technique
  • performance variation
  • reinforcement learning
  • throughput
  • worst-case latency

Fingerprint

Dive into the research topics of 'Maximizing I/O Throughput and Minimizing Performance Variation via Reinforcement Learning Based I/O Merging for SSDs'. Together they form a unique fingerprint.

Cite this