A Flexible Garbage Collection Scheme with Low Overhead for ZNS SSDs

Garbage collection (GC) has long impacted the ability of solid state disks (SSD) to function properly and has become an important issue that many memory manufacturers are eager to address. In recent years, with the advent of zoned namespace SSDs (ZNS SSD), the garbage collection process has become more controllable. The host side is able to selectively trigger the process of garbage collection and the area to be covered, so as to meet the real-time demand of applications. In this paper, we propose a flexible garbage collection scheme that aims to avoid affecting the normal read and write operations of applications as much as possible and further reduce the overhead of garbage collection. The experiments are conducted in a simulator capable of emulating zoned storage. Extensive experiments show that our approach is able to reduce read and write latency while reducing write amplification.

Authors

Institutions

Publication Details

Published
2024-12-20
DOI
https://doi.org/10.1145/3712623.3712631
Primary Topic
Advanced Data Storage Technologies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A Flexible Garbage Collection Scheme with Low Overhead for ZNS SSDs

Yaojie Chen, Weifeng Zou, Kang Jiu
Advanced Data Storage Technologies
article

A Flexible Garbage Collection Scheme with Low Overhead for ZNS SSDs

Yaojie Chen, Weifeng Zou, Kang Jiu
article en

Abstract

Garbage collection (GC) has long impacted the ability of solid state disks (SSD) to function properly and has become an important issue that many memory manufacturers are eager to address. In recent years, with the advent of zoned namespace SSDs (ZNS SSD), the garbage collection process has become more controllable. The host side is able to selectively trigger the process of garbage collection and the area to be covered, so as to meet the real-time demand of applications. In this paper, we propose a flexible garbage collection scheme that aims to avoid affecting the normal read and write operations of applications as much as possible and further reduce the overhead of garbage collection. The experiments are conducted in a simulator capable of emulating zoned storage. Extensive experiments show that our approach is able to reduce read and write latency while reducing write amplification.

Wuhan University of Science and Technology (CN)
Openalex Percentile: Top 68%
Advanced Data Storage Technologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.