基于延迟感知的混合ssd读性能优化页面迁移

Shicheng Li, Longfei Luo, Yina Lv, Liang Shi
{"title":"基于延迟感知的混合ssd读性能优化页面迁移","authors":"Shicheng Li, Longfei Luo, Yina Lv, Liang Shi","doi":"10.1109/NVMSA56066.2022.00015","DOIUrl":null,"url":null,"abstract":"The high-density flash memory makes it possible to store more information in a memory cell, while the drawbacks to it are degraded performance, less endurance, and wide variation of read latency. The prevalent hybrid solid state drives (SSDs) utilize a single-level cell (SLC) region as the write cache of high-density flash memory region and solve the problem of writes. However, this design did not take the read latency variation into consideration. Through evaluations on real storage device, we demonstrated the variation of read latency and presented the benefits and limitations on read performance optimization. Motivated by that, an efficient migration strategy, latency aware page migration (LAPM), is proposed. Its basic idea is to identify frequently read pages with high read latency and migrate them to the SLC region to improve the overall read performance. Experimental results show that LAPM can effectively reduce the read latency by 20% on average over the baseline while lowering the migration traffic to the device by up to 40%, compared with state-of-the-arts.","PeriodicalId":185204,"journal":{"name":"2022 IEEE 11th Non-Volatile Memory Systems and Applications Symposium (NVMSA)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Latency Aware Page Migration for Read Performance Optimization on Hybrid SSDs\",\"authors\":\"Shicheng Li, Longfei Luo, Yina Lv, Liang Shi\",\"doi\":\"10.1109/NVMSA56066.2022.00015\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The high-density flash memory makes it possible to store more information in a memory cell, while the drawbacks to it are degraded performance, less endurance, and wide variation of read latency. The prevalent hybrid solid state drives (SSDs) utilize a single-level cell (SLC) region as the write cache of high-density flash memory region and solve the problem of writes. However, this design did not take the read latency variation into consideration. Through evaluations on real storage device, we demonstrated the variation of read latency and presented the benefits and limitations on read performance optimization. Motivated by that, an efficient migration strategy, latency aware page migration (LAPM), is proposed. Its basic idea is to identify frequently read pages with high read latency and migrate them to the SLC region to improve the overall read performance. Experimental results show that LAPM can effectively reduce the read latency by 20% on average over the baseline while lowering the migration traffic to the device by up to 40%, compared with state-of-the-arts.\",\"PeriodicalId\":185204,\"journal\":{\"name\":\"2022 IEEE 11th Non-Volatile Memory Systems and Applications Symposium (NVMSA)\",\"volume\":\"44 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE 11th Non-Volatile Memory Systems and Applications Symposium (NVMSA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/NVMSA56066.2022.00015\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE 11th Non-Volatile Memory Systems and Applications Symposium (NVMSA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/NVMSA56066.2022.00015","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

高密度闪存使得在一个存储单元中存储更多的信息成为可能,但它的缺点是性能下降、续航时间较短、读取延迟变化大。目前流行的混合固态硬盘(ssd)利用单级单元(SLC)区域作为高密度闪存区域的写缓存,解决了写问题。然而,这个设计没有考虑到读延迟的变化。通过对实际存储设备的评估,我们展示了读延迟的变化,并提出了读性能优化的好处和局限性。在此基础上,提出了一种高效的迁移策略——感知延迟页面迁移(latency - aware page migration, LAPM)。其基本思想是识别具有高读延迟的频繁读页面,并将其迁移到SLC区域,以提高整体读性能。实验结果表明,与最先进的技术相比,LAPM可以有效地将读取延迟平均降低20%,同时将设备迁移流量降低高达40%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Latency Aware Page Migration for Read Performance Optimization on Hybrid SSDs
The high-density flash memory makes it possible to store more information in a memory cell, while the drawbacks to it are degraded performance, less endurance, and wide variation of read latency. The prevalent hybrid solid state drives (SSDs) utilize a single-level cell (SLC) region as the write cache of high-density flash memory region and solve the problem of writes. However, this design did not take the read latency variation into consideration. Through evaluations on real storage device, we demonstrated the variation of read latency and presented the benefits and limitations on read performance optimization. Motivated by that, an efficient migration strategy, latency aware page migration (LAPM), is proposed. Its basic idea is to identify frequently read pages with high read latency and migrate them to the SLC region to improve the overall read performance. Experimental results show that LAPM can effectively reduce the read latency by 20% on average over the baseline while lowering the migration traffic to the device by up to 40%, compared with state-of-the-arts.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信