面向闪存的树型索引综述

doi:10.3778/j.issn.1002-8331.2112-0066

摘要/Abstract

摘要： 随着闪存技术的不断成熟，基于闪存的固态硬盘（solid state drive，SSD）迅速发展。然而，SSD具有不同于磁盘的特性，使得传统基于磁盘设计的索引不适用于闪存环境，因此面向闪存索引机制的研究与优化迅速开展。通过对目前闪存索引的广泛调研，从索引更新策略的角度，分析了它们的优缺点，旨在为SSD算法设计和索引开发提供系统的、有价值的参考。最后讨论了该领域未来的发展趋势和新的研究方向。

关键词: 数据管理, 闪存, 树型索引, 性能优化

Abstract: With the continuous maturity of flash memory technology, solid state drive（SSD） based on flash memory has developed rapidly. However, the characteristics of SSD are different from those of disk, which makes the traditional indexing based on disk not suitable for the environment of flash memory. Therefore, the research and optimization of the indexing mechanism for flash memory are carried out rapidly. Based on extensive investigation of current flash index research, this paper analyzes their advantages and disadvantages from the view of index update strategy, in order to provide a systematic and valuable reference for SSD algorithm design and index development. In addition, the future development trends and new research directions in this field are also discussed.

Key words: data management, flash memory, tree index, performance optimization

孙鉴, 武晓晓, 谢开斌, 高锦涛, 刘凇佐, 巫思敏. 面向闪存的树型索引综述[J]. 计算机工程与应用, 2022, 58(22): 30-40.

SUN Jian, WU Xiaoxiao, XIE Kaibin, GAO Jintao, LIU Songzuo, WU Simin. Review of Tree Index Based on Flash Memory[J]. Computer Engineering and Applications, 2022, 58(22): 30-40.

参考文献

[1] 孙翠锋，王祎晨.闪存技术发展趋势及在数据中心的应用[J].互联网天地，2020（12）：57-60.
SUN C F，WANG Y C.Development trend and application of flash memory technology in data center[J].Internet World，2020（12）：57-60.
[2] 宋俊辉.面向高级命令的闪存转换层算法研究[D].武汉：华中科技大学，2017.
SONG J H.Research on flash translation layer algorithm for advanced commands[D].Wuhan：Huazhong University of Science and Technology，2017.
[3] 李红艳.针对固态盘的I/O优化技术研究[D].武汉：华中科技大学，2016.
LI H Y.Research on I/O optimization technology based on solid state disk[D].Wuhan：Huazhong University of Science and Technology，2016.
[4] 徐海娟.保留时间感知的闪存转换层设计与实现[D].武汉：华中科技大学，2017.
XU H J.Design and implementation of retention-time aware flash translation layer[D].Wuhan：Huazhong University of Science and Technology，2017.
[5] 李楚.固态盘缓存系统基于闪存特性的优化研究[D].武汉：华中科技大学，2017.
LI C.Exploring inherent features of flash memory for SSD-based cache optimization[D].Wuhan：Huazhong University of Science and Technology，2017.
[6] 周斯忠，陈耀武.一种应用于闪存数据库的高效B+tree索引机制[J].计算机工程，2013，39（9）：1-5.
ZHOU S Z，CHEN Y W.An efficient B+tree index mechanism for flash-based database[J].Computer Engineering，2013，39（9）：1-5.
[7] 曾思望.基于固态盘的分布式块存储系统缓存技术研究[D].武汉：华中科技大学，2015.
ZENG S W.The research of SSD-based cache techniques for distributed block storage system[D].Wuhan：Huazhong University of Science and Technology，2015.
[8] 罗蜜.基于数据生存期的固态盘性能优化策略研究[D].武汉：华中科技大学，2017.
LUO M.Research on performance optimization for solid-state drive based on data lifetime[D].Wuhan：Huazhong University of Science and Technology，2017.
[9] LEE Y G，JUNG D，KANG D，et al.μ-FTL：a memory-efficient flash translation layer supporting multiple mapping granularities[C]//Proceedings of the 8th ACM International Conference on Embedded Software，2008：21-30.
[10] 崔凯.混合结构闪存索引研究[D].合肥：中国科学技术大学，2010.
CUI K.Research on hybrid index structure for flash disks[D].Hefei：University of Science and Technology of China，2010.
[11] 林根.基于数据类型语义的闪存转换层的设计与实现[D].武汉：华中科技大学，2017.
LIN G.The design and implementation of flash translation layer based on semantic of data type[D].Wuhan：Huazhong University of Science and Technology，2017.
[12] 曾祥伟.基于闪存特性的固态盘缓存算法研究[D].广州：暨南大学，2020.
ZENG X W.Research on solid state disk cache algorithm based on flash memory characteristics[D].Guangzhou：Jinan University，2020.
[13] GALAKATOS A，MARKOVITCH M，BINNIG C，et al.Fiting-tree：a data-aware index structure[C]//2019 International Conference on Management of Data，2019：1189-1206.
[14] LEVANDOSKI J J，LOMET D B，SENGUPTA S.The BW-tree：a B-tree for new hardware platforms[C]//2013 IEEE 29th International Conference on Data Engineering，2013：302-313.
[15] ROH H，KIM S，LEE D，et al.AS B-tree：a study of an efficient B-tree for SSDs[J].Journal of Information Science and Engineering，2014，30（1）：85-106.
[16] WU C H，CHANG L P，KUO T W.An efficient B-tree layer for flash-memory storage systems[C]//9th International Conference on Real-Time and Embedded Computer Systems and Applications.Berlin，Heidelberg：Springer，2003：409-430.
[17] ROH H，PARK S，KIM S，et al.B+-tree index optimization by exploiting internal parallelism of flash-based solid state drives[J].arXiv：1201.0227，2011.
[18] ROH H，PARK S，SHIN M，et al.MPSearch：Multi-path search for tree-based indexes to exploit internal parallelism of flash SSDs[J].IEEE Data Engineering Bulletin，2014，37（2）：3-11.
[19] 梁新伟.基于NAND闪存的数据库索引机制研究与改进[D].沈阳：东北大学，2013.
LIANG X W.Research and improvement of database index based on NAND flash[D].Shenyang：Northeastern University，2013.
[20] JIN P，YANG C，JENSEN C S，et al.Read/write-optimized tree indexing for solid-state drives[J].The VLDB Journal，2016，25（5）：695-717.
[21] YIN S，PUCHERAL P，MENG X.PBfilter：indexing flash-resident data through partitioned summaries[C]//17th ACM Conference on Information and Knowledge Management，2008：1333-1334.
[22] BYUN S，HUH M，HWANG H.An index rewriting scheme using compression for flash memory database systems[J].Journal of Information Science，2007，33（4）：398-415.
[23] XIE B，LI Q，WANG Y，et al.EPLA-DSTree：extending piecewise linear approximation on a dynamic segmentation tree index in sensor-cloud systems[J].IEEE Access，2020，9：123742-123753.
[24] ATHANASSOULIS M，AILAMAKI A.BF-tree：approximate tree indexing[C]//40th International Conference on Very Large Databases，2014.
[25] 杨程程.基于闪存的索引机制研究[D].合肥：中国科学技术大学，2017.
YANG C C.Research on index mechanism for flash based devices[D].Hefei：University of Science and Technology of China，2017.
[26] 郝行军.物联网大数据存储与管理技术研究[D].合肥：中国科学技术大学，2017.
HAO X J.Reseach on technology of storage and management of IoT data[D].Hefei：University of Science and Technology of China，2017.
[27] JIN R，CHO H J，CHUNG T S.LS-LRU：a lazy-split LRU buffer replacement policy for flash-based B+-tree index[J].Journal of Information Science and Engineering，2015，31（3）：1113-1132.
[28] LEE H S，LEE D H.An efficient index buffer management scheme for implementing a B-tree on NAND flash memory[J].Data & Knowledge Engineering，2010，69（9）：901-916.
[29] LEE H S，PARK S，SONG H J，et al.An efficient buffer management scheme for implementing a B-tree on NAND flash memory[C]//3rd International Conference on Embedded Software and Systems.Berlin，Heidelberg：Springer，2007：181-192.
[30] XIANG X，YUE L，LIU Z，et al.A reliable B-tree implementation over flash memory[C]//2008 ACM Symposium on Applied Computing，2008：1487-1491.
[31] 刘颖杰，林子雨，赖永炫.CF-HNLBI：一种新的闪存数据库B-树索引[J].厦门大学学报（自然科学版），2015，54（2）：247-256.
LIU Y J，LIN Z Y，LAI Y X.CF-HNLBI：a new B-tree indexing for flash-based databases[J].Journal of Xiamen University（Natural Science），2015，54（2）：247-256.
[32] FEVGAS A，AKRITIDIS L，BOZANIS P，et al.Indexing in flash storage devices：a survey on challenges，current approaches，and future trends[J].The VLDB Journal，2020，29（1）：273-311.
[33] ON S T，HU H，LI Y，et al.Lazy-update B+-tree for flash devices[C]//2009 10th International Conference on Mobile Data Management：Systems，Services and Middleware，2009：323-328.
[34] ROH H，KIM W C，KIM S，et al.A B-tree index extension to enhance response time and the life cycle of flash memory[J].Information Sciences，2009，179（18）：3136-3161.
[35] 周大，梁智超，孟小峰.HF-Tree：一种闪存数据库的高更新性能索引结构[J].计算机研究与发展，2010，47（5）：832-840.
ZHOU D，LIANG Z C，MENG X F.HF-Tree：an update-efficient index for flash memory[J].Journal of Computer Research and Development，2010，47（5）：832-840.
[36] 付佳.基于LSM树的NoSQL数据库索引研究[D].北京：北京理工大学，2016.
FU J.Research on NoSQL database index based on LSM tree[D].Beijing：Beijing Institute of Technology，2016.
[37] 马文龙，朱妤晴，蒋德钧，等.Key-Value型NoSQL本地存储系统研究[J].计算机学报，2018，41（8）：1722-1751.
MA W L，ZHU Y Q，JIANG D Y，et al.A survey on local key-value store of NoSQL system[J].Chinese Journal of Computers，2018，41（8）：1722-1751.
[38] ZHOU X，SHOU L，CHEN K，et al.DPTree：differential indexing for persistent memory[J].Proceedings of the VLDB Endowment，2019，13（4）：421-434.
[39] LI Y，HE B，LUO Q，et al.Tree indexing on flash disks[C]//2009 IEEE 25th International Conference on Data Engineering，2009：1303-1306.
[40] LI Y，HE B，YANG R J，et al.Tree indexing on solid state drives[J].Proceedings of the VLDB Endowment，2010，3（1/2）：1195-1206.
[41] 马伟.基于FD-tree的闪存数据库索引技术研究[D].杭州：浙江大学，2011.
MA W.The FD-tree reexamined：a case for indexing flash database[D].Hangzhou：Zhejiang University，2011.
[42] CHOI W G，SHIN M，LEE D，et al.Optimization of a multiversion index on SSDs to improve system performance[C]//2016 IEEE International Conference on Systems，Man，and Cybernetics，2016：1620-1625.
[43] 吴桐.基于闪存的数据库索引技术研究[D].北京：北京邮电大学，2014.
WU T.Flash-based database indexing technology research[D].Beijing：Beijing University of Posts and Telecommunications，2014.
[44] 房俊华，王翰虎，陈梅，等.一种具有自适应机制的闪存数据库索引结构[J].计算机应用，2013，33（2）：563-566.
FANG J H，WANG H H，CHEN M，et al.Index structure with self-adaptive mechanism in flash-based database system[J].Journal of Computer Applications，2013，33（2）：563-566.
[45] NA G J，LEE S W，MOON B.Dynamic in-page logging for B+-tree index[J].IEEE Transactions on Knowledge and Data Engineering，2011，24（7）：1231-1243.
[46] NATH S，KANSAL A.FlashDB：dynamic self-tuning database for NAND flash[C]//6th International Conference on Information Processing in Sensor Networks，2007：410-419.
[47] AGRAWAL D，GANESAN D，SITARAMAN R，et al.Lazy-adaptive tree：an optimized index structure for flash devices[J].Proceedings of the VLDB Endowment，2009，2（1）：361-372.
[48] 房俊华，王翰虎，陈梅，等.DB-tree：一种高性能的闪存数据库索引结构[J].计算机应用与软件，2013，30（11）：243-246.
FANG J H，WANG H H，CHEN M，et al.DB-tree：an efficient index structure for flash-based database[J].Computer Applications and Software，2013，30（11）：243-246.
[49] 叶锋.基于闪存的混合存储系统缓存算法研究[D].武汉：华中科技大学，2015.
YE F.Research of cache replacement algorithms for flash-based hybrid storage system[D].Wuhan：Huazhong University of Science and Technology，2015.
[50] TANG C，DONG Z，WANG M，et al.Learned indexes for dynamic workloads[J].arXiv：1902.00655，2019.
[51] ZHU Z，MUN J H，RAMAN A，et al.Reducing Bloom Filter CPU Overhead in LSM-Trees on Modern Storage Devices[C]//17th International Workshop on Data Management on New Hardware，2021：1-10.
[52] THONANGI R，BABU S，YANG J.A practical concurrent index for solid-state drives[C]//21st ACM International Conference on Information and Knowledge Management，2012：1332-1341.
[53] LIU J，CHEN S，WANG L.LB+trees：optimizing persistent index performance on 3DXPoint memory[J].Proceedings of the VLDB Endowment，2020，13（7）：1078-1090.
[54] 杨国平，乔少杰，屈露露，等.学习型数据库索引推荐技术综述[J].重庆理工大学学报（自然科学），2022，36（6）：189-199.
YANG G P，QIAO S J，QU L L，et al.A survey of learning based database index advisor technology[J].Journal of Chongqing University of Technology（Natural Science），2022，36（6）：189-199.
[55] YAN Y，YAO S，WANG H，et al.Index selection for NoSQL database with deep reinforcement learning[J].Information Sciences，2021，561：20-30.
[56] 陈国栋.固态硬盘I/O性能抖动优化算法研究[D].合肥：安徽大学，2019.
CHEN G D.Research on I/O performance jitter optimization algorithm for solid state disk[D].Hefei：Anhui University，2019.
[57] 杨伟健.固态盘存储系统的性能优化研究[D].厦门：厦门大学，2017.
YANG W J.Research on performance optimization for SSD-based storage systems[D].Xiamen：Xiamen University，2017.
[58] TOSHNIWAL A，TANEJA S，SHUKLA A，et al.Storm@ twitter[C]//2014 ACM SIGMOD International Conference on Management of Data，2014：147-156.
[59] CARBONE P，KATSIFODIMOS A，EWEN S，et al.Apache Flink：stream and batch processing in a single engine[J].IEEE Data Engineering Bulletin，2015，38（1）：28-38.
[60] 何龙，陈晋川，杜小勇.一种面向HDFS的多层索引技术[J].软件学报，2017，28（3）：502-513.
HE L，CHEN J C，DU X Y.Multi-layered index for HDFS-based systems[J].Journal of Software，2017，28（3）：502-513.
[61] 李斌，郭景维，彭骞.面向大数据存储的HBase二级索引设计[J].计算技术与自动化，2019，38（2）：124-129.
LI B，GUO J W，PENG Q.Design of HBase secondary indexes for big data storage[J].Computing Technology and Automation，2019，38（2）：124-129.
[62] KONDYLAKIS H，DAYAN N，ZOUMPATIANOS K，et al.Coconut：sortable summarizations for scalable indexes over static and streaming data series[J].The VLDB Journal，2019，28（6）：847-869.
[63] WANG L，CAI R，FU T Z J，et al.Waterwheel：realtime indexing and temporal range query processing over massive data streams[C]//2018 IEEE 34th International Conference on Data Engineering，2018：269-280.
[64] 祝磊.NAND固态盘缓存区管理及磨损均衡算法研究[D].长沙：湖南大学，2018.
ZHU L.A study on buffer management and wear leveling algorithm for NAND solid state drives[D].Changsha：Hunan University，2018.
[65] 孙杰贤.华数传媒：创新，引领与傲腾[J].中国信息化，2019（1）：42-43.
SUN J X.Washu Media：innovation，leading and Optane[J].China Informatization，2019（1）：42-43.
[66] 小平.用好Intel傲腾内存[J].电脑知识与技术，2019（2）：81-82.
XIAO P.Make good use of Intel Optane memory[J].Computer Knowledge and Technology，2019（2）：81-82.