存算分离架构下Part元数据的单独管理策略

doi:10.3969/j.issn.1000-5641.2023.05.004

华东师范大学学报（自然科学版） ›› 2023, Vol. 2023 ›› Issue (5): 40-50.doi: 10.3969/j.issn.1000-5641.2023.05.004

存算分离架构下Part元数据的单独管理策略

刘丹琪, 蔡鹏*()

华东师范大学数据科学与工程学院, 上海　200062

收稿日期:2023-06-30 接受日期:2023-07-24 出版日期:2023-09-25 发布日期:2023-09-15
通讯作者: 蔡鹏 E-mail:pcai@dase.ecnu.edu.cn
基金资助:
国家自然科学基金(61972149, U22B2020)

Separate management strategies for Part metadata under the storage-computing separation architecture

Danqi LIU, Peng CAI*()

School of Data Science and Engineering, East China Normal University, Shanghai　200062, China

Received:2023-06-30 Accepted:2023-07-24 Online:2023-09-25 Published:2023-09-15
Contact: Peng CAI E-mail:pcai@dase.ecnu.edu.cn

摘要/Abstract

摘要：

针对ClickHouse存在的硬件资源无法被充分利用、缺少弹性和节点启动过慢的问题, 在存算分离架构下, 提出了一套针对描述数据信息的元数据 (Part元数据) 的管理策略. Part元数据是元数据中最重要的组成成分. 为了能够有效管理远程共享存储上的数据, 采集了所有Part元数据文件, 并将其合并后, 经过键值映射、序列化和反序列化, 存入分布式键值数据库中. 此外, 还设计了一套同步策略, 以确保远程共享存储上的数据与分布式键值数据库中的元数据的一致性. 利用Part元数据管理策略及相关的同步策略, 实现了一个针对Part元数据的管理系统, 解决了ClickHouse节点启动过慢的问题, 并支持高效的节点动态扩缩容.

关键词: 数据库系统, 存算分离架构, 元数据管理

Abstract:

To address the deficiencies of ClickHouse, including underutilization of hardware resources, lack of flexibility, and slow node startup, this paper proposes metadata management strategies under the storage-compute separation architecture, which focuses on the description of data information through Part metadata. Part metadata are the most crucial component of metadata. To effectively manage data on remote shared storage, this study collected all Part metadata files and merged them. After key-value mapping, serialization, and deserialization processes, the merged metadata were stored in a distributed key-value database. Furthermore, a synchronization strategy was designed to ensure consistency between the data on remote shared storage and the metadata in the distributed key-value database. By implementing the above strategies, a metadata management system was developed for Part metadata, which effectively addressed the slow node startup issue in ClickHouse and supported efficient dynamic scaling of nodes.

Key words: database systems, storage-computing separation architecture, Part metadata management

中图分类号:

TP392

刘丹琪, 蔡鹏. 存算分离架构下Part元数据的单独管理策略[J]. 华东师范大学学报（自然科学版）, 2023, 2023(5): 40-50.

Danqi LIU, Peng CAI. Separate management strategies for Part metadata under the storage-computing separation architecture[J]. Journal of East China Normal University(Natural Science), 2023, 2023(5): 40-50.

图/表 10

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

参考文献 18

1	HAN W S, NG J, MARKL V, et al. Progressive optimization in a shared-nothing parallel database [C/OL]// Proceedings of the 2007 ACM SIGMOD International Conference on Management of Data. Beijing: ACM, 2007: 809-820. [2023-05-16]. https://dl.acm.org/doi/10.1145/1247480.1247569.
2	VUPPALAPATI M, TRUONG D, MIRON J, et al. Building an elastic query engine on disaggregated storage [C]// Proceedings of the 17th USENIX Symposium on Networked Systems Design and Implementation. Santa Clara, CA, USA, 2020.
3	DAGEVILLE B, CRUANES T, ZUKOWSKI M, et al. The snowflake elastic data warehouse [C/OL]// Proceedings of the 2016 International Conference on Management of Data. San Francisco, CA, USA: ACM, 2016: 215-226. [2023-05-16]. https://dl.acm.org/doi/10.1145/2882903.2903741.
4	CLICKHOUSE. Fast Open-Source OLAP DBMS [EB/OL]. [2023-05-16]. https://clickhouse.com/.
5	GAO P X, NARAYAN A, AGARWAL R, et al. Network requirements for resource disaggregation [C]// Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation. Savannah, GA, USA, 2016.
6	GU J, LEE Y, ZHANG Y, et al. Efﬁcient memory disaggregation with INFINISWAP [C]// Proceedings of the 14th USENIX Symposium on Networked Systems Design and Implementation. Boston, MA, USA, 2017.
7	DONG S, CALLAGHAN M, GALANIS L, et al. Optimizing space ampliﬁcation in RocksDB [C/OL]// 8th Biennial Conference on Innovative Data Systems Research. Chaminade, CA, USA, 2017. [2023-05-16]. https://lrita.github.io/images/posts/database.
8	ZHOU J, XU M, SHRAER A, et al.. FoundationDB: A Distributed key value Store. ACM SIGMOD Record, 2022, 51 (1): 24- 31.
9	CAMACHO-RODRÍGUEZ J, CHAUHAN A, GATES A, et al. Apache hive: From MapReduce to enterprise-grade big data warehousing [EB/OL]. (2019-03-26)[2023-05-16]. http://arxiv.org/abs/1903.10970.
10	ARMENATZOGLOU N, BASU S, BHANOORI N, et al. Amazon redshift re-invented [C/OL]// Proceedings of the 2022 International Conference on Management of Data. New York, NY, USA: Association for Computing Machinery, 2022: 2205-2217. [2023-05-29]. https://dl.acm.org/doi/10.1145/3514221.3526045.
11	GUPTA A, AGARWAL D, TAN D, et al. Amazon redshift and the case for simpler data warehouses [C/OL]// Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data. Melbourne Victoria, Australia: ACM, 2015: 1917-1923. [2023-06-19]. https://dl.acm.org/doi/10.1145/2723372.2742795.
12	HALEVY A, KORN F, NOY N F, et al. Goods: Organizing Google’s datasets [C/OL]// Proceedings of the 2016 International Conference on Management of Data. San Francisco, CA, USA: ACM, 2016: 795-806. [2023-05-16]. https://dl.acm.org/doi/10.1145/2882903.2903730.
13	EDARA P, PASUMANSKY M. Big metadata: When metadata is big data[J]. Proceedings of the VLDB Endowment 14(12): 3083-3095.
14	ZAHARIA M, CHOWDHURY M, DAS T, et al. Resilient distributed datasets: A Fault-tolerant abstraction for in-memory cluster computing [C]// Proceedings of the 9th USENIX Conference on Networked Systems Design and Implementation. Publication History, 2012.
15	AMAZON. Amazon simple storage service (user guide) [Z/OL]. (2006-03-01)[2023-06-09]. https://docs.aws.amazon.com/AmazonS3/latest/userguide/.
16	BORTHAKUR D. HDFS architecture guide [Z/OL]. (2022-05-18)[2023-06-19]. https://docs.huihoo.com/apache/hadoop/1.0.4/hdfs_design.pdf.
17	MICROSOFT AZURE. Azure blob storage [EB/OL]. [2023-05-16]. https://azure.microsoft.com/en-us/products/storage/blobs.
18	MACKEY G, SEHRISH S, WANG J. Improving metadata management for small files in HDFS [C]// 2009 IEEE International Conference on Cluster Computing and Workshops. New Orleans, LA, USA, 2009.

[1]	俞融, 杨攀飞, 王清帅, 张蓉. 数据同步机制自适应优化的HTAP数据库原型系统[J]. 华东师范大学学报（自然科学版）, 2023, 2023(5): 11-25.
[2]	朱涛, 周敏奇, 张召. 面向OceanBase的存储过程实现技术研究[J]. 华东师范大学学报(自然科学版), 2014, 2014(5): 281-289.

存算分离架构下Part元数据的单独管理策略

Separate management strategies for Part metadata under the storage-computing separation architecture

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 18

相关文章 2

编辑推荐

Metrics

本文评价