arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2306.02604cs.DB
收藏

一种简单而高性能的磁盘上学习型索引:我们能否鱼与熊掌兼得?

A Simple Yet High-Performing On-disk Learned Index: Can We Have Our Cake and Eat it Too?

Hai Lan, Zhifeng Bao, J. Shane Culpepper, Renata Borovica-Gajic, Yu Dong

AI总结:

本文提出完全磁盘上、可更新的学习型索引 AULID,通过融合传统索引与学习型索引并遵循五项设计原则降低 I/O,在多类负载中显著优于 B+-tree 及现有学习型索引。

摘要:

尽管内存中的学习型索引与 B+-tree 相比已展现出可观的性能,但实际应用中最广泛使用的数据库仍然依赖基于磁盘的操作。根据我们的实验,我们观察到将现有学习型索引直接应用于磁盘存在若干缺陷,并且在大多数情况下无法优于标准 B+-tree。因此,本工作首次尝试展示学习型索引思想如何为磁盘上索引带来收益,提出了 AULID,这是一种完全位于磁盘上、可更新的学习型索引,能够在多种工作负载类型下实现最先进性能。AULID 方法结合了传统索引技术与学习型索引的优势,以降低磁盘设置下的主要开销——I/O 成本。具体而言,降低 I/O 成本考虑了三个方面:(1)减少更新索引结构的开销;(2)引入更短的从根节点到叶节点的路径;(3)实现更好的局部性,以最小化完成一次扫描所需的块读取次数。文中提出了五项原则来指导 AULID 的设计,使其表现出显著的性能提升,同时易于实现。我们的评估表明,AULID 的存储成本与 B+-tree 相当,并且远小于其他学习型索引;AULID 的效率分别最高可达 FITing-tree、PGM、B+-tree、ALEX 和 LIPP 的 2.11 倍、8.63 倍、1.72 倍、5.51 倍和 8.02 倍。

英文摘要:

While in-memory learned indexes have shown promising performance as compared to B+-tree, most widely used databases in real applications still rely on disk-based operations. Based on our experiments, we observe that directly applying the existing learned indexes on disk suffers from several drawbacks and cannot outperform a standard B+-tree in most cases. Therefore, in this work we make the first attempt to show how the idea of learned index can benefit the on-disk index by proposing AULID, a fully on-disk updatable learned index that can achieve state-of-the-art performance across multiple workload types. The AULID approach combines the benefits from both traditional indexing techniques and the learned indexes to reduce the I/O cost, the main overhead under disk setting. Specifically, three aspects are taken into consideration in reducing I/O costs: (1) reduce the overhead in updating the index structure; (2) induce shorter paths from root to leaf node; (3) achieve better locality to minimize the number of block reads required to complete a scan. Five principles are proposed to guide the design of AULID which shows remarkable performance gains and meanwhile is easy to implement. Our evaluation shows that AULID has comparable storage costs to a B+-tree and is much smaller than other learned indexes, and AULID is up to 2.11x, 8.63x, 1.72x, 5.51x, and 8.02x more efficient than FITing-tree, PGM, B+-tree, ALEX, and LIPP.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑