暂无图片
暂无图片
暂无图片
暂无图片
暂无图片
一种内存搜索树的键值保序压缩技术-熊艳辉.pdf
259
22页
35次
2022-12-01
免费下载
ENMOTECH
Order-Preserving Key Compression for
In-Memory Search Trees
作者: 熊艳辉
部门: 核心产品研发2部
日期: 2022/12/01
SIGMOD'20 论文学习分享
数 / 据 / 驱 / 动 成 / 就 / 未 / 来
Scenarios
In-memory search Tree(内存数据结构,速数据查询,解决内存footprint问题)
Solution
High-speed Order-Preserving Encoder (HOPE压缩框架)
Key Techniques
Fast dictionary-based compressor
Identify common key patterns at a fine granularity
Exploit the entropy to achieve high compression rates with a small dictionary
Encodes arbitrary keys while preserving their order
Contributions
Develop a theoretical model to reason about order-preserving dictionary designs
Six representative compression schemes implement model in HOPE
Evaluate HOPE on five data structures used in databases: SuRF, ART, HOT, B+tree, and Prefix B+tree
Conclusion
Lower query latency(up to 40% lower) and better memory efficiency (up to 30% smaller) simultaneously for most
string key workloads
latency
40%
memory footprint
30%
研究结论
数 / 据 / 驱 / 动 成 / 就 / 未 / 来
SIGMOD'20, June 14–19, 2020, Portland, OR, USA
Huanchen Zhang
Carnegie Mellon University(CMU)
Background
SSD have increased the performance demand on main-memory data structure
growing cost gap between DRAM and SSD storage($/GB ratio from 10x in 2013 to 40x in 2018)
Increasing database size
Changing
DBMS developers in turn are changing how they implement their systems' architectures
of 22
免费下载
【版权声明】本文为墨天轮用户原创内容,转载时必须标注文档的来源(墨天轮),文档链接,文档作者等基本信息,否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容,欢迎发送邮件至:contact@modb.pro进行举报,并提供相关证据,一经查实,墨天轮将立刻删除相关内容。