研究论文

Apache IoTDB 始于清华大学软件学院。IoTDB是一个用于管理大量时间序列数据的数据库，它采用了列式存储、数据编码、预计算和索引技术，具有类SQL的接口，可支持每秒每节点写入数百万数据点，可以秒级获得超过数万亿个数据点的查询结果。它还可以很容易地与 ApacheHadoop、MapReduce 和 ApacheSpark 集成以进行分析。

相关研究论文如下：

• PISA: An Index for Aggregating Big Time Series Data, Xiangdong Huang and Jianmin Wang and Raymond K. Wong and Jinrui Zhang and Chen Wang. CIKM 2016.
• Matching Consecutive Subpatterns over Streaming Time Series, Rong Kang and Chen Wang and Peng Wang and Yuting Ding and Jianmin Wang. APWeb/WAIM 2018.
• KV-match: A Subsequence Matching Approach Supporting Normalization and Time Warping, Jiaye Wu and Peng Wang and Chen Wang and Wei Wang and Jianmin Wang. ICDE 2019.
• The Design of Apache IoTDB distributed framework, Tianan Li, Jianmin Wang, Xiangdong Huang, Yi Xu, Dongfang Mao, Jun Yuan. NDBC 2019
• Dual-PISA: An index for aggregation operations on time series data, Jialin Qiao, Xiangdong Huang, Jianmin Wang, Raymond K Wong. IS 2020

Benchmark工具

我们还研发了面向时间序列数据库的Benchmark工具：

https://github.com/thulab/iotdb-benchmark