How to load history data to IoTDB without impacting query history(历史数据导入会导致IoTDB查询变慢)
Open
- Dominant language
- Java
- Stars
- 6.4k
- Forks
- 1.2k
- Avg merge
- 1d 23h
- Merged PRs (30d)
- 115
Description
当进行大量的历史数据导入时(TB级),IoTDB会出现明显的查询变慢问题,同时CPU的使用率明显上升。以目前我们的使用场景为例,导入80000个测点两年以上的历史数据,数据量大概在7-8TB。导入前,IoTDB中测点最新值查询耗时在10ms以下,在导入历史数据的过程中发现:同样的最新值查询竟花费了9秒,对应用的数据查询体验造成了较大影响。希望IoTDB团队优化历史数据导入处理流程,降低对查询的影响。此外,希望能给出历史数据导入过程的制约因素,指导用户进行数据库参数配置或者硬件资源的升级。
Contributor guide
Research direction
The issue names no files, tests, or entry points. Start by reproducing the reported TB-scale historical import alongside latest-value queries, then determine the resource constraints and configuration factors that cause query latency to rise; done means reduced query impact and documented import limits or tuning guidance.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java
- Domain
- databases, performance
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100