I’m working with Apache IoTDB version 0.12.2 and have encountered some issues with the out-of-order data files. I understand that IoTDB can merge these out-of-order data files into ordered files to optimize the writing process and improve query performance.
However, despite going through the official documentation and various resources, I couldn’t find a clear explanation of the specific conditions or configurations that trigger this merge process. Here are a few specific questions I have:
- What are the exact conditions or thresholds (e.g., file size, number of out-of-order files) that need to be met for the merge process to be initiated?
- Are there any configuration parameters in the
iotdb-engine.properties
file that influence or control this behavior? - Can manual intervention be used to trigger the merge process, or is it solely automated?
Here is an example of my current configuration in the iotdb-engine.properties
file related to merges:
# tsfile management
merge_interval_seconds=3600
merge_chunk_subthread_num=4
merge_fileSelection_time_budget=60000
merge_memory_budget=21474836480
Any insights or pointers to relevant sections in the documentation would be greatly appreciated.
Alice Mengyuan Wang is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.