TiDB READ Scheduling

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: TIDB READ调度

| username: SianoWang

[TiDB Usage Environment] Testing
[TiDB Version] 6.5.2
[Reproduction Path] Concurrent reading with 20 threads
[Encountered Problem: Phenomenon and Impact]
A single table with 200 million rows, indexed by employee number, reads about 2000 rows for 100 employees each time.
Under balanced leader and region conditions, query traffic is concentrated on a single TiKV instance, and the distribution of running tasks in the read pool is extremely uneven.
This causes the average wait duration of the coprocessor on that instance to exceed 10 seconds.
Could you please advise on how to rebalance the read traffic?
[Resource Configuration]


[Attachments: Screenshots/Logs/Monitoring]



| username: huhaifeng | Original post link

First, you can check the leader and region distribution of the table.
Additionally, you can look at the msg count monitoring in gRPC.

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.