High Disk Utilization When TiDB and DM Are Deployed on the Same Machine

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

Original topic: tidb和dm部署在同一台机器上磁盘的util值高

| username: love-cat

【TiDB Usage Environment】Production Environment
【TiDB Version】5.2.2
【Reproduction Path】What operations were performed when the issue occurred
Stopping some components with tiup dm stop dm1 -R prometheus will cause a drop, but it will rise again after a while.
【Encountered Issue: Problem Phenomenon and Impact】
We have some nodes with limited resources, so we deployed TiDB + DM on a single virtual machine. Now we find that even when no tasks are running, the disk IO util value is close to 100%.
【Resource Configuration】
Virtual Machine Configuration: 8 cores + 32GB memory
The IO util value is high but there is no significant read/write activity.

| username: h5n1 | Original post link

What type of disk is it? Check the disk performance.

| username: dba-kit | Original post link

You can refer to the explanations in this article for the output metrics of iostats: 深入理解iostat

The util metric does not truly represent “whether the IO bottleneck is reached.” It indicates how long there was IO activity over a period of time. If DM and TiDB are deployed together, especially when all components (tidb-server, tikv, pd, dm-master, dm-worker) are deployed together, it could be that frequent IO is caused by log or region state updates, making the util appear high. This does not necessarily mean that the disk bottleneck has been reached; specific analysis is required.

| username: love-cat | Original post link

Okay, thank you.

| username: system | Original post link

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