Single TiKV Node CPU Usage is Very High

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

Original topic: 单个tikv节点cpu很高

| username: Jolyne

[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version]
[Reproduction Path] What operations were performed that caused the issue
[Encountered Issue: Problem Phenomenon and Impact]
There are a total of 3 nodes. Currently, data is being read through interface polling, but the CPU usage of a single TiKV node is very high while the other 2 nodes are normal. Logically, since the data is distributed across different nodes, the CPU usage should not be high on just one node.

[Resource Configuration] Go to TiDB Dashboard - Cluster Info - Hosts and take a screenshot of this page
[Attachments: Screenshots/Logs/Monitoring]

| username: hey-hoho | Original post link

Typical read hotspot issue, reference handling methods:

| username: zhanggame1 | Original post link

Try enabling follower read to see if it alleviates the issue.

| username: Jolyne | Original post link

I have already split it, but it seems that split only applies to newly written data. Mine is fixed data :sweat_smile:

| username: Jolyne | Original post link

I’ll give it a try.

| username: Jolyne | Original post link

It doesn’t seem to be very useful.

| username: 像风一样的男子 | Original post link

This is the issue I encountered before with the single KV CPU increase. Take a look:

| username: Jolyne | Original post link

Okay, after some optimization, it has indeed come down.

| username: Fly-bird | Original post link

Slow query