How to Accelerate Migration Speed After Expanding TiKV Nodes

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

Original topic: tikv扩容节点后,如何加速迁移速度

| username: lxzkenney

tidb 5.0.3

Added one machine and deployed two TiKV nodes. I found that the balancing is a bit slow. From the 6th to the 9th, only one node has synchronized over 500GB. How can I speed up the balancing process? I adjusted several parameters but found no significant effect. CPU, IO, and network card usage are not high either.
Here is the latest configuration:
set config pd schedule.leader-schedule-limit= 8;
set config pd schedule.merge-schedule-limit= 16;
set config pd schedule.max-pending-peer-count= 32;
set config pd schedule.replica-schedule-limit= 100;
set config pd schedule.max-pending-peer-count = 128;
set config pd schedule.max-snapshot-count = 16;

Previously, the configuration was 1/2 or 1/4 of this.

| username: tidb菜鸟一只 | Original post link

Try adjusting these parameters: PD Control 使用说明 | PingCAP 文档中心

| username: yulei7633 | Original post link

I’ll give it a try too.

| username: WalterWj | Original post link

The main adjustment is the store limit parameter.

| username: system | Original post link

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