# Other Parameters | Parameter | Default | Description | | :-------------------------------------- | :------ | :----------------------------------------------------------------------------------------- | | `patrol-region-interval` | 100 ms | Controls the execution frequency that replica checker checks the health status of regions. | | `tolerant-size-ratio` | 0 | Controls the size of the balance buffer area. | | `{region_weight}` and `{leader_weight}` | 1 | Manually control the corresponding weights by `pd-ctl store weight` command. | -------------------------------------------------------------------------------------------------------------------------------------------------- Other parameters - `patrol-region-interval` Control the interval of the sweep region. The default is 100ms, and usually does not need to be adjusted. If the number of regions is large and you encounter some scenario such as speeding up region merge, node failover replica speed, etc., you can reduce it to 10ms to speed up the patrol region, at the cost of consuming more CPU - `tolerant-size-ratio` Controls the balance region buffer size. After 4.0, the default is 0, which means automatic adjustment, and usually does not need to modify. If you upgrade from version 3.0 or earlier, it will inherit the previous default setting of 5. If you have a large number of regions, a smaller value may cause frequent unnecessary scheduling, such as 100 or change to 0 after the upgrade version. - `region_weight` and `leader_weight` After PD calculates the region and leader scores, it divides the weight to get the final region and leader scores. The weight parameter defaults to 1, and usually does not need to modify. If necessary, it can be adjusted through the pd-ctl > store weight command to achieve the purpose of manually controlling the scheduling direction