Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: 关于TiKV数据量与硬盘容量的关系
I deployed a simple cluster using three servers, with each machine running one TiKV. I encountered an issue where two 3TB hard drives and one 500GB hard drive are mounted on the three machines, respectively. It seems that the data allocated to the 3TB drives is about five to six times that of the 500GB drive. I would like to ask if TiDB automatically allocates data based on the capacity of the hard drives? If so, can this ratio be adjusted? Or is the difference in data volume due to the write performance of the hard drives?
PD schedules regions based on the score of the store. A score of 5.0 can adjust the parameter leader-schedule-policy, which is based on size or count. Additionally, the related parameter region-weight will affect the score, thereby influencing the distribution of regions.
What does the count here refer to?
Does “size” refer to the remaining disk capacity?
I roughly understand, size refers to the total size of keys in a region. But it seems that these two allocation methods don’t have a direct relationship with the proportional allocation phenomenon mentioned above 
The total size of regions in the store
To understand how the specific score is calculated, you need to dig into the source code.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.
|
username: Leox
|
username: 
|
username: