Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.Original topic: Coprocessor 的缓存命中率是怎么计算出来的
|
username: Raymond
Dear teachers, could you please explain how the copr_cache_hit_ratio in explain analyze is calculated? I always see the result as either 0 or 1, but how is the 75% mentioned in the documentation calculated?
This article has a very detailed description.
More details:
Is it related to the number of nodes? Are you testing with multiple nodes over there?
The cache hit rate is the ratio of cop_task cache hits to the total number of cop_tasks, which is cop_task(cache hits)/total_cop_task. Whether the cache is hit depends on the region version information in TiKV (it is speculated that a new version is generated when the data in the region changes, and if the version does not change, it is a hit). After a hit, TiDB’s copr_cache uses the key generated by the cop_task request (copRequest) to find the value from the last request. Network overhead is unavoidable, but there is no need to rescan the data in TiKV.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.
|
username: 
|
username: 
|
username: 
|
username: