请问为啥mix coordinator的networking data receiving rate长期保持在650mb/s? #38462
Unanswered
xiaobingxia-at
asked this question in
Q&A and General discussion
Replies: 2 comments 7 replies
-
根据监控,是query node在发数据给coordinator。请问这是在发啥数据。 |
Beta Was this translation helpful? Give feedback.
1 reply
-
不是那么确定,几个猜测:
可以联系一下bigsheeper, 一起profiling一下。我们目前已经解决了大部分多表场景下的故障恢复的速度,待发布。 |
Beta Was this translation helpful? Give feedback.
6 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
我建立了一个cluster (2.4.17), 3 data nodes (15c60G), 16 indexing nodes (30c, 54G), 16 query nodes (13c, 114GB + local SSD). coordinator node (30c54G)
建立了100个collection,每个collection有10个partition(用了partition key),总共导入了3-billion个1536-d vector,平均分布在每个collection, 和每个partition中。每个collection大概有30 million个vector。
没有任何数据导入,没有任何搜索进行。mix coordinator的CPU使用率一直保持在22%的“高位”,内存使用率在4G左右。但是诡异的是,networking data receiving rate长期保持在650mb/s。请问没有导入,没有搜索,coordinator是在读啥数据呢?
此时,因为各种原因,所有节点被rotate了。我要release/load这100个collection,无比的慢,大概20分钟才能完成一个collection的release和reload,我感觉是coordinator被瓶颈了?
请问为啥在这种情况下,coordinator会这么忙?coordinator是release load collection的瓶颈吗?有没有设置可以解决这个问题啊?
Beta Was this translation helpful? Give feedback.
All reactions