Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.Original topic: 建表慢
|
username: 等一分钟
[TiDB Usage Environment] Production Environment
[TiDB Version] 6.1.5
[Reproduction Path]
|
username: 等一分钟
Sorry, I can’t translate the content from the image directly. Please provide the text you need translated.
The image you provided is not visible. Please provide the text you need translated.
|
username: My single-node k8s deployment of TiDB is faster than yours:
It feels like either the disk is too bad… or the network between nodes is not good.
|
username: Is your disk really bad? The read and write speeds are so slow.
The SSD performance of Alibaba Cloud is not very good. Tests have shown that it might be worse than the mechanical disks of physical machines. You need to use local disks.
It’s unlikely. IOPS also depends on disk space, and there are also several thousand.
|
username: Additionally, output:
trace format=‘row’ create table t1 (id int);
Sorry, I misread it 
Please provide the trace information above.
I just remembered a question. Previously, querying “select * from information_schema.tables” in the TiDB cluster was extremely slow. Could it be related to this?
That shouldn’t have much to do with it. The information_schema.tables query is an issue with the table design itself, and there are already issues in the repository being addressed for a fix.