How to search faster? #6204
-
大家好,我现在有10,000,000条256维的向量,然后我搜索一条向量的top10用时300ms,怎样才能更快呢?想达到一亿向量秒级,这个速度不太能满足,谢谢大家的指导! |
Beta Was this translation helpful? Give feedback.
Replies: 4 comments 3 replies
-
version: 0.5 storage: 这是server_config.yaml的一些信息 |
Beta Was this translation helpful? Give feedback.
-
Let do some testing. I start a milvus server(v1.1.1) on my machine, with the config:
The machine environment: The following is a script, the script do these things:
|
Beta Was this translation helpful? Give feedback.
-
On my machine, the result is:
For single thread search, each search() cost about 10 milliseconds. |
Beta Was this translation helpful? Give feedback.
-
The "nprobe" will affect search performance, if we set nprobe=128, the result will be:
|
Beta Was this translation helpful? Give feedback.
Let do some testing.
I did't know which kind of index you select, and what search parameters you choose.
I start a milvus server(v1.1.1) on my machine, with the config:
The machine environment:
Ubuntu18.04
Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz 12 cores
64GB memory
The following is a script, the script do these things: