Query Taking Too much time

Support
1

I am using weaviate version [1.24.4].
I have a schema with properties mention below :

“properties”: [
{
“name”: “text”,
“dataType”: [“text”],
“description”: “Text chunks”,
“tokenization”: “word”,
},
{
“name”: “prop 1”,
“dataType”: [“number”],
“description”: “Prop 1”,
“moduleConfig”: { # Module-specific settings
“text2vec-transformers”: {
“skip”: True,
“vectorizePropertyName”: False,
}
},
},
{
“name”: “prop 2”,
“dataType”: [“number”],
“description”: “Prop 2”,
“moduleConfig”: { # Module-specific settings
“text2vec-transformers”: {
“skip”: True,
“vectorizePropertyName”: False,
}
},
},
{
“name”: “prop 3”,
“dataType”: [“number”],
“description”: “Prop 3”,
“moduleConfig”: { # Module-specific settings
“text2vec-transformers”: {
“skip”: True,
“vectorizePropertyName”: False,
}
},
},
{
“name”: “prop 4”,
“dataType”: [“number”],
“description”: “Prop 4”,
“moduleConfig”: { # Module-specific settings
“text2vec-transformers”: {
“skip”: True,
“vectorizePropertyName”: False,
}
},
},
{
“name”: “prop 5”,
“dataType”: [“number”],
“description”: “Prop 5”,
“moduleConfig”: { # Module-specific settings
“text2vec-transformers”: {
“skip”: True,
“vectorizePropertyName”: False,
}
},
},
]
}

I need to filter on Prop 5 and do a vector search on property “text”.

My total number of records are around 30 millions. I ma using default settings and machine infra is 128 CPU and 256 GB Memory.

My query is taking around 2-3 seconds.
Can you please help me with optimisation. Optimisation related to Schema and Weaviate Setup so that I can achieve the performance within few hundred mili seconds.

2

Hi @Yogesh_Sharma !

Please, when opening an issue in Support, please, answer all the questions we ask on the template, as this has proven to improve our troubleshooting.

What is the version of the client you are using? Is it multinode? Single node?

Can you share the query?

Also, do you see any outstanding logs?

THanks!