Vector search index initialization very slow

[RodrigoE]

Hello,

I am creating a vector search index and selected Compute embeddings for a delta table with 19M records.  Delta table has only two  columns: ID (selected as index) and Name (selected for embedding). Embedding model is databricks-gte-large-en.

Index initialization estimated time to complete is 15 days!

I'm looking for advice to speed up initialization.

Thank you,

Rodrigo Escamilla

[iyashk-DB]

Hi Rodrigo,

The issue that you are seeing is because these embeddings are computed on the Databricks-GTE-Large-EN endpoint, which is a Pay-Per-Token endpoint. These have very high latency when used. So if speed is a concern, we suggest you use the models present in system.ai schema and create a Provisioned Throughput endpoint with a larger number of Model Units to have higher throughput and faster computations of embeddings. Then use that endpiont for computing the embeddings.

[RodrigoE]

Thank you very much for the fast response.  I am new to databricks (and vector search).  How do I go about "use the models present in system.ai schema and create a Provisioned Throughput endpoint with a larger number of Model Units"

Thank you,

Rodrigo Escamilla

[iyashk-DB]

@RodrigoE please follow this document - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/deploy-prov-throughput-fou...