The Milvus destination in Fivetran enables automated ELT pipelines that load data into Milvus as a vector database, supporting AI and similarity search workloads.
Loading more......
Category: Data Integration & Migration
Brand: Fivetran
Type: Destination (Vector Database)
Milvus Destination for Fivetran lets you use Fivetran ELT pipelines to load data into Milvus or Zilliz Cloud as a vector database, supporting AI and similarity search workloads such as semantic search, RAG, and multi‑modal search.
Milvus and Zilliz Cloud support
Fivetran can write to self‑managed Milvus or Zillan Cloud clusters as the destination.
GenAI and vector search workloads
Data is loaded into Milvus as vectors suitable for high‑performance similarity search and AI applications (semantic search, RAG, multi‑modal use cases).
Only the following Fivetran connectors are supported as sources for this destination:
The documentation also references guidance on how to sync data from sources not directly supported by the Milvus destination, implying workflows or intermediate steps for unsupported sources.
Step‑by‑step setup guide
A dedicated Milvus setup guide walks through connecting a Milvus vector database (or Zilliz Cloud cluster) as a Fivetran destination.
Partner‑built destination
The Milvus destination is identified as partner‑built, with documentation and implementation maintained by the Milvus/Zilliz team.
Fivetran’s standard data types are mapped to Milvus data types as follows:
| Fivetran Data Type | Milvus Destination Data Type |
|---|---|
| BOOLEAN | BOOLEAN |
| SHORT | INT16 |
| INT | INT32 |
| LONG | INT64 |
| FLOAT | FLOAT |
| DOUBLE | DOUBLE |
| DECIMAL | VARCHAR |
| LOCALDATE | VARCHAR |
These mappings define how source schema fields are stored in Milvus when synced via Fivetran.