Milvus | Respan Docs

Milvus is an open-source vector database built for scalable similarity search. It supports billion-scale vector data, multiple index types (IVF, HNSW, DiskANN), and hybrid search combining vector similarity with scalar filtering. Respan gives you full observability over every collection operation, search, and insert.

Set up Respan

Create an account at platform.respan.ai and grab an API key.

Run npx @respan/cli setup to set up with your coding agent.

Example projects

Python examples

Tracing

Setup

Install packages

$ pip install respan-ai opentelemetry-instrumentation-milvus pymilvus

Set environment variables

$ export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"

RESPAN_API_KEY is used to export traces to Respan.

Initialize and run

1 from pymilvus import MilvusClient
2 from respan import Respan
3 from opentelemetry.instrumentation.milvus import MilvusInstrumentor
4 
5 respan = Respan(instrumentations=[MilvusInstrumentor()])
6 
7 client = MilvusClient(uri="http://localhost:19530")
8 
9 client.create_collection(collection_name="documents", dimension=4)
10 
11 data = [
12     {"id": 1, "vector": [0.1, 0.2, 0.3, 0.4], "text": "AI observability with Respan"},
13     {"id": 2, "vector": [0.5, 0.6, 0.7, 0.8], "text": "Vector similarity search"},
14     {"id": 3, "vector": [0.2, 0.3, 0.4, 0.5], "text": "Scalable vector indexing"},
15 ]
16 client.insert(collection_name="documents", data=data)
17 
18 results = client.search(
19     collection_name="documents",
20     data=[[0.1, 0.2, 0.3, 0.4]],
21     limit=2,
22     output_fields=["text"],
23 )
24 for hits in results:
25     for hit in hits:
26         print(f"{hit['entity']['text']} (distance: {hit['distance']:.4f})")
27 
28 respan.flush()

View your trace

Open the Traces page to see your Milvus operation spans.

Configuration

Parameter	Type	Default	Description
`api_key`	`str \| None`	`None`	Falls back to `RESPAN_API_KEY` env var.
`base_url`	`str \| None`	`None`	Falls back to `RESPAN_BASE_URL` env var.
`instrumentations`	`list`	`[]`	Plugin instrumentations to activate (e.g. `MilvusInstrumentor()`).
`customer_identifier`	`str \| None`	`None`	Default customer identifier for all spans.
`metadata`	`dict \| None`	`None`	Default metadata attached to all spans.
`environment`	`str \| None`	`None`	Environment tag (e.g. `"production"`).

Attributes

In Respan()

1 from respan import Respan
2 from opentelemetry.instrumentation.milvus import MilvusInstrumentor
3 
4 respan = Respan(
5     instrumentations=[MilvusInstrumentor()],
6     customer_identifier="user_123",
7     metadata={"service": "vector-api", "version": "1.0.0"},
8 )

With propagate_attributes

1 from respan import Respan, propagate_attributes
2 from opentelemetry.instrumentation.milvus import MilvusInstrumentor
3 
4 respan = Respan(instrumentations=[MilvusInstrumentor()])
5 
6 def search(user_id: str, query_vector):
7     with propagate_attributes(
8         customer_identifier=user_id,
9         thread_identifier="conv_abc_123",
10         metadata={"plan": "pro"},
11     ):
12         return client.search(collection_name="documents", data=[query_vector], limit=2)

Attribute	Type	Description
`customer_identifier`	`str`	Identifies the end user in Respan analytics.
`thread_identifier`	`str`	Groups related messages into a conversation.
`metadata`	`dict`	Custom key-value pairs. Merged with default metadata.