Observability for LlamaIndex with Splunk

Step 1

Install and initialize the SDK in your code. It will automatically detect the workflow structure in build traces.

import asyncio

import chromadb
import os

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, StorageContext
from llama_index.vector_stores.chroma import ChromaVectorStore
from llama_index.embeddings.huggingface import HuggingFaceEmbedding
from traceloop.sdk import Traceloop

os.environ["TOKENIZERS_PARALLELISM"] = "false"

Traceloop.init(app_name="llama_index_example")

chroma_client = chromadb.EphemeralClient()
chroma_collection = chroma_client.create_collection("quickstart")

# define embedding function
embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-base-en-v1.5")

# load documents
documents = SimpleDirectoryReader("./data/paul_graham/").load_data()

# set up ChromaVectorStore and load in data
vector_store = ChromaVectorStore(chroma_collection=chroma_collection)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = VectorStoreIndex.from_documents(
    documents, storage_context=storage_context, embed_model=embed_model
)


async def main():
    # Query Data
    query_engine = index.as_query_engine()
    response = await query_engine.aquery("What did the author do growing up?")

    print(response)


if __name__ == "__main__":
    asyncio.run(main())

import * as traceloop from "@traceloop/node-server-sdk";
import {
  OpenAIEmbedding,
  SimpleDirectoryReader,
  SimpleVectorStore,
  VectorStoreIndex,
  serviceContextFromDefaults,
  storageContextFromDefaults,
} from "llamaindex";

traceloop.initialize({
  appName: "sample_llamaindex",
  apiKey: process.env.TRACELOOP_API_KEY,
  disableBatch: true,
});


const documents = await new SimpleDirectoryReader().loadData({
  // Download from:
  // https://github.com/traceloop/openllmetry-js/tree/main/packages/sample-app/data/paul_graham
  directoryPath: "data/paul_graham", 
});

const serviceContext = serviceContextFromDefaults({ embedModel });
const storageContext = await storageContextFromDefaults({ vectorStore });

const index = await VectorStoreIndex.fromDocuments(documents, {
  storageContext,
  serviceContext,
});

const queryEngine = index.asQueryEngine();

const res = await queryEngine.query("What did the author do growing up?");
console.log(res)

Step 2

Configure your collector to collect traces and send them to Splunk Observability Cloud; point OpenLLMetry to send traces to your Splunk collector

# In your Splunk collector configuration
receivers:
  otlp:
    protocols:
      grpc:
        endpoint: "0.0.0.0:4317"
      http:
        endpoint: "0.0.0.0:4318"
exporters:
  # Traces
  sapm:
    access_token: "${SPLUNK_ACCESS_TOKEN}"
    endpoint: "https://ingest.${SPLUNK_REALM}.signalfx.com/v2/trace"
    sending_queue:
      num_consumers: 32
 pipelines:
  traces:
    receivers: [jaeger, otlp, sapm, zipkin]
    processors:
    - memory_limiter
    - batch
    #- resource/add_environment
    exporters: [sapm]
        
# In the environment of your application
TRACELOOP_BASE_URL=http://splunk-otel-collector:4318

Docs

Observability for LlamaIndex with Splunk

Step 1

Step 2

Discover use cases

Trace a LlamaIndex-based RAG pipeline

Trace a LlamaIndex-based Agent

More things you can connect to {platform}

Bedrock

OpenAI

Watsonx

together.ai

Anthropic

Chroma

Mistral AI

Cohere

Gemini

LangChain

Ollama