Quick Start

Data Lake Import

Use the SessionManager to project data from your data lake into a Neptune Analytics graph, run algorithms, and export results.

1. Install nx-neptune

   pip install nx_neptune

2. Set environment variables

   export AWS_REGION=us-west-2
   export NETWORKX_S3_IMPORT_BUCKET_PATH=s3://my-bucket/import/
   export NETWORKX_S3_EXPORT_BUCKET_PATH=s3://my-bucket/export/
   export NETWORKX_STAGING_BUCKET=s3://my-bucket/staging

3. Project data and run algorithms

   from nx_neptune.session_manager import SessionManager
   
   session = SessionManager.session("my-session")
   graph = await session.get_or_create_graph(config={"provisionedMemory": 32})
   
   NODES = """
   SELECT DISTINCT "~id", 'customer' AS "~label"
   FROM (
       SELECT "nameOrig" as "~id" FROM transactions
       UNION ALL
       SELECT "nameDest" as "~id" FROM transactions
   );"""
   
   EDGES = """
   SELECT "nameOrig" as "~from", "nameDest" as "~to", "type" AS "~label",
          "amount" AS "amount:Float"
   FROM transactions;"""
   
   await session.import_from_table(graph, s3_import_path, [NODES, EDGES],
                                   catalog=catalog, database=database)
   
   # Run Louvain community detection
   result = graph.execute_query(
       'CALL neptune.algo.louvain.mutate({writeProperty:"community"}) '
       'YIELD success RETURN success'
   )

Note

Data lake integration requires an S3 bucket for staging import/export data. The CloudFormation deployment creates this automatically.

NetworkX Backend

Use nx-neptune as a drop-in NetworkX backend to run algorithms on Neptune Analytics.

1. Install nx-neptune

   pip install nx_neptune

2. Set your Neptune Analytics graph ID

   export NETWORKX_GRAPH_ID=your-neptune-analytics-graph-id

3. Run a graph algorithm

   import networkx as nx
   
   G = nx.Graph()
   G.add_node("Bill")
   G.add_node("John")
   G.add_edge("Bill", "John")
   
   r = nx.pagerank(G, backend="neptune")
   print(r)

4. Execute

   python nx_example.py

Tip

You can also pass the graph ID inline:

NETWORKX_GRAPH_ID=your-graph-id python nx_example.py

Running tests

Unit tests:

make test

Integration tests (requires a live Neptune Analytics instance):

export NETWORKX_GRAPH_ID=g-test12345
make integ-test

Next steps

• Session Manager — project data lake sources into a graph
• NetworkX Backend — offload algorithms to Neptune Analytics
• Deployment — deploy with CloudFormation