Technology Partner

Query VeloDB as a graph — without moving a single row.

VeloDB is great at flat, aggregation-heavy analytics — dashboards, reporting, time-series. But some of the highest-value questions are about relationships: who is connected to whom, what depends on what, which chain of events led here. PuppyGraph plugs into VeloDB over JDBC, maps your existing tables into a graph, and lets you run Cypher or Gremlin against the same warehouse — no ETL, no separate graph database.

Talk to us →Visit PuppyGraph ↗

// About PuppyGraph

What PuppyGraph does

PuppyGraph is a graph analytics engine that connects directly to existing data sources. It reads metadata and rows from VeloDB through the MySQL-compatible protocol (default port 9030, MySQL 8 JDBC driver), maps tables and views into graph vertices and edges, and serves Cypher and Gremlin queries. Because the graph is a virtual view over VeloDB, every change in the warehouse is immediately reflected in graph results — no batch ETL, no replication lag.

CategoryGraph analytics engine

ConnectionJDBC · MySQL protocol (port 9030)

LanguagesCypher · Gremlin

Data modelVertices and edges over VeloDB tables/views

// Better Together

Why VeloDB + PuppyGraph

Zero ETL, single source of truth

Your VeloDB warehouse stays the source of truth. PuppyGraph maps tables and views into graph vertices and edges directly — no copies, no pipelines, no drift.

Multi-hop queries that scale

SQL JOINs degrade exponentially as you stack hops. Cypher and Gremlin in PuppyGraph express path-finding, pattern-matching, and recursive traversal natively — and run them against your existing VeloDB data.

Always-fresh graph

Because PuppyGraph queries VeloDB live, every insert and update is reflected in graph results immediately. No nightly rebuilds, no replication lag — critical for fraud, alerting, and security workloads.

One stack instead of two

No separate graph database to scale, monitor, secure, and pay for. Run graph queries on the same warehouse you already operate, with the same access controls.

// Use Cases

What you can build

Fraud and risk graphs

Trace transactions, accounts, and devices across many hops to surface rings, mules, and unusual patterns — on warehouse data that is seconds old.

Identity and entity resolution

Link customer, device, and account records into a unified entity view on top of your VeloDB tables, queryable with Cypher.

Knowledge graphs for AI

Give agents and RAG pipelines a graph-shaped view of business data, queryable in real time alongside the SQL surface they already use.

Use cases that benefit from graph search over VeloDB — Examples of relationship-heavy questions that benefit from graph traversal on top of VeloDB. Source: VeloDB blog.

// Reference Architecture

How the data flows

PuppyGraph over VeloDB integration workflow: connect, model, query — Three-phase workflow: connect PuppyGraph to VeloDB via JDBC, model graph vertices and edges over existing tables/views, then query with Cypher or Gremlin.

Ready to get started?

Tell us about your stack and use case. Our team will help you stand up VeloDB with PuppyGraph for your environment.