# Brahmand #### Brahmand (Sanskrit for Universe) **A high-performance, stateless graph-analysis layer for ClickHouse.** > **Note:Brahmand is under active development and not yet production-ready. Some Cypher features are still missing.** --- ## Features - **ClickHouse-native** Extends ClickHouse with native graph modeling, merging OLAP speed with graph-analysis power. - **Stateless** Offloads all storage and query execution to ClickHouse—no extra datastore. - **Cypher-query** Industry-standard Cypher syntax for intuitive, expressive property-graph querying. - **Analytical-scale** Optimized for very large datasets and complex multi-hop traversals. --- ## Architecture Brahmand runs as a lightweight graph wrapper alongside ClickHouse: ![acrhitecture](./architecture.png) 1. **Client** sends a Cypher query to Brahmand. 2. **Brahmand** parses & plans the query, translates to ClickHouse SQL. 3. **ClickHouse** executes the SQL and returns results. 4. **Brahmand** sends results back to the client. --- ## Docs and Installation Check [Docs](https://www.brahmanddb.com/introduction/intro) here. ## Benchmark Preliminary informal tests on a MacBook Pro (M3 Pro, 18 GB RAM) running Brahmand in Docker against a ~12 million-node Stack Overflow dataset show multihop traversals running approximately 10× faster than Neo4j v2025.03. These early, unoptimized results are for reference only; a full benchmark report is coming soon. ## License Brahmand is licensed under the Apache License, Version 2.0. See the LICENSE file for details. ## Issues & Contributing Feel free to submit issues and enhancement requests. All contributions are welcomed.