In v0.12.0, the engine filters out records where age <= 25 immediately during the scan, rather than pulling all persons into memory first.
UNWIND [1, 2, 3] AS x RETURN x * 2 AS result;
and LlamaIndex for immediate ingestion of agentic memory and structured retrieval chains.
: Construct and query large knowledge graphs to improve data integration, query performance, and insight generation. kuzu v0 120 better
| What’s New | Why It Matters | |------------|----------------| | (up to 3× faster on typical workloads) | Faster analytics, lower latency for real‑time apps | | Native CSV/Parquet import (no external ETL needed) | One‑click data onboarding | | Hybrid storage layer (in‑memory + on‑disk) | Bigger graphs, smaller memory footprints | | Cypher 1.2 compliance + new MATCH … WHERE optimizer | Easier migration from Neo4j & richer pattern matching | | Built‑in graph analytics library (PageRank, Betweenness, Community detection) | Do more inside the DB, fewer round‑trips | | Rust‑first client SDK (and refreshed Python/Go/JS bindings) | Safer, more idiomatic client code | | Transparent clustering & replication (beta) | Scale‑out without rewriting your app |
Kùzu is built for analytical (OLAP) graph workloads. In v0.12.0, its core query engine utilizes to process data in batches rather than row-by-row, which significantly reduces CPU overhead GitHub - kuzudb/kuzu.
This release marks a significant milestone in our journey to build the world's fastest and most embeddable property graph database management system. Over the past few months, our team has been hard at work optimizing the query engine, enhancing standard compliance, and smoothing out the developer experience. | What’s New | Why It Matters |
Kùzu is an designed for analytical workloads on large, highly connected datasets. Its architecture is built for speed and scalability through several modern design choices:
Kuzu’s underlying architecture is uniquely tailored to complex analytical workloads (OLAP). Version 0.12.0 further optimizes its execution engine to outpace traditional Graph Database Management Systems (GDBMSs) like Neo4j in multi-hop pathfinding. kuzudb/kuzu: Embedded property graph database ... - GitHub
As an embeddable database, Kuzu integrates directly into your application's process. It runs in-memory or on-disk right alongside your Python, Node.js, Rust, Go, Java, or even browser-based (WASM) applications. This native integration eliminates network latency and simplifies your architecture dramatically. Over the past few months, our team has
+-----------------------------------------------------------+ | Kùzu Core Optimization | +-----------------------------+-----------------------------+ | Vectorized Processing | Factorized Execution | | (Batched CPU operations) | (Compressed data paths) | +-----------------------------+-----------------------------+ | Worst-Case Optimal Joins | | (Cyclic query optimization) | +-----------------------------------------------------------+ Vectorized Execution Engine
Previously, managing database files could involve managing multiple files or directories. With v0.12.0, Kùzu supports a more streamlined structure, making it easier to transport, back up, or embed the entire graph database within an application.
Being embedded means it runs within your application process, eliminating the overhead of client-server communication. The improvements in 0.12.0 make this embedded experience even smoother.
Traditional graph databases were designed as standalone, client-server applications. While functional for Online Transaction Processing (OLTP), they incur significant network latency, serialization overhead, and suffer from poor scalability when running multi-hop, complex analytical queries (OLAP). Kùzu v0.12.0 bypasses these constraints by executing completely in-process.
: A new mechanism that reclaims disk space as you update the database, preventing excessive storage growth.
In v0.12.0, the engine filters out records where age <= 25 immediately during the scan, rather than pulling all persons into memory first.
UNWIND [1, 2, 3] AS x RETURN x * 2 AS result;
and LlamaIndex for immediate ingestion of agentic memory and structured retrieval chains.
: Construct and query large knowledge graphs to improve data integration, query performance, and insight generation.
| What’s New | Why It Matters | |------------|----------------| | (up to 3× faster on typical workloads) | Faster analytics, lower latency for real‑time apps | | Native CSV/Parquet import (no external ETL needed) | One‑click data onboarding | | Hybrid storage layer (in‑memory + on‑disk) | Bigger graphs, smaller memory footprints | | Cypher 1.2 compliance + new MATCH … WHERE optimizer | Easier migration from Neo4j & richer pattern matching | | Built‑in graph analytics library (PageRank, Betweenness, Community detection) | Do more inside the DB, fewer round‑trips | | Rust‑first client SDK (and refreshed Python/Go/JS bindings) | Safer, more idiomatic client code | | Transparent clustering & replication (beta) | Scale‑out without rewriting your app |
Kùzu is built for analytical (OLAP) graph workloads. In v0.12.0, its core query engine utilizes to process data in batches rather than row-by-row, which significantly reduces CPU overhead GitHub - kuzudb/kuzu.
This release marks a significant milestone in our journey to build the world's fastest and most embeddable property graph database management system. Over the past few months, our team has been hard at work optimizing the query engine, enhancing standard compliance, and smoothing out the developer experience.
Kùzu is an designed for analytical workloads on large, highly connected datasets. Its architecture is built for speed and scalability through several modern design choices:
Kuzu’s underlying architecture is uniquely tailored to complex analytical workloads (OLAP). Version 0.12.0 further optimizes its execution engine to outpace traditional Graph Database Management Systems (GDBMSs) like Neo4j in multi-hop pathfinding. kuzudb/kuzu: Embedded property graph database ... - GitHub
As an embeddable database, Kuzu integrates directly into your application's process. It runs in-memory or on-disk right alongside your Python, Node.js, Rust, Go, Java, or even browser-based (WASM) applications. This native integration eliminates network latency and simplifies your architecture dramatically.
+-----------------------------------------------------------+ | Kùzu Core Optimization | +-----------------------------+-----------------------------+ | Vectorized Processing | Factorized Execution | | (Batched CPU operations) | (Compressed data paths) | +-----------------------------+-----------------------------+ | Worst-Case Optimal Joins | | (Cyclic query optimization) | +-----------------------------------------------------------+ Vectorized Execution Engine
Previously, managing database files could involve managing multiple files or directories. With v0.12.0, Kùzu supports a more streamlined structure, making it easier to transport, back up, or embed the entire graph database within an application.
Being embedded means it runs within your application process, eliminating the overhead of client-server communication. The improvements in 0.12.0 make this embedded experience even smoother.
Traditional graph databases were designed as standalone, client-server applications. While functional for Online Transaction Processing (OLTP), they incur significant network latency, serialization overhead, and suffer from poor scalability when running multi-hop, complex analytical queries (OLAP). Kùzu v0.12.0 bypasses these constraints by executing completely in-process.
: A new mechanism that reclaims disk space as you update the database, preventing excessive storage growth.