Why KaiwuDB Lite?
Lightweight and Efficient
- Ultra-low resource consumption: Runs smoothly on a single CPU core with minimal memory usage — ideal for resource-constrained edge devices and embedded systems.
- Million-level throughput: A proprietary storage engine and query optimizer deliver millions of data points written per second and millisecond-level queries over tens of millions of data points, enabling real-time edge analytics.
- Adaptive compressed storage: Built-in real-time compression algorithms achieve 5×–30× storage reduction, substantially lowering hardware costs and extending device lifecycles.
- Flexible deployment: Supports server, serverless, and embedded modes, spanning use cases from smart hardware to industrial gateways.
Secure and Reliable
- Write-Ahead Logging (WAL): Every data modification is first persisted to the WAL log, ensuring that committed data can be fully recovered by replaying the log after a crash or sudden power loss.
- Checkpoint optimization: Dirty pages in memory are periodically flushed to data files in batches, shortening log replay during recovery and enabling second-level failover.
- Full ACID compliance: Fully supports ACID transactions, guaranteeing atomicity, consistency, isolation, and durability across concurrent operations.
Cloud-Edge Collaboration
- Edge-side real-time computing: Device data is processed locally, cutting response latency to milliseconds; sensitive data remains on-premise to meet data-security and compliance requirements.
- Seamless cloud synchronization: Changes are detected in real time and synced on demand to the cloud-side KaiwuDB instance, forming a closed loop of "edge sensing" to "cloud decision-making."
Full-Stack Compatibility
- Mainstream ecosystem alignment: Compliant with the technical and regulatory requirements of industrial internet, energy, power, and other regulated sectors.
- Zero-ETL migration: Reads SQLite data files directly — no ETL pipeline or data migration required — dramatically lowering migration effort and cost.
- Developer-friendly interfaces: Adheres to standard SQL and provides JDBC/ODBC drivers, with bindings for mainstream languages including Java, C, and Python, minimizing learning curves and integration effort.
Core Features
Lightweight Embedded Architecture
Specifically designed for edge IoT and embedded scenarios, KaiwuDB Lite offers flexible deployment options. It can run as a standalone database, providing standard services via the PostgreSQL protocol; alternatively, it can be embedded into applications, too. KaiwuDB Lite can offer plug‑and‑play services for edge devices such as Raspberry Pi, enhancing data processing efficiency and response speed.

High‑performance Data Processing
- KaiwuDB Lite was specifically optimized for data read/write performance during its design, delivering exceptional efficiency in both data writing and querying. It supports standard SQL writes as well as batch importing, and can write millions of data in seconds.
- It has a complete built‑in function ecosystem to meet diverse data query and analysis requirements.
- By adopting a column-based storage structure, it only reads data on target column, significantly reducing disk I/O and response latency.
- It has a built‑in intelligent query optimization mechanism—including parallel execution, dynamic plan adjustment, adaptive JOIN, multi‑level caching, and runtime statistics collection—effectively reduces data scanning volume and further improves computational efficiency.

Diversified Compressed Storage Methods
KaiwuDB Lite adopts multiple technical approaches to achieve cost-effective storage.
- The column-based storage compression technique intelligently selects the optimal compression algorithm for different data types, achieving a 5–30× data compression ratio.
- It supports online real‑time compression, which means compression is performed simultaneously during data writing to balance storage efficiency and writing performance.
- It supports table‑level data lifecycle configuration, enabling automatic cleaning of historical data beyond a specified retention period, effectively control disk usage and further reduce storage and O&M costs.

Cloud‑edge-end Collaboration
KaiwuDB Lite can work with Enterprise KaiwuDB to build a comprehensive edge-to-cloud collaboration system. For key scenarios such as energy production and industrial manufacturing, KaiwuDB Lite is deployed on the edge side and end side, enabling real-time collection, storage, and millisecond-level data query of sensors and devices data. KaiwuDB, as the central database deployed on cloud, supports aggregation and integrated analysis of massive time-series data from various edge and end nodes, as well as relational data such as personnel statistic, device profiles, and business information. Taking the energy industry as an example, this solution delivers precise decision support for peak shaving and frequency regulation, operational monitoring and early warning, as well as other actionable business insights for enterprises.By integrating KaiwuDB Lite(high-performance end-side TSDB) with NeuronEX(an industrial edge gateway product of EMQ), this solution offers data capabilities including real-time data collection, processing, local storage and cloud‑edge-end collaboration, and thus forming a complete, efficient, low‑latency, and secure data foundation for industrial digitalization.

Open Ecosystem
- KaiwuDB Lite is compatible with mainstream programming languages including C/C++, C#, Java and Python, allowing developers to choose accordingly.
- It is also compatible with mainstream operating systems including Ubuntu, Kylin, and Windows, ensuring stable operation and multi‑environment deployment.
- It can be integrated with third‑party tools such as EMQX and Telegraf, enabling efficient data reading/writing with simple configuration.

KaiwuDB Lite & EMQX Joint Solution
Built on KaiwuDB and its data service platform, KaiwuDB Lite Plus is an all-in-one data solution designed for industrial edge-computing scenarios.