OpenMLDB Provides a Full-Stack FeatureOps Solution for Production
SQL-Centric Development and Management
Feature extraction script development, deployment, and maintenance are all based on SQL with great ease of use.
The Unified Online-Offline Execution Engine
Offline and real-time online feature extraction use a unified execution engine, thus online-offline consistency is inherently guaranteed.
Customized Optimization for Feature Extraction
Offline feature extraction is performed based on a tailored Spark version that is particularly optimized for batch-based feature processing. Online feature extraction provides tens of milliseconds latency under high throughput pressure, which fully meets the online performance requirements.
OpenMLDB has been implementing important production features for large-scale applications, including fault recovery, high availability, seamless scale-out, smooth upgrade, monitoring, heterogeneous memory support, and so on.
OpenMLDB is Driving the AI Transformation for Enterprisesmore
Akulaku: Real-Time Feature Extraction for AI-Powered Risk Control
In financial technology scenarios, OpenMLDB not only doubles the team's human efficiency and saves millions of costs, but also is the only solution with linear scale compared with Spark and Flink.
OpenMLDB Helps Build an AI-Powered Anti-Fraud System in Banking Affairs
Helping the commercial bank leverage AI to build an anti-fraud system, driving the AI-based anti-fraud to achieve effectiveness and efficiency
Practice of Artificial Intelligence for IT Operations Based on OpenMLDB in a Financial Institution
Building the next-generation AIOps powered by OpenMLDB with resource consumption significantly reduced, scaling the business at low cost
OpenMLDB introduction video
Publication in VLDB 2021: Optimizing OpenMLDB Based on Persistent Memory
The paper titled "Optimizing In-memory Database Engine for AI Powered On-line Decision Augmentation Using Persistent Memory" was accepted as a regular research paper by the top database conference VLDB 2021. This paper describes the system design of OpenMLDB to efficiently support trillion-dimensional sparse feature online prediction. It further utilizes the persistent memory for fast recovery and cost saving.