YMatrix
Quick Start
Connecting
Benchmarks
Deployment
Data Usage
Manage Clusters
Upgrade
Global Maintenance
Expansion
Monitoring
Security
Best Practice
Technical Principles
Data Type
Storage Engine
Execution Engine
Streaming Engine(Domino)
MARS3 Index
Extension
Advanced Features
Advanced Query
Federal Query
Grafana
Backup and Restore
Disaster Recovery
Guide
Performance Tuning
Troubleshooting
Tools
Configuration Parameters
SQL Reference
Deeply optimized for time-series workloads in electric vehicles, IoT, and smart factories. In customer production environments, YMatrix stably ingests 20 billion rows of data per day. It supports advanced SQL features such as CTEs and window functions, along with native time-series-specific functions. High compression ratios and support for tiering cold data to object storage significantly reduce storage costs.
Handles data volumes from terabytes to petabytes, delivering reliable and high-performance data processing and serving capabilities for enterprise reporting, BI, and other analytical applications. The proprietary Domino stream processing engine enables real-time data processing using standard SQL—replacing frameworks like Flink or Spark.
Provides full ACID compliance, enhanced by multiple patented technologies to deliver high performance. Meets the stringent requirements of critical systems such as financial and ERP applications for data correctness, consistency, and performance—while also offering strong analytical capabilities. Supports geographically distributed disaster recovery to ensure data safety under extreme conditions.
Enables large language models (LLMs) with vector search capabilities, helping enterprises rapidly build AI agents based on their business data. Eliminates the need for Spark: machine learning workflows can run directly inside the database using PL/Python, maximizing hardware utilization and improving ML efficiency.
In recent years, Farasis Energy—a leading listed manufacturer of new energy vehicle batteries—has steadily increased its production volume. The company faced growing complexity in production line management and urgently needed to enhance its in-house data capabilities to achieve "data-driven operational efficiency."
In Phase 1, YMatrix replaced Greenplum in the battery traceability system, resolving issues where complex reports previously failed to return results. Simple, frequently used reports saw query performance improve by several times. In Phase 2, YMatrix consolidated data from production lines using disparate systems into a unified data platform for upper-layer reporting and BI. Compared to the legacy Oracle system, YMatrix eliminated failures in exporting large time-range datasets and running complex reports. Query response times are now consistently in the seconds to milliseconds range.
In 2021, Li Auto adopted YMatrix to replace its original OpenTSDB-based vehicle telematics platform, enabling storage, querying, and computation of real-time vehicle driving metrics to support use cases such as vehicle analytics, fault tracing, and the owner mobile app.
YMatrix’s leading time-series performance not only resolved Li Auto’s issue of data ingestion backlog exceeding 3 hours during traffic peaks but also significantly reduced hardware costs—saving over one million RMB in the first year alone. As Li Auto’s vehicle sales have grown, the YMatrix cluster supporting its telematics platform has scaled accordingly. It now manages over 1 PB of historical data and supports telematics ingestion and querying for more than 1.5 million vehicles.
Shortly after OpenAI’s initial release, a leading domestic ERP vendor began exploring how to apply AI in enterprise services. While recognizing the market potential, the vendor sought to minimize R&D and maintenance costs, accelerate time-to-market, and deeply integrate AI capabilities with existing data platforms to maximize value.
As a key technology partner, YMatrix already provided core data services for the vendor’s main products. Leveraging YMatrix’s native vector data capabilities, the vendor launched its AI product within months—without introducing a new database. After extensive internal testing and refinement, the solution has now been deployed with multiple paying customers.
For more case studies, visit YMatrix Success Stories.
