GBase Partners with Beijing Mobile to Build a Homegrown Core Lakehouse Big Data Platform
Facing the urgent need for massive data integration, real-time analytics, and cost control, China Mobile Beijing (Beijing Mobile) partnered with GBase to build a next-generation lakehouse big data platform based on GBase 8a MPP Cluster and the cloud data warehouse, achieving full replacement of foreign databases and the unified, efficient release of data value. This implementation has been selected as an authoritative "Xinghe" (Star River) Promising Case in the big data field, providing a homegrown benchmark for the upgrade and reconstruction of core data systems in large enterprises.
Breaking Through: The Four Core Challenges
Beijing Mobile's digital transformation journey faced clear and formidable challenges:
The Integration Dilemma: How can data scattered across more than ten different databases and big data components be unified so that business departments can perform global analysis as if using a single database?
The Performance and Elasticity Dilemma: How can fast response for real-time analytics and ad-hoc queries be achieved against massive data (petabyte scale)? Can the platform scale elastically like "LEGO" blocks as business grows?
The Cost and Efficiency Dilemma: How can the status quo of redundant data storage across different clusters and rigid allocation of computing resources be changed, thereby reducing massive storage and O&M costs and improving data development efficiency?
The Sustainable Development Dilemma: How can the new platform remain technologically relevant for the next 5-10 years and seamlessly support future scenarios such as AI analytics and real-time data warehousing?
GBase Database: Delivering "Four-in-One" Breakthrough Capabilities
After rigorous selection and verification, Beijing Mobile ultimately chose GBase as its core engine. GBase 8a MPP Cluster and its cloud-native data warehouse product GCDW precisely address the above challenges with a combination of capabilities:
Capability 1: Build a "Logical Data Warehouse" for Seamless Data Integration
The "Logical Data Warehouse" capability offered by GBase is a key innovation. It acts as an intelligent data virtualization layer that logically integrates data scattered across Oracle, MySQL, Hadoop, and its own MPP clusters without physically moving the raw data. For business users, they only need to use standard SQL statements to transparently correlate and query all data, completely eliminating the cumbersome process of exporting data across systems and reconciling definitions, truly realizing "data visible everywhere, globally analyzable."
Capability 2: Distributed Engine + Storage-Compute Separation, Balancing High Performance and Extreme Elasticity
GBase 8a MPP Cluster itself is a powerful distributed analytics engine. Its parallel computing capabilities significantly improve complex query performance, efficiently taking over the core business from the original foreign data warehouse. More importantly, its cloud-native architecture supports storage-compute separation—computing and storage resources can be scaled independently. This means that during business peaks akin to "Singles' Day," computing capacity can be rapidly expanded to cope, while independently scaling storage capacity for continuously growing historical data. This elasticity optimizes resource utilization and cost control.
Capability 3: Unified Data Governance, Making Data Assets "Visible, Controllable, and Well-Used"
Based on the unified data foundation built with GBase, Beijing Mobile established an enterprise-level data governance system. Through unified metadata management, the lineage of all data tables is clear at a glance; built-in data quality rules automatically detect and alert on problematic data; and unified task scheduling and development standards significantly improve data development efficiency. This transforms data from a "cost burden" into a trusted and usable core asset.
Capability 4: Lakehouse Architecture, Laying a Future-Proof Technical Foundation
The lakehouse architecture supported by GBase database is a critical future-proof design. It can support high-performance, high-concurrency precise analytics (BI reports) like a data warehouse, while also inexpensively storing massive amounts of diverse data in raw format (such as logs and text) like a data lake, for use by AI models and exploratory analysis. This architecture ensures that the platform meets today's rigorous business analytics needs and has the potential to support cutting-edge scenarios such as machine learning and real-time stream processing.
The successful practice of Beijing Mobile and GBase database signifies that a converged data platform centered on a homegrown distributed database is fully capable of supporting carrier-grade critical business. It is not merely a successful technology substitution but a paradigm shift in the evolution of enterprise data infrastructure towards cloud-native and intelligent.
Looking ahead, this solid and flexible data foundation will continue to unleash its power: it will become the "brain" empowering precision marketing for thousands of faces, the "nerve" sensing network quality in real time, and the "soil" incubating innovative services such as smart cities and IoT. GBase database will also continue to deepen its technology and explore together with industry leaders, making data intelligence a powerful engine driving social development.