GBase 8c

Multi-Model Distributed Database

Introduction

Product Overview

GBase 8c is a multimodal and polymorphic third-generation intelligent distributed database independently developed by General Data Technology Co., Ltd.. It achieves database intelligence through intelligent optimization, intelligent operation and maintenance, and intelligent security, endowing GBase 8c with intelligent features such as high performance, high availability, elastic scalability, and high security.
It supports multiple storage modes including row storage, column storage, and in-memory storage, as well as various deployment architectures such as standalone, active-standby, and distributed modes, along with compatibility modes for ORACLE, PG, and MYSQL. Deployable on physical machines, virtual machines, containers, private clouds, and public clouds, GBase 8c provides secure, stable, and reliable data storage and management services for core systems of key industries, Internet business systems, and government-enterprise business systems, meeting diverse application scenarios.
GBase 8c Multi-Model Distributed Database Architecture (with Compute-Storage Decoupling, GTM, and Replica Design

Product Architecture

GBase 8c supports two deployment modes (active-standby and distributed) and three storage engines (row storage, column storage, and in-memory storage), meeting diverse business requirements of users.
Multiple Storage Modes
  1. GBase 8c integrates multiple storage engines to address scenario-specific business needs:
  2. Row Storage Engine: Optimized for OLTP (Online Transaction Processing) scenarios such as order management, shipping systems, and banking transaction systems.
  3. Column Storage Engine: Designed for OLAP (Online Analytical Processing) scenarios including data statistical reporting and analysis systems.
  4. In-Memory Engine: Tailored for extreme-performance scenarios like bank risk control systems.
Multiple Deployment Architectures
GBase 8c enables diverse deployment architectures through multi-tenancy, offering three options—standalone, active-standby, and distributed—all managed via a unified operation and maintenance (O&M) platform. These architectures target enterprise core transaction systems and future massive transactional scenarios, delivering differentiated competitiveness.
Standalone Architecture
GBase 8c supports standalone deployment directly on a single server. This architecture offers the advantages of minimal cost and simple implementation.
Active-Standby Architecture
GBase 8c supports one-primary-multiple-standby deployment, with synchronous or asynchronous replication between the primary and standby nodes. Featuring simple deployment and efficient delivery, it is suitable for scenarios with moderate data volumes, demand for extreme single-machine performance, and requirements for data backup.
Distributed Architecture
The distributed mode supports full-component redundant high availability and separation of storage and computing. It enables independent horizontal scaling of computing and storage capabilities based on business needs, making it ideal for scenarios with large data volumes, high concurrency, and stringent data security requirements.

Commercial Value

Full Independent Innovation Ecosystem
GBase 8c's independent research and development rate fully complies with national policy requirements. It supports a variety of independent innovation (Xinchuang) CPUs and servers based on these CPUs, as well as multiple independent innovation (Xinchuang) operating systems and middleware.
Meeting Cost Reduction Requirements
GBase 8c offers multiple deployment options to address diverse cost needs of users: Adopts general-purpose servers to reduce hardware costs; Features customer-friendly license fees to lower software costs; Supports on-demand configuration to reduce total cost of ownership (TCO); Enables cloud deployment and intelligent operation and maintenance (O&M) to minimize maintenance costs. It fully meets customers' various requirements for cost reduction.
Horizontal Scalability
GBase 8c's horizontal scalability enables customers to flexibly meet system expansion needs, effectively accommodating the growing data volume requirements of business operations.
Elastic Scalability
Scenarios such as flash sales and promotional events impose extreme performance requirements on systems. GBase 8c database is equipped with elastic scalability, enabling flexible scaling up and down to seamlessly meet these demands.
Continuous Availability
Core business systems have extremely high requirements for high availability. GBase 8c distributed database can achieve and even exceed the 99.999% high availability level through its multi-replica strong consistency capability.
Virtualization Deployment
With the continuous development of cloud computing and virtualization technologies, the cloudification of enterprise IT infrastructure has become an irreversible trend, and customers have an urgent demand for deploying distributed databases on the cloud. GBase 8c supports flexible deployment options, including physical machines, containers, virtual machines, private clouds, hybrid clouds, and public clouds, fully meeting customers' cloud migration needs.

Core Advantages

Multimodal & Polymorphic

Supports two deployment architectures (active-standby and distributed) and three storage engines (row storage, column storage, and in-memory storage), meeting diverse business and scenario requirements of users.

Extreme Performance

A single GBase 8c node can achieve a performance level of millions of TPM (Transactions Per Minute), and relying on distributed scalability, it provides users with cluster-level extreme performance capabilities.

Ultra-High Availability

  • Active-Standby Architecture: Offers multiple deployment modes including synchronous replication, asynchronous replication, and cascaded standbys.
  • Distributed Architecture: Features no single point of failure (SPOF) across the cluster, achieving 99.99% high availability. Supports 3DC (Three Data Centers) deployment across two locations and remote active-active mode, ensuring continuous business operation and zero data loss.

    • Elastic Scalability

    The distributed cluster supports online scaling (scale-up/scale-down) and separation of storage and computing.

    Ultra-High Security

    Provides comprehensive end-to-end data security protection with features such as fully homomorphic encryption, access control, encryption authentication, database auditing, and dynamic data masking.

    Distributed Transactions

    Supports strongly consistent distributed transactions with RC (Read Committed) and RR (Repeatable Read) isolation levels.

    Distributed Parallel Processing

  • Coordinators push down operators to data nodes for parallel processing.
  • Multi-threaded architecture enables parallel business processing.
  • Adopts MVCC (Multi-Version Concurrency Control) technology to achieve read-write concurrency.

    • Multi-Tenancy Capability

    GBase 8c supports resource isolation based on multi-tenancy. Resources are isolated between different tenants, and resource quotas can be configured for each tenant. Each tenant can freely choose deployment modes including standalone, active-standby, and distributed.

    Spatial Database Capabilities

    Equipped with spatial database functionality, GBase 8c provides GIS (Geographic Information System) with application-related geospatial data access and fast geographic information retrieval capabilities.

    High Compatibility

  • Supports standard SQL syntax.
  • Fully compatible with PostgreSQL syntax.
  • Compatible with common Oracle syntax.

    • Applicable Scenarios

    Core Businesses in Key Industries

    Core systems in key industries have stringent requirements for database performance, stability, and security. With the continuous growth of business volumes in these core systems, traditional centralized databases lack the capacity to support high concurrency and high throughput demands. Distributed databases have emerged as a powerful solution to this challenge.
    • GBase 8c boasts core capabilities including strong consistency in distributed transactions, and linear scalability of computing and storage capacities.
    • It supports multi-site/multi-center deployment and remote active-active mode, delivering 99.999% high availability.
    • Complemented by a visual intelligent operation and maintenance (O&M) platform, the database significantly reduces customers' O&M complexity.

    Internet Services

    With the rapid development of internet services, especially mobile internet services, internet applications have increasingly high performance requirements for databases under massive data volumes, such as concurrency and throughput. Moreover, to support extreme performance demands in specific scenarios like flash sales, auto-scaling has become an essential requirement for internet applications.
    • GBase 8c's elastic scalability enables online scaling up/down based on business needs, seamlessly meeting performance requirements in extreme business scenarios.
    • Its gray release feature provides robust support for the frequent deployment and decommissioning of internet services.

    Government and Enterprise Services

    With the deepening development of cloud computing and virtualization technologies, migrating OA systems in the financial industry and enterprise applications to the cloud has become a mainstream architectural choice.
    • GBase 8c supports multiple deployment modes including physical machines, containers, private clouds, and public clouds, offering flexible options for financial businesses moving to the cloud.
    • It supports data transmission encryption and storage encryption, providing strong guarantees for data security on the cloud.
    View More Cases →

    Product FAQ

    Explore GBase 8c's architecture, performance, deployment options, scalability, and high availability.

    Q1: What is GBase 8c, and how does it differ from traditional databases?

    GBase 8c is an AI-ready, multi-model database designed to support diverse data types, processing patterns, and deployment architectures. Its multi-model capabilities bring together row-oriented and column-oriented storage, hybrid transactional and analytical processing (HTAP), GIS, vector data, and time-series workloads within a unified database platform. Its flexible deployment options include primary-standby, distributed, and compute-storage disaggregated architectures.

    Rather than moving data between separate systems for transactions, analytics, and AI-related workloads, GBase 8c can support these use cases on a unified platform. Applications can work with the database without needing to manage the underlying data distribution directly.

    Q2: Is GBase 8c a distributed database? How much application modification is required?

    Yes. GBase 8c supports a native distributed architecture in which coordinators distribute query operations to data nodes for parallel execution. Its multithreaded architecture supports concurrent workloads, while Multi-Version Concurrency Control (MVCC) helps manage concurrent reads and writes. It also supports complex cross-node JOIN queries without requiring an external middleware layer for distributed query processing.

    GBase 8c is compatible with PostgreSQL syntax and offers compatibility with SQL dialects from Oracle, MySQL, and SQL Server. With the appropriate migration tools and planning, application changes can be minimized. Certain operations, such as cross-node DDL statements, may require adjustments to comply with deployment-specific execution rules.

    Q3: What level of performance can GBase 8c deliver?

    GBase 8c can deliver million-level TPMC performance on a single node, with distributed scaling providing additional capacity at the cluster level. In the HyBench HTAP benchmark, GBase 8c achieved an H-Score of 3810.36, demonstrating its performance in hybrid transactional and analytical workloads.

    Q4: What deployment options does GBase 8c support, and how many servers are required?

    GBase 8c supports several deployment models. A single-node deployment offers a simple, cost-effective starting point but does not provide high availability on its own. A primary-standby deployment uses one primary node and one or more standby nodes, with synchronous or asynchronous replication, making it suitable for workloads that prioritize single-node performance and data redundancy.

    A distributed deployment uses redundant components to support high availability and can separate compute from storage. The compute-storage disaggregated model uses shared object storage and stateless compute resources, enabling rapid scaling and scale-to-zero capabilities.

    For production distributed deployments, a typical minimum configuration starts with three nodes: one coordinator node (CN) and two data nodes (DNs). The final configuration should be based on the required redundancy, fault tolerance, and workload capacity.

    Q5: Does GBase 8c support online scaling without disrupting business operations?

    Yes. Distributed deployments support online expansion and reduction of data nodes, allowing cluster capacity to scale horizontally. Compute and storage resources can also be scaled independently where supported by the deployment architecture. Online scaling is designed to minimize data movement and reduce the impact on running workloads.

    In the compute-storage disaggregated deployment model, compute capacity can be adjusted in seconds and scaled down to zero when there is no workload. This flexibility can help accommodate temporary demand spikes, such as major sales events and flash-sale traffic, while optimizing resource utilization during quieter periods.

    Q6: What high-availability and disaster-recovery options does GBase 8c offer?

    For primary-standby deployments, GBase 8c supports a range of high-availability and disaster-recovery topologies, including single-site deployments, dual-site configurations within the same city, two-site three-center architectures, three-site five-center architectures, and multi-site active-active deployments.

    Distributed deployments also support multiple high-availability topologies, including single-site clusters, dual-center and three-center configurations within the same city, two-site three-center and two-site four-center architectures, and geographically distributed active-active deployments. The appropriate design depends on recovery objectives, network conditions, replication requirements, and application continuity needs.