Meeting Core Cloud Database Needs: GBase 8a MPP V9 Delivers Multi-Tenancy

As enterprise IT evolves, data is growing at an unprecedented pace. The drive toward centralized data architectures has gained wide acceptance in finance, telecom, energy, and other industries, with many large organizations already investing in data centers. These users expect their big data platforms or data warehouses to support multi-tenancy, enabling branch offices to consume resources on demand. GBase 8a MPP V9 clusters meet this requirement perfectly through virtual cluster technology and resource management.

First, let's look at GBase 8a MPP V9 virtual cluster technology. A virtual cluster is a resource isolation mechanism. Logical sub-clusters can be planned, deployed, scaled, and resourced independently, yet all logical sub-clusters share a unified access entry, a unified metadata view, unified resource management, unified execution scheduling, and unified authentication and authorization. Most importantly, data can be easily shared between virtual clusters—making it simple for headquarters to distribute shared data to branch offices.

In a real-world scenario, multiple logical sub-clusters correspond to multiple branch offices. Each branch can independently deploy, start/stop, and scale its logical sub-cluster. For branch users, using a logical sub-cluster is no different from using a standalone cluster. For headquarters, the physical resources and data assets of all branches are centrally managed, centrally utilized, uniformly accessed, and uniformly operated—with on-demand allocation, physical data isolation, and fault isolation.

GBase 8a MPP V9 clusters also perfectly support multi-site data center architectures spanning cities. Suppose a bank operates three data centers in Beijing, Shanghai, and Xi'an. The management nodes responsible for scheduling can be deployed across all three sites—for example, three nodes in Beijing, three in Shanghai, and three in Xi'an. These nine management nodes form a single scheduling cluster and serve requests together. Similarly, logical sub-clusters can be deployed locally at each site based on proximity. Under this model, the cluster can accept requests from nearby branches, enabling wide-area load balancing. Combined with the GBase 8a MPP V9 mirror cluster feature, this makes cross-site high availability for massive data warehouses effortless.

Next, let's examine the resource management technology. The virtual cluster technology described above partitions and manages branch resources at the enterprise architecture level. To fully achieve multi-tenancy, resource management further refines the physical resources within logical sub-clusters. Within the physical machines allocated to each branch, CPU, memory, storage, network, and priority can be organized into resource pools based on user needs. These resource pools are then bound to users, delivering resource isolation and dynamic control. As a result, branches satisfy headquarters' centralized management requirements while retaining the flexibility to allocate and use their assigned resources, smoothing the transition to data centralization.

GBase 8a MPP V9 also implements the Open Service Broker standard protocol. Through cloud platforms or PaaS platforms, it can rationally and effectively manage the relationship between resources and their consumers in a multi-tenant model, enabling open resource sharing and meeting the flexible business demands of resource consumers.