This topic describes the system architecture of Realtime Compute in exclusive mode.

System architecture

The following figure shows the system architecture of Realtime Compute in exclusive mode.System architecture
  • For Realtime Compute in exclusive mode, all your purchased ECS instances are fully hosted in the VPC of your Realtime Compute cluster. In this mode, you cannot log on to ECS instances.
  • When you create a Realtime Compute cluster, Realtime Compute applies for an Elastic Network Interface (ENI) under your Realtime Compute account. You can use this ENI to access all resources in your VPC.
  • To enable Internal access for your Realtime Compute cluster, you can bind a network address translation (NAT) gateway and an Elastic IP address (EIP) to the ENI. For more information, see Bind an EIP to a NAT Gateway and Overview.
  • The ENI belongs to an independent security group under your Realtime Compute account. To access services of another security group in the VPC, you must configure inbound and outbound rules for the security group.
Note You are charged for the use of the ENI only when your Realtime Compute cluster accesses the Internet.


  • End-to-end real-time data computing development
    • Provides a real-time data processing capability based on Flink SQL, which implements automatic data recovery. This ensures accurate data processing if failures occur.
    • Supports multiple built-in functions, such as string, date, and aggregate functions.
    • Supports various window types, such as tumbling, sliding, and session windows.
    • Implements accurate control over computing resources, which ensures resource isolation for jobs.
    • Provides the following key performance metrics that are superior to those of open source Flink:
      • The data computing latency can be measured in sub-seconds.
      • The throughput of a single job can reach millions of records per second. A single cluster can consist of thousands of servers.
    • Deeply integrates all types of cloud data storage systems such as DataHub, Log Service, ApsaraDB for RDS, Tablestore, and AnalyticDB for MySQL so that you can conveniently read and write data from and to these systems.
  • Fully managed Realtime Compute service
    • Uses a fully managed stream computing engine.
    • Allows you to run and query streaming data without the need to provision or manage any infrastructure.
    • Allows you to enable streaming data processing services with one click.
    • Integrates features such as data storage, data development, data O&M, and monitoring and alerts. This reduces both the trial and migration costs of stream processing.
    • Effectively isolates and protects the managed and running services of different tenants.
  • Low manpower and cluster costs
    • Significantly optimizes the SQL execution engine to deliver computing jobs that are more cost-effective than native Flink jobs.
    • Significantly reduces development and operation costs, which are much lower than those of open source stream frameworks.
  • High availability
    Assume that an ECS instance or a job of Realtime Compute fails. You can use the JobManager and TaskManager on an available ECS instance in the same zone to achieve high availability of jobs. You can also use the JobManager and TaskManager on an available ECS instance in another zone or region to achieve high availability in zones.High availability of clusters