Traditional R&D and O&M systems using the SOFA technology stack
Applications in these systems are developed with SOFABoot or use SOFA Mesh directly. These systems have complex inter-system dependencies and heavily use SOFAStack products. They must integrate seamlessly with the publishing and deployment capabilities of your existing Platform as a Service (PaaS).
These systems support simultaneous publishing and O&M for multiple applications. They also provide advanced capabilities, such as application grouping and dependency adjustments.
They require blue-green and cell-based deployment capabilities.
These systems have the following characteristics:
The application architecture is complex and has inter-application dependencies. This requires coordination and fine-grained control within the release ticket. This includes features such as pod grouping (with beta), group retries, single-pod retries, and retrying or ignoring single tasks on a pod.
They involve scenarios with mixed deployments of virtual machines and containers, often during migration.
They have a strong dependency on Ant middleware and require lossless publishing. This means operations such as switching Server Load Balancer traffic, managing software-based load balancing traffic, adding to a whitelist, and publishing must be performed in a specific order.
In this scenario, create an application service in AKS. Use the SOFABoot runtime image as the base image to build your application image. Then, use in-place upgrades for release-level deployments. This allows the application service to run on AKS, interact with services on virtual machines, and integrate seamlessly with existing SOFAStack products.
Lightweight R&D and O&M systems using the SOFA technology stack
Applications in these systems are developed with SOFABoot or use SOFA Mesh and are tightly integrated with SOFAStack products. These applications have the following characteristics:
They do not require simultaneous publishing of multiple applications because each application can be published independently.
They require lossless publishing.
They have no legacy constraints and are suitable for cloud-native O&M practices.
In this scenario, create an application service in AKS. Use the SOFABoot runtime image as the base image to build your application image. Manage deployments using in-place upgrades. This allows the application service to run on AKS and integrate seamlessly with existing Ant Financial Technology Cloud products.
Cloud-native technology systems
Applications in these systems typically use traditional technology stacks such as Spring or Spring Boot. They use Eureka and ZooKeeper for service registration and discovery. They also use monitoring and tracing tools from the Cloud Native Computing Foundation (CNCF) ecosystem to create a self-contained system. These applications have the following characteristics:
They are for business services with high fault tolerance or are non-critical systems.
They are published independently.
They have no legacy constraints and are suitable for cloud-native O&M practices.
They do not need to be integrated with SOFAStack products for features such as monitoring, elastic scaling, disaster recovery, or middleware.
In this scenario, create an application service directly in AKS to manage your existing code or images. Use native deployment methods for management. Associated tools, such as configuration centers, monitoring tools, tracing tools, and Service Mesh, can remain the same. Deploy and manage them directly using Kubernetes-supported methods, such as Helm Charts.