SOFAStack Service Mesh is an enhanced service mesh platform from Ant Group. It is built on financial-grade production practices and deeply integrates traditional microservices with Service Mesh technology. Its core technology has been proven in large-scale production environments at Ant Group. The platform seamlessly integrates with SOFAStack classic and containerized application services to provide an easy-to-use support platform for a Service Mesh architecture.
SOFAStack Service Mesh overview
SOFAStack Service Mesh = Traditional Microservices + Service Mesh.
Service Mesh provides a platform-agnostic, language-agnostic, lightweight, and non-intrusive solution for cloud-native distributed architectures. It is compatible with the Kubernetes and Istio ecosystems, supports virtual machine scenarios, and provides service administration for microservice frameworks, such as SOFA, Dubbo, and Spring Cloud. Integration requires no modification to business code. This helps enterprises adopt cloud-native technologies at a minimal cost and accelerate their digital transformation.
Service Mesh diagram
From a global perspective, Sidecars connect to form a network, which forms the Service Mesh.
The TCP protocol led to distributed systems, and distributed systems led to microservices. Service Mesh is the next generation of microservice technology, often described as the TCP protocol for the microservice era. Using a Sidecar proxy, Service Mesh decouples service administration from business logic and moves it into a separate process. This enables unified administration and network security for heterogeneous systems.
Sidecar pattern diagram
Service Mesh moves the functionality of the SDK into a separate process and deploys it as a Sidecar. This allows the business process to focus on business logic.
A Sidecar is a lightweight network proxy deployed alongside an application. It intercepts all incoming and outgoing network requests to enforce network policies. These policies enable features such as service discovery, load balancing, traffic splitting, fault injection, circuit breaking, and phased releases.
Features
Service Mesh includes the following features for service management and monitoring.
Provides unified registration and detailed views for heterogeneous services. You can view basic information, service providers, and consumers.
Supports service throttling, routing, circuit breaking, and authentication. Provides fine-grained service routing and throttling to ensure high availability (HA) and continuous business operations.
You can view and manage Sidecar instances in the current workspace. This information includes the Sidecar status, pod status, and injection time.
Provides a visualization of the call relationships and dependencies between application services. It also displays real-time monitoring information for each node, such as request volume, response time, and error rate.
Provides overall statistics and performance metrics for all applications. You can monitor application service throughput, response time, RPS, and status in real time to promptly detect service abnormalities.