1 views collected around this technical thread.
This article introduces and compares four common microservice architecture patterns—aggregation, shared database, proxy (including sidecar), and asynchronous messaging—detailing their designs, advantages, disadvantages, and typical use cases for large‑scale systems.
This article explains asynchronous messaging patterns, compares RabbitMQ and Apache Kafka in terms of architecture, queues, topics, and use cases, and provides guidance on selecting the appropriate platform for different scenarios in modern systems.
This article explains the fundamental asynchronous messaging patterns, compares RabbitMQ and Apache Kafka in terms of architecture, queue and pub/sub models, and provides guidance on selecting the appropriate technology for different backend scenarios.
This article explains the fundamental differences between RabbitMQ and Apache Kafka, covering their architectures, message models, ordering guarantees, routing capabilities, timing controls, retention policies, fault‑tolerance mechanisms, scalability, and consumer complexity, and then provides practical criteria to help architects decide which solution best fits a given use case.
To solve the scalability, visibility, and debugging challenges of heterogeneous asynchronous calls after Alibaba Momma’s advertising services shifted to micro‑services, the team created the AMB (Alibaba Momma Application Message Bus), a console‑driven, bus‑style platform with layered connectors, reusable components, isolated JStorm topologies, and centralized monitoring that turns hours‑long coding into rapid visual workflow configuration.
This article explains the importance of asynchronous messaging and event‑driven communication for coordinating changes across multiple microservices and bounded contexts, covering single‑receiver and multi‑receiver patterns, message brokers, eventual consistency, publishing/subscribing mechanisms, and practical implementation considerations.
The article uses Starbucks' coffee‑ordering process to illustrate asynchronous order handling, correlation identifiers, exception‑handling strategies, two‑phase commit, and conversation patterns, showing how real‑world practices inform the design of reliable distributed messaging systems.
The article examines how to reduce tight coupling between microservices by applying module‑division principles, converting synchronous calls to asynchronous messaging, leveraging event‑driven architectures and CQRS, employing local caching and data landing, and refactoring overly coupled services for more resilient backend systems.
The article uses Starbucks' coffee‑ordering process as a concrete analogy to explain asynchronous messaging, correlation, error‑handling strategies such as write‑off, retry and compensation, and why two‑phase commit is unsuitable for high‑throughput services.
The article explains how microservice architectures increase testing complexity and presents practical strategies—including orchestration testing, service virtualization, handling API changes, asynchronous command patterns, event firehose, and CI/CD environment configuration—to reliably test individual services and their interactions in cloud‑native environments.
The article summarizes the design principles of reactive microservices, covering isolation, single responsibility, autonomy, private state, asynchronous messaging, service discovery, API management, and security, and explains how these concepts enable scalable, resilient cloud‑native systems.
The article explains the widespread use of message oriented middleware, outlines its core challenges such as synchronous vs. asynchronous delivery, security, persistence, redelivery, ordering, and describes the two main transmission models—point‑to‑point and publish/subscribe—while comparing JMS and AMQP standards.
This article details the three-stage evolution of Xiaomi's e‑commerce architecture, introduces the design of the Notify asynchronous message system built on Redis and MySQL, and explains the subsequent upgrades—including agent proxy, Go‑based modules, and MyCAT integration—to improve scalability, reliability, and performance.