Microservice Architecture for Payment Scenarios: Design, Implementation, and Performance Optimization

This excerpt is taken from the book "High‑Availability and Scalable Microservice Architecture: Based on Dubbo, Spring Cloud, and Service Mesh" and introduces the practical application of microservices in payment systems.

1. Introduction to Payment Scenarios

The chapter explains various internet payment methods such as card swipe, QR code, public account, WAP, app, gateway, and quick payment, and states the goal of decoupling a monolithic payment system into reliable, concurrent microservices.

2. Payment Business Modeling and Service Partitioning

Key domain‑driven design concepts (domain, sub‑domain, bounded context) are introduced, followed by a detailed service layer diagram (Fig 11‑1) that divides the system into business core domains and supporting sub‑domains, including integration, gateway, product, business, and infrastructure layers.

3. Detailed Microservice Architecture Analysis

The architecture is split into business and technical layers (Fig 11‑2). The business layer shows service decomposition (e.g., payment, transaction, refund, billing, risk control). The technical layer discusses framework choices (Spring Boot + Spring Cloud, Dubbo, gRPC), messaging middleware (RabbitMQ, Kafka), configuration centers (Apollo, Diamond, Disconf), and bank channel monitoring.

4. Code‑Level Performance Enhancements

Common pitfalls and solutions are presented:

Database deadlocks caused by improper SELECT ... FOR UPDATE usage (see Table 11‑1).

Long‑running transactions mixing remote calls; recommendation to keep transactions short.

Thread‑pool misuse: avoid unbounded Executors.newCachedThreadPool(), prefer fixed pools with bounded queues.

Improper logging causing synchronization bottlenecks; adjust Log4j pattern to reduce contention.

Indexing best practices for MySQL (left‑most principle, avoid NULL columns, use appropriate operators).

Sample code snippets are preserved below:

public void test() {
    Transaction.begin //事务开启
    try {
        dao.insert //插入一行记录
        httpClient.queryRemoteResult() //请求访问
        dao.update //更新一行记录
        Transaction.commit() //事务提交
    } catch (Exception e) {
        Transaction.rollFor //事务回滚
    }
}

private static final ExecutorService executorService = Executors.newCachedThreadPool();
/**
 * 异步执行短频快的任务
 * @param task
 */
public static void asynShortTask(Runnable task) {
    executorService.submit(task);
    //task.run();
}

5. Operational Practices and Fault Diagnosis

Discusses service degradation strategies, big‑data reporting across microservices, automated deployment (DevOps), and troubleshooting tools such as hprof for CPU/memory profiling and pidstat for per‑thread CPU usage.

Overall, the chapter provides a practical roadmap for designing, implementing, and maintaining high‑performance, highly available microservice‑based payment systems.