Evolution of Server‑Side Architecture from Single‑Machine to Cloud‑Native for High Concurrency

Introduction: defines basic concepts such as distributed systems, high availability, clustering, load balancing, and forward/reverse proxy.

Evolution steps:

1. Single‑machine architecture – initial deployment of Tomcat and database on one server.

2. Separate Tomcat and database – improves performance by isolating resources.

3. Add local and distributed cache – uses memcached and Redis to reduce database load, introducing cache consistency, penetration, and avalanche challenges.

4. Introduce reverse‑proxy load balancing – Nginx or HAProxy distributes traffic across multiple Tomcat instances.

5. Database read/write separation – MyCAT middleware enables read replicas and synchronizes writes to reads.

6. Database sharding – splits large tables; MPP databases such as TiDB, Greenplum, PostgreSQL‑XC provide horizontal scalability.

7. LVS/F5 load balancing – layer‑4 balancers handle tens of thousands of concurrent connections with high availability.

8. DNS round‑robin across data centers – distributes traffic at the DNS level to achieve data‑center‑level scaling.

9. Introduce NoSQL and search engines – HBase, Redis, Elasticsearch, Kylin, Druid, etc., address big‑data storage, key‑value access, full‑text search, and multidimensional analysis.

10. Split monolith into smaller applications – modularize by business domain for clearer responsibilities.

11. Extract shared functions as microservices – use Dubbo, Spring Cloud for service governance, rate limiting, circuit breaking, and degradation.

12. Enterprise Service Bus (ESB) – unify protocol conversion and reduce coupling between services.

13. Containerization – Docker packages applications; Kubernetes orchestrates dynamic deployment and scaling.

14. Cloud platform – migrate to public cloud (IaaS/PaaS/SaaS) for elastic resources, reducing operational cost and enabling on‑demand scaling.

Finally, the article lists architectural design principles such as N+1 redundancy, rollback capability, feature toggles, monitoring, multi‑active data‑center design, mature technology adoption, resource isolation, horizontal scalability, buying non‑core components, using commercial hardware, rapid iteration, and stateless design.