Evolution of System Architecture: From Single‑Machine to Distributed Designs

The author divides the evolution of system architecture into four main stages, each driven by performance, reliability, continuity, and technological advances.

1. Single‑Machine Mode

In the early, fast‑growth phase of IT construction, organizations built whatever systems they needed (ERP, HIS, etc.) without high‑availability requirements.

2. Dual‑Machine Hot‑Standby and Mirroring

As users could no longer tolerate hours of downtime, active‑standby configurations appeared, providing rapid failover but incurring resource waste, storage dependence, and single‑point data risks. Numerous products (RoseHA, NEC ExpressCluster, MSCS, Symantec VCS, Legato, RHCS, etc.) emerged to address these issues.

To solve data‑single‑point problems, primary‑secondary and dual‑active storage solutions also appeared.

3. Multi‑Node Active Cluster (Node Multi‑Active)

With ever‑growing business load and data volume, single‑node scaling became insufficient, leading to horizontal scaling via multi‑node clusters such as Oracle RAC, Microsoft AlwaysOn, and Moebius, which distribute load and separate OLAP/OLTP workloads.

4. Distributed Architecture

Distributed systems further decompose data and services (horizontal/vertical sharding, database‑level splitting, business‑level splitting) to handle massive scale, though they require sophisticated design and skilled personnel.

The author notes that many traditional enterprises remain stuck in the first two stages, while modern enterprises adopt multi‑active and distributed solutions when resources allow.

Other Technical Discussions

Various data replication technologies (DG, OGG, Log Shipping, Replication) and infrastructure solutions (virtualization, hyper‑convergence, dual‑active storage) support high availability but sit beneath the database layer.

How to Choose an Architecture

Selection starts by identifying the generation needed: redundancy (second generation), coarse splitting (third), or fine splitting (fourth). The appropriate architecture balances business requirements, cost, and technical capability.

Construction and Maintenance

Building and maintaining complex architectures demand specialized staff; without them, enterprises often rely on third‑party products, and ongoing maintenance becomes the critical factor for stability, scalability, and cost control.

Conclusion

Choosing the right architecture—whether single‑machine, hot‑standby, multi‑node active, or distributed—depends on business needs, ensuring stability, security, efficiency, and sustainability while avoiding unnecessary complexity.

-END-

© Content originally from cnblogs.com; all rights belong to the author.