# Articles

> Every article, newest first.

---

LLMS index: [llms.txt](/llms.txt)

---

Section pages:

- [From Hardware to Software: Rethinking DR Networking in the Cloud Era](/blog/post/cloud-dr-networking/): How public cloud transforms disaster recovery from physical network engineering to software-defined infrastructure
- [Beyond Lift and Shift: Why Cloud Migration Demands Database Code Optimization](/blog/post/beyond-lift-and-shift-database-code-optimization/): Moving Oracle databases from physical servers to cloud ECS with EVS storage is not just an infrastructure change. Without optimizing SQL and database-interaction code, cloud deployments will hit storage I/O bottlenecks that didn't exist on bare metal. Here's why — and how much effort the fix really takes.
- [First Time on Managed MySQL: A Practical Guide to Huawei Cloud RDS](/blog/post/huawei-cloud-rds-mysql-getting-started/): Migrating from self-hosted MySQL on ECS to Huawei Cloud RDS for MySQL — a step-by-step guide covering connectivity, configuration, and the key pitfalls that catch first-time RDS users.
- [Why Most Enterprise Digitalization Fails — and the Only Way Out](/blog/post/why-digitalization-fails-and-the-only-way-out/): Enterprises spend millions on ERP, OA, and CRM systems, yet employees keep using WeChat groups, Excel sheets, and paper forms for real work. This is not an edge case — it is the most common and most fatal pattern in enterprise digital transformation. The root cause is that informatization was never truly executed, and without clean data in the system, digitalization is an illusion and AI is useless.
- [Purdue Model on Public Cloud — Not a Compromise, But an Evolution](/blog/post/purdue-model-on-public-cloud/): Traditional industries moving OT workloads to public cloud fear violating the Purdue Network Security Model. But the Purdue Model defines logical data-flow controls, not physical infrastructure. Public cloud can implement every Purdue boundary more rigorously than on-premises data centers ever could.
- [Load Balancing First: Why Traditional Industry Owners Should Resist the Microservices Jump](/blog/post/load-balancing-vs-microservices-enterprise/): Microservices solve organizational complexity at internet scale. For most enterprise systems serving a few thousand users, load-balanced horizontal scaling is the right architecture — lower risk, faster to deliver, and cost-controllable on public cloud.
- [From Monolith to Distributed: Best Practices and Effort Estimation](/blog/post/monolith-to-distributed-architecture/): Traditional industry software — factory MES, hospital HIS, power SCADA, AMI metering platforms — was built for a different era. Here's a grounded roadmap for migrating to distributed architecture, with realistic effort estimates for each stage.
- [PostgreSQL + TimescaleDB on Huawei Cloud vs Oracle for AMI Projects](/blog/post/timescaledb-vs-oracle-ami/): Oracle has been the default choice for AMI platforms for years. TimescaleDB on Huawei Cloud RDS for PostgreSQL offers better time-series performance, native hypertable partitioning, and a dramatically lower total cost. Here's the full comparison.
- [From On-Premise to Cloud: Why Distributed Architecture Is the Right Move](/blog/post/cloud-migration-distributed-architecture/): Traditional primary/standby server deployments create hard ceilings on scalability. Distributed architecture on public cloud removes those ceilings — and eliminates forced downtime for capacity changes. Here's why and how.