Application Refactoring · IT Operations Automation
Shenda Intelligent needed to reduce the risk of all-at-once releases, prevent service avalanches during holiday traffic surges, eliminate the operational bottlenecks of a self-managed Nacos cluster, and free engineering time for AI-agent, intelligent-customer-service, and crowd-forecasting initiatives.
Reported impact
Shenda Intelligent raised core-system availability to more than 99.95%, improved service stability by 60%, cut operations labor by 60%, increased cloud resource utilization by 30%, raised development iteration speed by 50%, and gave engineers more time to build AI-driven smart-tourism capabilities.
Zhejiang Shenda Intelligent Technology Co., Ltd. rebuilt its tourism-platform foundation on Alibaba Cloud so it could operate high-concurrency scenic-ticketing and guidance systems while moving faster on AI applications. The company adopted Microservices Engine (MSE) as the control plane for lossless instance offlining, gradual traffic warmup, and full-link canary routing, which let new versions and freshly scaled instances absorb traffic safely during holiday peaks. Shenda also migrated its self-managed Nacos cluster to MSE's managed Nacos so service registry and configuration no longer bottlenecked microservice stability. Running on Container Service for Kubernetes (ACK), the platform uses MSE's AI-native service governance to register AI components through MCP and hot-update prompts and model parameters without restarts, while Application Real-Time Monitoring Service (ARMS) and Log Service provide the telemetry needed to spot faults quickly and keep releases stable.
Traffic for ticketing, guidance, and crowd-management services runs on ACK. MSE routes tagged canary traffic and gradually warms new instances before they take full production load, while managed Nacos inside MSE maintains service discovery and AI-component configuration. ARMS and Log Service collect runtime telemetry so releases and scaling events can be monitored and corrected quickly.
"After migrating self-built Nacos to managed Nacos on MSE, the company gained 99.99% high availability and AI-native governance such as MCP-based AI service registration and dynamic prompt and parameter hot updates. (original: 深大智能将自建 Nacos 集群迁移至 MSE 托管版 Nacos，获得 99.99% 高可用保障。利用 Nacos 3.0 内置的 MCP 服务注册与发现能力，统一纳管 AI 组件；支持动态 Prompt 与参数热更新，无需重启即可实时调整大模型参数。)"
"Core-system availability reached more than 99.95%, service stability improved by 60%, development iteration speed improved by 50%, operations labor input fell by 60%, and cloud resource utilization improved by 30%. (original: 核心系统可用性达 99.95%+，服务稳定性提升 60%。发布效率倍增：高频发布零回滚、零事故，研发迭代速度提升 50%。成本优化：运维人力投入减少 60%，云资源利用率提升 30%。)"
"MSE provides lossless offlining and traffic warmup so even during million-QPS holiday peaks, elastic scaling and version updates can remain stable. (original: MSE 提供无损下线能力，在应用停止前自动摘除服务注册并等待存量请求处理完毕；同时利用流量预热（Warmup）能力，在新实例启动后逐步导入流量。即使在国庆、春节等百万级 QPS 业务高峰期进行弹性扩缩容，系统依然平稳运行。)"
"Traditional operations could not support intelligent transformation because engineers were tied up in middleware maintenance, incident response, and release assurance instead of AI agents, intelligent customer service, and crowd forecasting. (original: 大量人力被消耗在基础中间件维护、故障排查和发布保障上，无法聚焦于更具价值的 AI Agent 开发、智能客服、客流预测等创新业务。)"
"Using MSE's full-link canary release, Shenda tags specific traffic and routes it precisely to new-version services for live validation, reducing release risk by more than 90%. (original: 借助 MSE 的全链路灰度发布能力，深大智能通过为特定流量打标，精准将流量路由至新版本服务进行真实验证，其余用户仍使用稳定版本。发布风险降低 90%+。)"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Alibaba Cloud Sample Jul 28, 2026 Zhuiyin Technology (追因科技) Zhuiyin Technology had already built an agent for cross-border ecommerce analysis, but once the agent was exposed to customers, the team discovered three hard problems: results were hard to judge objectively, reasoning… inference quality improved 20% AgentLoop · Trekin Agent Alibaba Cloud Sample Jun 9, 2026 AsiaInfo Technologies (亚信科技) AsiaInfo had to raise software-engineering productivity while maintaining large, long-lived telecom BSS and OSS systems whose logic spanned multiple generations from 2G to 5G. The codebase and architecture were highly… average end-to-end efficiency from requirements to testing improved 8.6% Qoder · Qoder CN Alibaba Cloud Sample Apr 25, 2025 Hillstone Networks (山石网科) Hillstone Networks needed to accelerate software delivery for its security product portfolio while protecting source code and keeping AI assistance inside enterprise governance boundaries. Traditional development… 90%+ of developers use Qoder CN daily Bailian · DeepSeek-R1 · Qoder CN Dedicated Edition 