知识卡片
函数细粒度增加观测对象
把服务操作拆成函数,再按租户复制部署,会大幅增加运维对象。指标和排障要能定位具体函数、租户及其部署,不能只保留微服务总览;计算管理变简单并不代表整个系统的运营自动变简单。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- Serverless SaaS: Architecture Patterns and Strategies · 查看原书依据
we have distributed deployments. However, I feel like serverless magnifies the potential impacts of this problem. With serverless, we can have much more fine-grained units of deployment and management. As an example, in a traditional compute model, my unit of management and operational visibility tends to be more at the microservice level where the microservice represents the composite of all
参考来源
- 位置:《Building Multi-Tenant SaaS Architectures》第11章《Serverless SaaS: Architecture Patterns and Strategies》(来源项目 a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Building Multi-Tenant SaaS Architectures/_epub-src/OEBPS/ch11.html)
- 结论依据:原文具体说明把服务操作拆成函数,再按租户复制部署,会大幅增加运维对象;本卡据此归纳机制及适用边界。
- 原始内容:we have distributed deployments. However, I feel like serverless magnifies the potential impacts of this problem. With serverless, we can have much more fine-grained units of deployment and management. As an example, in a traditional compute model, my unit of management and operational visibility tends to be more at the microservice level where the microservice represents the composite of all