知识卡片
预生产部署检查实际候选
每次合适的代码变更形成候选并自动部署到预生产,通过部署后检查确认服务真正可运行。测试包通过与目标环境可用是两个不同问题。保持候选随时可发布的能力,也要保留生产专属配置和数据差异的验证。
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
- · 查看原书依据
CD4ML takes the ability to automate model deployments further by recommending teams: (i) trigger automated deployments—minimally to preproduction environment(s)—on every code push, and (ii) run post-deployment tests to verify that the deployment succeeded and is ready for production at any time.
参考来源
- 位置:《Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)》第09章《MLOps与持续交付》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)/_epub-src/OEBPS/ch09.html)
- 结论依据:摘录中关于预生产部署检查实际候选的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:CD4ML takes the ability to automate model deployments further by recommending teams: (i) trigger automated deployments—minimally to preproduction environment(s)—on every code push, and (ii) run post-deployment tests to verify that the deployment succeeded and is ready for production at any time.