请问HN:你如何确定数据在100个微服务中分布的位置?

1作者: doodledood大约 2 小时前原帖
每次我需要使用一个不熟悉的服务时,都会经历同样的循环:在代码库中搜索,找到过时或不存在的文档,联系可能记得细节的人。对于任何复杂的事情,这通常会消耗2到3个小时,才真正开始工作。<p>我很好奇,大家在大规模应用中实际使用的是什么。是类似Backstage或Cortex的目录?基于AI的方法?还是仅仅依赖于部落知识并接受这种成本?
查看原文
Every time I need to work with an unfamiliar service, it’s the same loop: grep through repos, find stale or nonexistent docs, Slack someone who might remember the nuances. For anything complex this eats 2-3 hours before actual work starts.<p>Curious what’s actually working for people at scale. Backstage&#x2F;Cortex-style catalogs? AI-based approaches? Or just tribal knowledge and accepting the tax?