Comment by markus_zhang

This is pretty damning, if half of it is true. I don’t work at Microsoft and I don’t have the knowledge to judge the reliability of Azure, but I do have friends who work as users of Azure and the words are not kind, especially the new Fabric database which is said to be crazy to pick for production at this stage — while MSFT switched the certification to Fabric already, pushing its customers to use it.

I’ll never work in a company that uses Azure as its main cloud services, just for the sake of quiet nights.

I do wonder what does it look like inside AWS and GCP, though. Is it the same level of chaos, but just because they started early they got more success? If that’s the case, maybe we can conclude, that very large cloud operation is not sustainable under the current company structure — because either the technical knowledge required is too dense, but companies won’t be able to retain workers, or because companies are forced to join the horn of the marketeers, eventually.