I am honestly wondering just how relevant "vendor lock-in" really is in the majority of the cases. I have personally only worked with one org that was seriously considering porting their cloud infra over to another provider (AWS->GCP, for example). I have however worked with several orgs where concern about lock-in bubbled up, but only as a strawman argument for/against XYZ tech.<p>So, my question to the community:<p>"Have you/your org been faced with vendor lock-in, specifically for cloud services such as AWS Lambda?"<p>A follow up would be:<p>"If so, how did it impact the project/product?"
Years ago, I worked at a large stock trading website. We deliberately used both Oracle SQL and Sybase SQL, with an interface layer. We deliberately did not use any special features from either. We ran both in production, some services on one, rest on other. We would occasionally test all services on each one. Reasons: price protection; the product or the vendor would disappear; future upgrades would break our system; sticking with 'standard' SQL features (no shiny new undebugged functions); and easier developer training.
What do you do when your sole vendor doubles or triples the price?