4 ms·
the conclusion you stated here is what everyone says --> "Fine for tinkering, not practical for actual work." And that is the exact reason why i posted this he
by patchthecode 5y ago
the conclusion you stated here is what everyone says --> "Fine for tinkering, not practical for actual work." And that is the exact reason why i posted this here. Because i do not have a good reason as to "why NOT have it as a production server" when (1) those guys in the article did just that and served millions of requests per day (2) they even have a hosting service of Raspberry PIs and have had this going pretty well for some time. These are all examples of PIs used in production, but everyone still says "dont do it" and my question is why? The difficulties you mentioned above is easily accounted for, so i was wondering if there was any other points i was missing?
- PragmaticPulp 5y agoSame reason nobody hosts their production website on a Celeron box in the corner of the office: Production hosting is about more than just requests per second. The Pi server thing is a proof of concept that ignores everything else that goes into production hosting. You can do it, it’s just going to be miserable like any self hosting, slower than any x86 hardware, and require the equivalent of thousands or tens of thousands of dollars of engineering time just to accomplish what you could get with a few clicks and 15 minutes with AWS. There’s a reason their hosted Rasperry Pi service is more expensive than getting an EC2 micro instance: https://www.mythic-beasts.com/order/rpi https://www.mythic-beasts.com/order/rpi
- patchthecode 5y agohmm. ok then. thanks. I will take the things you have said into consideration.