Thank you for this article!
I have a question regarding the 15 containers vs 750 pods, let's say you have a system of 5 microservices with one being the client-facing backend that needs to scale according to number of outside requests, and the other 4 being internal but also need to scale depending on the traffic on the backend instance.
And let's imagine you really have a lot of requests, so you need many bigger machines in order to serve all the workload.
Would it be better to run each of the 5 services once on each machine (or even just one service per machine) but have them configured to handle a lot of requests, or to have the same service multiple times on the same machine and configured to serve less requests?
The question is not about the aspect that the latter of course is better for load balancing, more if one process handles lots of requests turns out to be more error prone than running the same service multiple times on the same machine.
(Or in other words, why would one need 750 pods on the same machine)
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.