After a long time observing my application and analyzing why sometimes I obtained less than 100 ms per request in most of the actions of my application and on other occasions it did not go below approximately 2 seconds, I realized that everything depended on what machine was running. machine 17xxxxxxxx was the slow one and 3dxxxxxxx was super fast, if I want my app to go fast then I stop 17x and make sure that 3dx is running… Can someone explain to me why it behaves like this, both are in the same region, they are of the same type. Thank you
Hi @cluzstudio
What kind of app are you running?
One possibility could be that the majority of requests go to machine 3dx and only a small number go to 17x, causing database connections on 17x to be closed for inactivity then needing to be reopened when a request comes in.
No, at that time, the activity was generally on 17xxx, the 3d was barely used… What I think it was is the type of machine, I upgraded to performance and I think that shared machine was left, I already destroyed it and so far everything is fine… What I am afraid will happen next… With this machine that is now I am getting less than 100 ms per second… The application is a social network on Node and Apollo server with Prisma as ORM… https://creo.red
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.