Recently I noticed that some of my Endpoints started having increased runtimes on my project, but it seems to be pick-and-choose to when it decides to have an increased runtime.
I do know of the limitations of the M0 clusters Data Transfers being 10GB in/10GB out, and right now my project is within that 10GB, so is there any other reason as to why I would be experiencing this kind of performance issue? Is there any way I can fix this issue?
Hi @Travis_Engle,
M0 is, like M2 and M5, a shared tier: if any of the other tenants on the same VM is putting any kind of load, you may well observe inconsistent response times.
In a nutshell, don’t consider any of the responsiveness you get from a shared tier as indicative of the general performance of the system, these kind of load tests would better be executed on a dedicated tier (M10 or M20), while for a sustained load, like in a production environment, a M30 is advised (M10 and M20 can be throttled)