alexandre - aldeia digital <adaldeia(at)gmail(dot)com> wrote:
> Notice that we have no idle % in cpu column.
So they're making full use of all the CPUs they paid for. That in
itself isn't a problem. Unfortunately you haven't given us nearly
enough information to know whether there is indeed a problem, or if
so, what. What was throughput before? What is it now? How has
latency been affected? And all those unanswered questions from my
first email....
The problem *might* be something along the lines of most of the
discussion on the thread. It might not be. I just don't know yet,
myself.
> 14:26:47 up 2 days, 3:26, 4 users, load average: 48.61,
> 46.12, 40.47
This has me wondering again about your core count and your user
connections.
> My client wants to remove the extra memory... :/
Maybe we should identify the problem. It might be that a connection
pooler is the solution. On the other hand, if critical production
applications are suffering, it might make sense to take this out of
production and figure out a safer place to test things and sort this
out.
-Kevin