The win is _also_ about scale. The server CPU usage went down significantly - see http://bit.ly/1vqp1ki
So, more request can be served with a smaller number of servers.
So, more request can be served with a smaller number of servers.
The un-shortened URL is https://ganglia.wikimedia.org/latest/graph.php?r=year&z=xlar...
(I should've put a disclaimer there: I am one of the WMF engineers involved in the project).