On 6-Apr-06, at 10:51 PM, Charlie Brady wrote:
On Thu, 6 Apr 2006, Matt Sergeant wrote:
On 5-Apr-06, at 8:15 PM, John Peacock wrote:
I that's all it is, we can easily(?) change forkserver to prefork
some
percentage of its max connections and dole them out as needed. I
may try to do
that anyways because the inbound machines I use won't be getting
Apache2 anytime
soon.
I think that's where the big performance gains come from.
I haven't seen yet any evidence of performance gains, big or
otherwise. Nor a definition of what constitutes a performance gain.
Reduced latency? CPU load? Message throughput? Memory footprint?
Should be reduced CPU load. Though it's a bit more complex than that,
as the kernel simply copes better when re-using sockets.
Matt.
