DragonFly BSD
DragonFly users List (threaded) for 2006-11
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

Re: load balancing


From: Joerg Sonnenberger <joerg@xxxxxxxxxxxxxxxxx>
Date: Fri, 03 Nov 2006 08:38:58 +0100
Mail-followup-to: users@crater.dragonflybsd.org

On Thu, Nov 02, 2006 at 10:48:36PM -0800, David Cuthbert wrote:
> Most large sites don't use keep alive.  It ties up a worker thread 
> waiting for another request to come from your home PC, whose link is 
> this slow megabit-plus high-latency straw compared to our OC-48 firehose 
> which is 3-4 hops from MAE East.

Only if you use a broken web server. With a proper O(1) event
notification mechanism and async web servers, keep-alive is a huge win,
if you don't have a single server load high enough to run out of file
descriptors. Keep in mind that especially the high latency links are
gaining most by using keep-alive as TCP needs start to stabilise near
the bandwidth delay product -- and most end users have lots of both
facors.

Joerg



[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]