10,000? The last I heard, FB had over 60,000. This is not where I initially saw it but it says the same.
http://www.datacenterknowledge.com/archives/2010/06/28/facebook-server-count -60000-or-more/ - - - Anytime someone tells me CF can't handle big apps with heavy loads, I make them clarify what they think "heavy load" actually is. Then I usually blow that theory out of the water with a single CF app serving 650,000+ requests per hour. (load balanced across 13 servers) It's nothing compared to FB or mySpace but nothing to "pffft" at either. I'm sure many of you have seen bigger numbers... so why people still like to say CF cant handle it is beyond me. I still remember a conference call with an adobe rep not long ago: Rep: How many requests per hour do you handle on a server? Us: About 50,000. Rep: No, just for one hour. Us: About 50,000. Rep: No, not for all servers, just one... Us: About 50,000... He still didn't act like he believed it. Do those numbers seem high to any of you? I'd have to think there are plenty of people out there seeing much more load than that on a CF server. .:.:.:.:.:.:.:.:.:.:.:.:.:. Bobby Hartsfield http://acoderslife.com -----Original Message----- From: rex [mailto:[email protected]] Sent: Wednesday, July 28, 2010 1:45 PM To: cf-talk Subject: Re: www.cbo.gov down. CF to blame? Any technology can be setup so that it doesn't scale well, or not scale at all. This can be on PHP, dotNet, Rails, IIS, Apache, Nginx... I could go on... One server can only handle so much. Getting linked on drudge would send so many people to a site, and depending on how that site is setup, it may be sending all those people to one server! Anyone hear of getting "slashdotted?" Same difference: http://en.wikipedia.org/wiki/Slashdot_effect It has the same effect as a "distrubuted denial of service" attack. This happens to the best of the web. It happened to AT&T when the iPhone 4 was available for download (http://gizmodo.com/5563909/apple-iphone-4-pre+ordering-is-a-total-disaster) . Same thing brought down Amazon during black friday and some other instances (http://www.betanews.com/article/Amazon-goes-down-for-the-count-twice/121303 0157) Or how about killing the datacenter itself: http://www.theregister.co.uk/2007/11/13/rackspace_texas_truck/ Yeah, sure, blame all these on ColdFusion... btw, I'm being sarcastic :-) Look at how many servers are needed to run high-traffic servers: http://www.paragon-cs.com/wordpress/?p=144 Do you have 10,000 servers for your website? Facebook does. They need them to serve 500 million users. How about 5 CF servers that serve 48,000 concurrent users? Check that out here: http://www.cfwhisperer.com/post.cfm/coldfusion-9-and-48-000-concurrent-users Now, that one we can blame on ColdFusion! Greg Luce wrote: > down. Drudge linked to it I guess and crashed the site and they are blaming > CF. The error is HTTP Error 503. The service is unavailable. Isn't this an > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Order the Adobe Coldfusion Anthology now! http://www.amazon.com/Adobe-Coldfusion-Anthology-Michael-Dinowitz/dp/1430272155/?tag=houseoffusion Archive: http://www.houseoffusion.com/groups/cf-talk/message.cfm/messageid:335830 Subscription: http://www.houseoffusion.com/groups/cf-talk/subscribe.cfm Unsubscribe: http://www.houseoffusion.com/groups/cf-talk/unsubscribe.cfm
