Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Suppose that a hypotetical webapp keeps one connections per user (eg: chat app using comet, long pooling, etc...). How can i load balance such thing to scale beyond the connection limit of a single machine (C10[0]K) ?

I can see round robin DNS as a solution, would like to know what else would be possible...

share|improve this question

migrated from May 24 '11 at 16:16

This question came from our site for professional and enthusiast programmers.

Are there not a million dupes of this question? – Lightness Races in Orbit May 24 '11 at 14:46
I also thought so, but couldn't find... could you point some? – Julio - AWS DevRel May 25 '11 at 13:16
Look to the right. – Lightness Races in Orbit May 25 '11 at 13:18

The far extreme of this is Geographic DNS distributing discrete round-robin IP address pools which in turn point to clustered hardware load-balancers using Virtual IP to handle failover. I say hardware since the big hardware load-balancers are designed for truly massive connection-counts.

share|improve this answer
That's what i think too... do you have an idea of how much "massive" this count is? – Julio - AWS DevRel May 25 '11 at 13:17

dns load balancing will do but you don't have much control over it - better use a load balancer.

share|improve this answer
The problem is what to do when you exceed the capacity of a single load balancer. – Julio - AWS DevRel May 25 '11 at 13:16
have two or more load balancers handling the traffic, these can as well be mixed with dns round robin – silviud May 25 '11 at 15:01

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.