Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

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

some time ago a friend showed me his concept of a virtualized web application:

  • 2 Apaches - load-balanced via IPVS/LVS
  • 2 MySQL - HA via DRBD and Heartbeat

The virtualized machines all run under the same hypervisor on the same host. So I was wondering if this setup is reasonable because I see no advantage.

  • Host is SPOF
  • Load-Balancing via IPVS? Host CPU is the bottleneck anyway or can 2 virtualized CPUs handle more interrupts than the Host CPU alone?
  • MySQL HA? Isn't it faster for the hypervisor restarting a crashed guest than failover via DRBD etc...?

I don't see any advantage in network load-balancing under the same hypervisor since network I/O between guests is mapped memory anyways and therefore fast.

Any ideas, suggestions?

For me it is just overhead and single, beefier guests make more sense...

share|improve this question
Unless you have an imminent migration to more hosts or very odd isolation requirements (like 2 sets of completely seperate human administrators), I see little use to this scenario. – Joris Sep 8 '10 at 8:20

Agreed on every point. If you had a hypervisor cluster, then it might have made sense, especially with a policy to always keep the guests on different hosts

share|improve this answer

Like dyasny already said, the scenario does not make sense at all unless there's a hypervisor cluster. For a single host that kind of hoopla is just plain stupid.

share|improve this answer

I think this configuration is only for testing and learning purpose.

share|improve this answer

Your Answer


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