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

Around here the clusters (not hyper-v clusters) are typically configured as n+1, so they are asking me to create a n+1 hyper-v r2 clusters. These will configured with both csv's and live migration and managed via scvmm r2. My thinking is that its a waste in having a node sitting there idle. In my opinion it would be better to have headroom left over for what would traditionally the +1 server spread amongst the N nodes. Anybody have an opinion on this. thanks

share|improve this question
up vote 5 down vote accepted

I'm with you too. If your hot server experiences a failure ALL the machines running on it will have to migrate to the other box, but if you load balance only half will have to migrate. Just make sure you have enough headroom on both boxes for either to fail-over.

share|improve this answer
I'm trying to explain this to my co workers and they are not getting it! thanks – tony roth Apr 23 '10 at 23:23
+1 Plus you'll generally get better performance spreading the load out. – Chris S Dec 2 '10 at 18:36

You are dead on. Problem with a "hot" spare is you never know it is ready to run until something fails. Better to spread the n+1 capacity across all cluster hosts so you know it is ready to run when needed.


share|improve this answer

I agree. That said, I would always go with n+2 - n+1 +1 that you can take down yourself (service pack, patching etc.). At least in business critical environments - there is no reason no to do that ;)

Whether you keep the spares active or just on hot standby is your decision - at the end, using them gives you more CPU power (if you are CPU bound). Not a lot, but it can make applications a little snappier. OTOH keeping them warm makes sure you really CAN run without problems on N systems - especialyl that you are not getting CPU allocation problems later ;)

At the end, a lot runs into "just decide" here, IMHO.

share|improve this answer

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.