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

First a bit of background - We have a full redundant setup for a client consisting of 2 web servers (win 2008), 2 database servers (SQL server - one active, one passive), an SSL load balancer and a firewall. These servers host multiple websites - with more coming down the line. The current visits to the server is around 200k/m with bandwidth hitting just over 250GB/m.

Everything has worked flawless for the customer and we have had zero downtime. They love the setup but hate the price (circa £70k/yr).

Key points they have to have are: Fully redundant setup - they can't have downtime. They want to be able to host multiple websites each with their own SSLs and have the option to easily add new domains with SSLs in the future i.e. continue to use it as their own server. Much Lower costs.

They are asking about a virtualized/cloud setup to reduce costs. Is this the right path to choose? Personally, I think it could end up costing them as much in the long term but would like to get your take on this.

Thanks for your help.

share|improve this question
Personally, I found the blog post from dotcloud about EC2 and the outage in April very insightful. Using cloud services will just work very differently than using datacenter services and as such will show different failure and performance characteristics, thus requiring completely different approaches to HA. – the-wabbit Jun 17 '11 at 11:06
up vote 3 down vote accepted

There's a lot to be said for sticking with things that work - I'm pleased you're maintaining such high availability, it's not that easy to achieve. While you can maintain the same levels of availability using 'cloud' services you really need to understand the cost to their business of outages and then compare that to the potential cost savings of moving to 'cloud' services plus the cost of the move - only then will you be able to judge if the move is worth the effort. Generally speaking it makes more sense to build new platforms on virtualised/cloud systems than to move existing ones, especially where service availability is key.

In terms of moving to a self-owned virtualised platform I'd say that the benefits to this particular setup would be minimal - you'd need 2 physical servers for resilience and you only have 4 today, the move would work I'm sure but not save you too much, if anything.

There's nothing wrong with looking for better options but when you've got a rig you're happy with staying-put is often the right decision for today.

share|improve this answer
Thanks for the feedback Chopper. I agree with your points too. I just wanted to double check I wasn't overlooking an obvious setup that would have helped tick all the boxes (without the cons!). – Richard Reddy Jun 17 '11 at 11:00
Understand, and as I mention you should indeed be looking around periodically to find better things, and don't just take my word for this, listen to other opinions if you get them, it's just not in our nature to accept that what we've built already might be the best for now :) – Chopper3 Jun 17 '11 at 11:03

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.