Could you elaborate more about why you needed to spend $10K up front? And how you take advantage of your one giant server?
Are you using Xen to virtualize a bunch of servers on it?
I've always figured colocation + System Admin ends up costing more than a cloud solution. Especially when you need to have enough redundancy so that you don't fall over if a single server fails. In the cloud if a single server fails, it's their problem. In physical space it's your problem or your system admin's problem. How much time can you spare?
My wording wasn't too clear, sorry. I didn't mean I have spent that money, just that my planned hardware buildout will cost somewhere around $10k to start. Our single server was about $2k to put together 1.5 years ago. We use it for everything - mysql, redis, pool of unicorns, nginx proxying to unicorn pool and a handful of background workers (resque). I have already dealt with drive failures, but I loaded up the 8 hotswappable bays with drives, so it wasn't too scary. However, I am scared for other hardware failure, which is why I want some more machines.
I am not using Xen. I can see the benefit for some stuff, but I don't believe it's beneficial for my use case. However, I will make that decision in a more informed manner when I am actually in a position to expand our hardware (money and need).
I don't think the colocation costs more. Seriously, I have had great use out of my current server. I figure with a colocated set of 4 servers, $10k initial expenditure + $300 a month for 2 years is $17200, or I can go with rackspace and pay $3000 a month for 2 years, and be $72,000 in the hole. Oh, and after that two years, I don't even have spare hardware to play with!
Sorry to jump in, but are there cloud solutions that don't require sysadmin time? All the ones I have looked at give you a "bare" VM or set of VMs, which you then need to configure for your application.
Are you using Xen to virtualize a bunch of servers on it?
I've always figured colocation + System Admin ends up costing more than a cloud solution. Especially when you need to have enough redundancy so that you don't fall over if a single server fails. In the cloud if a single server fails, it's their problem. In physical space it's your problem or your system admin's problem. How much time can you spare?