The Cloud, and RackSpace
// August 11th, 2010 // No Comments » // Other
The last 15+ years of server infrastructure has been pretty incredible to say the least. I’ve been in the field and spent a lot of money on a lot of different hardware and providers. I clearly remember spending tens of thousands on hardware to build critical infrastructure to then colocate it so I could handle rapid growth when it hit. I’ve had dedicated servers, colocated servers, and more recently cloud servers.
During that time there was nothing worse then receiving that 2 a.m. page that a customer’s server was down, and they couldn’t reach a critical web page. The frantic SSH session that failed to connect lead to using an APC MasterSwitch to reboot that outlet and to hope it brought the server back so you could work on it remotely. Once in a blue moon you were so lucky, but all to often you’d have to leave your house to go work on the server. Even worse you may be on vacation and stuck paying $150+ hr for remote hands and eyes to read back what they see so you can advise them how to fix it. Incidents like those quickly convince you to pay for the Remote KVM over IP’s that cost thousands and to start buying built in HP ILO (Integrated Lights Out) and Dell DRAC (Dell Remote Access Card) cards, regardless of cost. If the server had hardware problems it could take even longer to get parts on site to fix the server..
Thankfully that’s starting to become the distant memory as we now have Amazon EC2, RackSpace Cloud Servers, Microsoft Azure, VMware, and oh so many other players emphasizing redundant virtual servers and cloud servers. Being the technology geek that I am I’ve signed up for all of the above solutions above and evaluated them. There are different niches in the market for different needs, but as a general “cloud server” provider RackSpace does a great job.
With RackSpace I got started back when I signed up with SliceHost.com in 2008. They were bought out by RackSpace and provided the technology to power RackSpace’s cloud solutions. Admiring RackSpace’s reputation in the industry for solid customer service I quickly signed up for a cloud server when they introduced them. During that time I’ve tested their functionality to grow and shrink servers, setup automated backups, perform restores, used their iPhone application to remotely manage the server, and their web based console has saved me a few times. These are all standard features that haven’t cost me a dime extra, but have been incredibly convenient to have.
I’ve been using 3rd party monitoring and have experienced excellent performance and uptime for the last couple of years. My first real issue came up on July 4, 2010 at 3:16 AM CST (8:16 UTC). The cloud server went down, and at 3:25 AM CST RackSpace contacted me to inform me of the issue. They were quick to tell me that a chassis swap was necessary and that was completed by 4:14 AM. At which point they immediately notified me that the work was complete. In this case they dealt with a hardware failure while I was sleeping and I never had to leave the house to resolve it.
I think it goes without saying that their service was superb during this issue. Although this is a very pro-RackSpace story it might as well be a very pro “cloud” review. If your servers and solutions are architected well up front you’ll save yourself a lot of time and grief later on. Virtualized servers in the cloud aren’t the right fit for everything, but when the fit is right they can be a life saver. This is just one example of that fact, I could have just as well posted about the ease of scaling up, quickly handling large deployments, providing redundancy throughout the world, and so on. The cloud isn’t the next big thing; the cloud continues to be the current big thing. Its just time for the rest of the marketplace to catch on.
