Over the course of the 2010 year, I’ve had the pleasure of speaking with many owners of web hosting companies around the world. There are a few things (topics) that keep on popping up, one of which is cloud computing and how a hosting company can harness that to make it an actual viable solution for their current client base.
Most of us, if not all, have heard of cloud computing, but from what I have gathered is that very few have a clear understanding as to what that is, what it can do for the company, for the customers and above all, does the hosting company actually need it. Usually, large corporations are early adopters of new technologies because of the larger budgets and the human resources to make a case for it. This helps the firm know in which direction to proceed, technologically speaking. If you are running a smooth operation and overhead is low with not so many employees, you’re ahead of the game. Why break it? Why fix it, if it ain’t broke?
Here is why you should at least take a look at what is out there and understand why this would be a very solid investment into the future of your hosting business. Sure, at this point, no, you won’t need it and if you run a tight ship with top service so the support issues are limited, there may not be a reason at this point to invest into new technologies.
FYI, Cloud computing is for your future, not for the present. Already, you see new pricing features for cloud-based services on a “utility” business model, which are hourly costs for resources instead of a flat rate billing option. The message I received from other owners is that, “I don’t charge per hour, why do I need it?”. Now, you may not need it, but this is a gradual implementation into the industry and has a few very positive points to it. Resources based on the customers needs and not limited to a predefined hosting package. The applications of this business model are still being defined and I suggest that you, as a hosting company, not ignore this. You may be successful now, but the industry is changing, especially now, by more than leaps and bounds. It’s changing so much so that a whole new set of computing jargon is being invented and in my case (old guy), sometimes they are the same as age old acronyms and buzz words, so I’m learning a new language (almost).
A hosting company stepping into the cloud computing world needs to evaluate their business model and apply a new way to run a business. I’ll be adding more business oriented blogs later on, but I wanted to cover a few points of things to think about.
1. Are you big enough to think about cloud hosting as an option?
You can actually invest into cloud technology at any size, however, it really depends on your business model and what your target market is. For example, if you are an enterprise style business, your clients will be few, but pay quite a bit of money for services. Cloud computing inherently consists of failover (HA or High Availability) and most have the option of Hot Migration as well, therefore you will need at least 2 servers for that and a separate data device(s) (without a mutually exclusive data store device, there is no HA (DRBD does not work very well and local storage defeats the purpose), plus whatever other hardware that you wish to attach, ie controller, management devices, load balancing etc.
* I’ll write more on DRBD (Distributed Raid Block Device) in another blog. It does have some promise, but it’s not ready yet for the market to embrace (just yet).
2. Do I need to invest and if so, what am I looking at?
Yes, predominantly, you will need to invest into new hardware, but your new hardware will have better capacity, more energy efficiency on top of cloud computing’s inherent energy saving model and you can also utilise your legacy gear to enhance the new cloud infrastructure that you have built by increasing redundancy and failover.
3. That sounds like a lot to try something that I am not familiar with.
It’s not a lot really, but it’s rather a different way of thinking and a different way to deploy your business model. Cloud computing offers working technology to handle the problems of the past (like fail over and hot migration) which reduces downtime, if any and also it allows you to work more on the business, not trying to keep things up and have many sleepless nights. I’m sure the majority of you have slept in the datacenter at least once in your life! I know I have (a few times). Don’t forget about automation as it is a key feature and will only get better as time goes on. It’s not just clapping your hands and the lights turn on. This goes deeper into network, server, client-side and beyond. I’ll shed some light on this and go deper into detail in future blogs.
4. How much is this all going to cost me?
My suggestion is to ignore price and go for ease of use and functionality. You might want to look at opensource like Openstack or the Nimbus project, but that just creates more work for you as you will need to completely configure and add code for it to work the way you want it. IF you have the resources, then go ahead! If not, then you have the option to look for proprietary IaaS or one of those hybrid companies that have taken an open sourced cloud platform and built proprietary technology around it. The range for minimum deployment is anywhere from a few hundred per month to a few thousand. Again, it all depends on what you are looking for. I can’t even begin to tell you pricing models as there are hundreds of cloud providers out there and most have a completely different pricing models.
Cloud computing does make the difference, but it also creates more micromanagement, so keep that in mind when planning your new business model. If you have anything to add or want to contact me for specific questions, please leave feedback in the comments area.