What is the cloud, anyway?
- Short answer: it’s the place where cloud hosting companies do your computing for you.
- Longer answer: the cloud is the place where the resources of many thousands of computer servers are combined to create the appearance of unlimited numbers of individual computers, each running its own operating system and software services; each fully controlled by its unique account owner. Administrating your little bit of the cloud is, by and large, no different than configuring and maintaining your own physical computer. To someone browsing the Internet, your service will look just like any other.
Have you got stuff you’d like to share with the whole world or with co-workers or clients or customers? You could connect that old, unused computer in your basement directly to the Internet (using a web server software package like Apache), associate it with a web address, and invite the world in. Or you could rent space from a web hosting company so they’ll put your stuff on that old, unused computer in their basement, and invite the world in.
Or, rather than just renting storage space and Internet connectivity, you could rent computing resources from a cloud host, where you, along with many thousands of other customers, will share vast hard drives, network servers and bandwidth, allowing you to provide your stuff to the world through a virtual computer whose size and cost can efficiently grow or shrink according to your changing needs.
What’s in it for you?
- Suppose that old PC in your basement just can’t provide what you’re after, and the machine that will do the job will cost you upwards of $750.
- Suppose your Internet connection goes down from time to time – not to mention your electricity supply – but you want to be sure that your web service availability is uninterrupted.
- Suppose you’re just not the kind of all-star networking guru who’s comfortable creating a properly functioning and secure connection between what’s on your hard drive and the big, bad Internet outside.
- Suppose you’re not sure what your needs might be in six months or a year, and you’d like to avoid the costs of over and under supply.
- Suppose you live a highly mobile lifestyle, leaving you with few opportunities to access and manage the physical computers in your home or office.
- Suppose you work as part of a widely dispersed team. You all need equal and constant access to common documents and resources.
You could probably find solutions to each of these problems individually, but cloud computing offers them all in one convenient package. And a package that, in many cases, will turn out to be a great deal cheaper and simpler.
What can the cloud deliver?
For our purposes, let’s say that there are three broad categories of cloud services: “IaaS” (infrastructure as a Service), “PaaS” (Platform as a Service) and “SaaS” (Software as a Service).
- An IaaS provider offers you an empty virtual computer. You are often given a choice of quickly and easily installing one of a number of well-known operating systems (various flavors of Linux or, perhaps, a couple of recent versions of MS Windows), and then the computer is yours. Practically speaking, you only pay for the time that your virtual computer (or “instance”) is actually running, and for the memory and bandwidth resources you actually use. Well-known IaaS examples include: Amazon Web Services’ EC2 (Elastic Compute Cloud), Rackspace, and Google Compute Engine.
- Rather than a whole “computer,” PaaS providers offer an always-running platform for which you can create (i.e., program) your own applications. These platforms are not always as versatile as IaaS environments, but they will often require less maintenance and are, in some cases at least, cheaper. Examples include: Amazon Web Services’ Elastic Beanstalk, Google App Engine, and Windows Azure Cloud Services.
- SaaS services provide the kinds of specific software functionality that you would normally associate with local PC software packages – like business productivity applications – but on the web. A good illustration of this kind of computing would be the Google Chromebook, a cheap laptop designed to run nothing more than a web browser (Google Chrome), through which web-based software (Google Docs) can be accessed for document editing and storage. Examples include Adobe Creative Cloud, Google Apps, and Microsoft Office 365.
How does it work?
Much of the configuration and control of cloud services is carried out in your web browser, usually through a dashboard-like interface on the provider’s web site. Using this interface, you will set up your account, choose the kind of environment you want, determine the resources you’d like to access, and launch your service.
If you’re running a full IaaS, it’s likely that your day-to-day management will be performed through a secure, text-based terminal session. For users who aren’t familiar with terminal sessions, there might be a rather steep learning curve to consider.
What does it cost?
Light service is often offered free or for as little as a few pennies an hour. At such rates, it’s not hard to see why more and more businesses are choosing cloud computing over local servers. Remember our $750 server? Well add to that the costs for your electrical supply, Internet access, cooling (powerful computers need to be constantly cooled – often by combinations of fan systems and extra air conditioning), and hours and hours of your labor.
However, a few pennies per hour, when multiplied by the 8,760 hours in a year, can come to quite a sum on its own. Before leaping into cloud computing, you should really do a good, old fashioned cost-benefit analysis of your own to make sure that it’s a good match for your particular business model. You should also explore any alternatives that might be out there.
What about the risks?
On the one hand, Big Cloud is very good at providing reliable service. Cloud-based platforms like Amazon offer a high level of redundancy (that is, if one of their servers should fail, your data will already be backed up on hundreds of others, with the switchover happening so quickly as to be undetectable) and don’t suffer much down time. But even they can fail: occasionally, a major power outage could take out an entire data center – as happened to Amazon’s Virginia operation back in 2012 – killing service for thousands of companies.
There is also the possibility that your critical and sensitive data could be intercepted somewhere between your computer and the cloud; that your system integrity might be compromised by an attack from the web; or that any one of the countless workers employed by the cloud providers themselves could go rogue. While such fears are common to all Internet-connected computers, the cloud does present its own built-in vulnerabilities that must be addressed.
Back up your data. Back it up again. Back it up regularly, every day, or at least every week. Back it up to multiple, safe locations (both physically distant from your computer and from your home or office).
And did I mention that you should perform regular backups? You bet I did.
Make sure that you use good passwords for all your sensitive accounts. That means your password should never contain words from the dictionary, names, birth dates, anniversary dates, etc. They should also contain both lower-case and upper-case characters, numbers and other characters, like $, %, and &.
Finally, make sure that you use only public/private key secure connections to your cloud activities (SSL for Linux and PuTTY for Windows).
It ain’t perfect, but at least you will have done your best.
Image Credit: Stan Wiechers