Everywhere you look nowadays there is a reference to “The Cloud” yet there is no real clear definition as to WHAT “The Cloud” actually is.  You see references to public Clouds, private Clouds, hybrid Clouds … you name it and there is probably the “Cloud” moniker attached to it.  So just what IS “The Cloud”, anyway?

When you come right down to it, most people would equate the Cloud with being a set of accessible services that is not hosted on in-house hardware.  And it just explodes in all directions from that vague description.  And if the services are hosted elsewhere then what on earth is a “private Cloud”???  I can understand why people are confused and headaches start when the discussions starts!

Going back to the just-mentioned vague description, “Cloud” really refers to compute resources that you make use of that are not hosted on your own equipment or maintained in-house.  And let’s add that you should be able to expand or contract the services consumed “at will” (add capability, reduce capability) without giving any thought to the mechanics of doing so.  Yes, there are all sorts of exceptions to that particular general statement but lets agree to use it as the basis for this discussion going forward.

So where does that leave the average small business?  Is it possible to have ALL of your resources out in the Cloud and not have your own servers?  The answer is a qualified “maybe” because in the end it really depends on your line of business applications.  It is a given that “generic” applications like email and websites can easily be Cloud based; there has been a roaring business in that kind of thing since long before the Cloud moniker was bandied about.  And there are more specific applications that are or can be cloud based such as CRM (Salesforce, Sugar) and collaboration (SharePoint through Office365 and other vendors) just to name a few.  But the real roadblock for most organizations are the specific line of business applications that drive their business.  Accounting applications are the big hold back in most cases and/or line of business apps that are bound to an organization’s specific industry or market niche.

We (itgroove) deal with a few accounting firms in our SMB practice and these firms are perfect examples of users of specialized line of business applications that simply don’t exist in the cloud.  Yes, these firms can “cloudify” some of their applications such as email but they hit a real solid wall at what can be done with their public-accounting specific applications.  Applications that still rely on “fat clients” and “client/server” data models don’t lend themselves to a cloud model in any way.  That is not to say that a firm that uses these types of applications cannot move everything they use into a “hosted” environment (note the term “hosted” and not “cloud”) because they can and do all the time but a hosted environment is a much different animal from a Cloud environment.

A hosted solution usually refers to a situation where a firm’s applications and data live inside rented equipment (or a rented virtual environment) in someone else’s datacenter.  In the case of “fat client” applications there may be a Citrix XenApp or Windows Remote Desktop Server (RDS) set up that allows for full fat client operation (as far as the user is concerned) without having massive amounts of data moving across the WAN connection.  In essence, you just move the applications that you would have in-house on your own iron to someone else’s iron in a remote datacenter.  The XenApp or RDS servers then allow your users to run those apps in a remote session on the XenApp or RDS server.  This would be no different than the same process taking place on XenApp or RDS servers in-house, it’s just a matter of WHERE the servers reside.  Obviously, there are other hosted app scenarios that may not involve XenApp or RDS but I’m sure you get the picture.

A Cloud solution, on the other hand, is usually a bit more abstract in the sense that you just connect to the “service”, usually with your web browser, and just start working.  There is no sense of connecting to a server as in the XenApp or RDS scenario,  its just a service.  And with a Cloud service you are never really aware of where your data resides, its just “out there” somewhere. With a hosted solution its a pretty good bet that you know where your data lives (or your tech people do, at the very least).  There are many variations on this scenario, of course, as you look at things such as “platform as a service” (PaaS) and “software as a service” (SaaS) but, again, I’m sure you get the gist of it.  Cloud solutions are just that, “cloudy” and somewhat amorphous whereas hosted solutions are pretty cut and dried.

And that brings us to another point about the Cloud.  In many cases you simply have no idea where your data lives.  In the case of some Cloud providers your data can be constantly shifting around between various global data centers depending on load conditions.  If your operate under some form of privacy law you may actually be legally prohibited from using certain Cloud resources because your data is stored outside of the country of origin.  A concrete example of this would be a Canadian law firm using a Cloud service that stored sensitive client data outside of the country which is an absolute no-no under Canadian federal law (and many Province’s, too).

One other aspect of the Cloud that no one seems to discuss is the fact that there really is no such thing a s a free lunch.  Yes, in many cases, a Cloud-based service or application will look to be much cheaper than the same service or application running on your own in-house iron but you have to look at ALL the costs. When services and applications are moved out to the Cloud the network load generally moves from the local LAN to WAN which usually means you need to beef up your WAN connections which in turn means you have to spend more (sometimes much more) on said connections.  As an example, we have a client that has offices locally in Victoria along with a number of regional offices across Canada and the US.  When they moved from their own internal Exchange servers (there were local Exchange servers in a number of their offices) to a Cloud-based Exchange service they were shocked to see how much their WAN bandwidth jumped (skyrocketed would be a better term).  Their E10 connection locally was almost totally consumed with traffic to/from the Cloud service.  Now, to be fair to all, they made some configuration changes to both Exchange and their Outlook clients which reduced the traffic on the WAN but overall WAN traffic was still considerably higher than before the move to the Cloud service (and this could have also happened with a “hosted” service, just to be clear). 

There may be other costs as well such as subscription fees, use fees, storage fees, the list goes on.  In general, you really just move the dollar costs around going from a mixture of CapEx/OpEx to OpEX only but the dollars, ultimately, still go out the door.  It is even possible over a set period of time that a Cloud solution (or hosted) may actually cost you more than the same solution in-house!  Again, you have to analyze all the costs of the solutions at hand before you can make the decision as to whether or not the Cloud makes sense for your organization.

So, with all of this in mind, what is a “private” Cloud?  There are many possible definitions but I’d suggest that if you are running internal applications on virtualized gear then you can say you have a private Cloud.  Why, you ask?  Because you can generally dial up or dial back the resources you assign to virtual machines/apps just like you can with Cloud apps.  I realize this is a very “thin” definition but it does work for purposes of this discussion.

Small Business can definitely take advantage of the Cloud (and private Cloud) with a little judicious planning.  Office365 is a great Cloud offering that can work really well for small biz (in fact it is the cornerstone of Microsoft’s “hybrid” Cloud offering for small biz, Small Business Server 2011 Essentials).  Other things like SugarCRM, salseforce.com and Google Apps (the paid-for version) also offer big bang for the buck. But you probably won’t be able to migrate your line-of-business app to the Cloud at this point unless your application vendor provides a Cloud offering OR you make the move to a vendor that does have such an offering.  You CAN put your line-of-business apps (and other services) in to your own private Cloud, the tools are available to virtualize almost everything you do and, by extension, give you the ability to dial-up and dial-back resources (and services/applications) at will.

In the end it really doesn’t matter if a resource is local, lives in a private or public Cloud or is hosted.  What does matter is that the resource does the job you need it to do at a cost that makes sense for your business.  There is no one model that works for all resources for all businesses all of the time.  There is no point in pursuing a Cloud resource for an application that just isn’t suited to the Cloud model but, equally, there is also no point in staying rooted where you are simply because that’s “always the way we have done it”.  We are in the midst of a computing revolution thanks to Cloud resources (all types of Clouds) and the emergence of new types of devices that can consume those resources (phones, tablets, you name it).  Smart businesses will look at what ALL of these various resources can do and work with what works best for them without dwelling too much on the semantics of Cloud vs everything else.  And this works just as well for small business as it does for the global multinationals.

The “Cloud” and Small Biz
Tagged on: