What is the Azure Cloud?Alex Fields
It would be quite literally impossible to list all of the Azure-based products and their corresponding feature sets, since the list would be out of date as soon as I hit “Publish.” Microsoft is constantly releasing new features into Azure everyday–some are all new products, while others build on existing ones. Therefore, we have to break it down in terms of solutions. One broad-stroke way to do that quickly is to look at the various cloud service models.
I find that most of us in the SMB space tend to think of Microsoft’s Azure first of all as an IaaS product (Infrastructure-as-a-Service). You click a button, and get a virtual machine–all the hardware is abstracted from your view, and you do not have to manage it. While that characterization is true and accurate, the full truth is, Azure has products in every major type of cloud service offering, including Platform-as-a-Service (PaaS) and Software-as-a-Service (SaaS).
For example you can also develop and deploy your own applications on top of Azure’s database and mobile services–in this case you as the customer are responsible only for the data and apps that ride on top of Azure’s PaaS stack. Click a button, get a database management layer (and everything underneath it).
Something like Azure Active Directory, on the other hand, is best classified as SaaS–in this case the entire stack is taken care of by Microsoft (the service provider). With just a few clicks to provision a new directory and assign applications, administrators can have their users logging into new services in no time, with zero additional hardware, operating systems, middleware or other software necessary. Click a button, get a full application or service, all the way down to user login capability. Office 365 is another example of SaaS.
As already mentioned, Infrastructure-as-a-Service includes things like Azure Virtual Networks and Azure Virtual Machines. I would also add Azure Backup Vaults and Azure Site Recovery to this list. What it means to the SMB customer is very simple: you no longer have to be in the datacenter business. With IaaS, hardware is all owned and managed by Microsoft. You just take care of your connection to the cloud, and the virtual infrastructure you’ve deployed there (likely with the help of a Microsoft partner).
How I like to explain Azure
Azure is basically a storefront. What makes it a “cloudy” storefront is simply the fact that its products can be deployed quickly–automated and on-demand–with no major capital expenditures required (well, maybe some time commitment / labor to get started).
At the end of the day, the Azure cloud (or any cloud) is just more tools for the tool belt. Well, except that you do not have to own the belt–you just rent the tools that you need when you need them. Perhaps the cloud could best be characterized as a… public… tool… library… belt? Sorry–not my day for making up analogies.
As with any product, adoption should be driven first and foremost by business objectives. It’s about what you need to get done in your business, and why. Market forces, industry trends, security, compliance requirements and so forth can also have a major influence on defining your business objectives.
Here are some of the basic trends that are driving adoption of cloud-based service models, and which I think also strongly influence the design of modern-day business objectives, even for small-to-mid-sized businesses:
- High availability / lower tolerance for downtime
- Mobility / modern work lifestyle
- Quicker time to market / on-demand deployment
- Reduced management requirements
- Elasticity (scalability up and down)
It is important to note that we have on-premises tool sets that can line up with the above to fulfill related business objectives also. It’s just that cloud and these trends were born together and grew up together, so in some cases, certain problems find a “more natural home” in native cloud solutions.
Even so, the most important thing is to start with the end in mind (as Stephen Covey would say)–define that end as clearly as possible before proceeding to a solution. I can help you understand what Azure and other technologies are capable of, but you need to have a focused set of objectives that will act, ultimately, as the litmus test–do I or do I not need this tool (or one like it)?