Manage Learn to apply best practices and optimize your operations.

Big data and cloud computing: Understanding vendor options

Even though big data and cloud computing have been around for a dozen years, learn why and how it is just now getting commercialized.

Social, mobile, analytic, real time and “big data” — the attributes of our lives and the new requirements for all of the software we use on our computers and mobile devices. Undergirding all that is the cloud. Cloud computing is where data and information come from and they host to our modern communication infrastructure. The cloud has been around in one form or another for more than a dozen years and it is now making huge commercial in-roads.

So, the era of big data and cloud computing and all the rest is upon us and for many people it is all confusing. They’re the lucky ones because they are questioning the paradigm. The unlucky ones are not formulating questions at all, assuming that things will be fine.

Surprisingly, the cloud computing options available today are already well matched to the major themes of need, though some of us might not see it. Big data forms a framework for discussing cloud computing options. Depending on your need you can go into the marketplace and buy infrastructure services from providers like Google and Amazon, Software as a Service (SaaS) from a whole crew of companies starting at Salesforce and proceeding through NetSuite, Cloud9, Jobscience and Zuora — a list that is almost never ending. 

Infrastructure vendors provide hardware and a software stack including your favorite operating system, database, middleware and perhaps your favorite single instance of a conventional application. They’ll do the daily backup and keep all the green lights on but this is really just about moving your data center to another location. You have a great deal of responsibility for managing the system so you need some expertise. 

I recently had an interesting conversation with an infrastructure company about the cost of API calls and how to manage them to save money. That seems a little too down in the weeds for me and while I like saving money, I think my IT deployment ought to be helping me make it and hiding from me the complexities of API calls.  

One of the great benefits of cloud solutions is elasticity, the ability to add users and additional applications as your needs change and to reverse course if your demand slows.  In a conventional situation you buy capacity based on maximum demand and then sit on the investment when demand slows. Cloud computing is better for that.

Today, most vendors have a “have it your way” strategy. They support several forms of cloud computing including single instance solutions that are not much more than your data center in the cloud (affectionately called your mess elsewhere) all the way to multitenant cloud computing practiced by the major innovators of the space such as Salesforce, Zuora and NetSuite to name just three. 

Despite what anyone might think of the single instance cloud deployment, there are reasons for it and people buy it. I’d advise you to investigate carefully and ask some hard questions of the vendors, as well as yourself before selecting a solution.

Very often you can find what you are looking for from one of the established cloud vendors and there is little reason for building your own private cloud, but that doesn’t mean you shouldn’t. Regulations and corporate policies can rule out some kinds of cloud computing which is why so many approaches thrive.


Another kind of cloud computing, which we don’t think about much but which is very important in a big data world doesn’t even have a name that I am aware of. When you send a digital video demo of your product to a YouTube or Vimeo channel, you are sending a big chunk of data to a third party to manage it for you. 

We are increasingly depending on third parties to do this sort of thing and while we may keep a copy on a server behind the firewall, the third party is responsible for making the video or other content available to the world. Most companies link back to their Web properties from these third parties too. 

This is a convenient situation. You get storage and bandwidth to push your video to millions of people and you probably also get analytics to capture information about the video’s use. Keep in mind that your cloud channel still needs to be managed. You need to be able to add and delete content and to analyze its use. Nothing is completely free.

Dig Deeper on Big data, machine learning and AI

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.