July 24, 2014 | Written by: Franz Freidrich Liebinger Portela
Share this post:
For me, strawberries and chocolate are foods that are good by themselves or together (my apologies to those who have food allergies and can’t enjoy them!). They complement each other in a wonderful way, and I think the same is true of big data and cloud.
It is undeniable that there has been a large shift in the way that most enterprises work, and many of us have seen this shift start with moving some basic workloads to the cloud. For example, a company might put its storage or development and testing systems there to try out this new trend. And this can evolve to more complex and important workloads being moved out to the cloud.
Data, data everywhere
Certain workloads, because of their structure, or lack thereof, make wonderful candidates to be moved onto the cloud. I think big data analytics is one of these workloads that was born to be on the cloud.
We can gather considerable knowledge by analyzing the wealth of unstructured data around whatever subject we are studying. For example, we might be looking at what makes our stock price go up or down, or analyzing how a certain demographic feels about the new flavor of our product, or considering how many people just got hungry because of the strawberries and chocolate analogy I used in my title! All of this data is out there, hidden in between millions of Facebook, Twitter and blog posts.
Every person writes in a different manner though, and the data exists in many different languages. So we need to start by understanding that not all of the data out there is relevant, and in some cases the data could even be misleading (for example, a sarcastic comment may be taken as a falsely positive sentiment if you are not careful). Data analysis can be extremely valuable to help us identify how good our data is and build better predictive models and simulations. It can also allow us to enrich our business intelligence with data that was previously beyond our capacity to work with and therefore enable us to better understand our customers and clients.
But let me take a small step back and talk a little about what big data actually is.
What exactly is big data?
Just like in the case of cloud, “big data” has a fuzzy description. Gartner made a good first pass at trying to describe it in its IT glossary:
Big data is high volume, high velocity, and/or high variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization.
If you are completely lost with this description, here is what it means (to me, at least): it means that we can make use of all the available data (structured and unstructured) from internal data warehouses, databases, forms and other structured systems and enrich it with all of the unstructured data that comes from, let’s say, a social media site (or all of them if you so choose) to gain a better understanding of our customers, services, products and so on.
This sounds simple enough, but when you look at how to use all this data for decision making, you can quickly find yourself overwhelmed by the actual volume of data, the speed at which it changes, the variety of data sources you need to integrate and, finally, the speed at which you need to have the data analyzed so that you can make decisions in a timely manner.
If your infrastructure is not flexible, you will find yourself in a bit of a bind, because the speed of change and volume of data varies (sometimes drastically), and this will affect the speed at which you can convert data into valuable information. This is where cloud makes its entrance. The moment that you enter into the realm of big data, your infrastructure has to scale so that it can support this workload, and here is the link between big data and cloud.
How cloud and big data work together
Cloud provides you with the flexible infrastructure you need to be able to handle fluctuations in workloads effectively (scaling up and down). It also allows for the integration of new workloads in a convenient manner based on changing business needs.
Big data analytics opens up new ways to work with the data around us, and like in the case of strawberries and chocolate, when we pair our big data with cloud, it just works.
If you’d like to learn more about big data analytics solutions, you can refer to this webpage, which lists IBM SaaS offerings for big data.
Have you seen how cloud computing could complement big data analytics? I’d love to hear your thoughts. Continue the conversation below or on Twitter.