You’ve likely heard of software as a service (SaaS) or infrastructure as a service (IaaS). Big data as a service (BDaaS) follows in the footsteps of many other X-as-a-service technologies, using cloud computing to provide something that was previously available only to large enterprises to anyone on a pay-as-you-go basis.

Like those other services, BDaaS has the potential to open up space for smaller companies. Now, those interested in what they can learn when using big data and analytics can find out, even if that access had been previously unavailable to them due to cost or other resource constraints.

What is big data?

Before delving into BDaaS and how it works, let’s start with the key concept in the phrase: Big Data. What is it?

First, ‘data’. In his 2019 book Big Data, Naseer Rahim defines data as ‘information that has been translated into a form that is efficient for movement or processing.’ Today, that means ‘data is information converted into digital binary form.’

Data can be organized in three ways:

• Structured—something like a relational database
• Semi-structured—pairs in a database program
• Un-structured—unprocessed emails, text files, or logs

Big data, then, is the ‘large volume of data […] that inundates a business on a day-to-day basis.’

Big data has also been described using the so-called ‘five V’s’:

• Volume—these are big data sets and have associated challenges
• Variety—there are many different types of data in these data sets
• Velocity—the data comes in at different rates and the system needs to be able to account for that
• Value—all the data in the world isn’t worth much if it can’t be used
• Veracity—the trustworthiness of the data and the integrity and security of its transmission

What is BDaaS?

So, what is big data as a service? BDaaS is the integration of big data and cloud computing, which offers the same ‘easy pay-as-you-go model, offering scalability and availability,’ as the authors of a 2016 paper describe, as other cloud services. BDaaS builds on previous services, including:

• IaaS—Infrastructure as a Service, giving a user access to ‘storage, processing power, and virtual machines’ without having to operate them
• PaaS—Platform as a Service, building on infrastructure to help a user to develop an application
• SaaS—Software as a Service, delivering an app to a consumer that runs somewhere else

The three build, one on top of the other. BDaaS involves each of these in different ways, from the storage required to keep the data, to the platforms on which to make it accessible, and the software used to analyse it.

How does it work?

Since big data requires ‘scalability, fault tolerance and availability,’ cloud computing could be a good fit. BDaaS can provide ‘available, scalable and fault tolerant’ access to your data and the tools that you need to analyse it. Plus, rather than investing the necessary resources to set up your own Hadoop framework, for instance, you can have someone else provide it for you.

Big data as a service allows you to do whatever analytics or storage task you need to, but in an on-demand, pay-as-you-go fashion:

• If a hospitality company wants to collect customer data, analyse it, and then identify problems or gauge customer interest in a given offer, it can.
• If a hospital wants to provide doctors with the ability to quickly analyse patient records, health plans, etc., it can.
• If a retailer wants to predict trends, understand customer needs, or recommend new products to their customers when they need them, it can as well.

And none of those entities need to run their own data centre to do so.

Final Thoughts

While big data may have business value, it is also complex, and approaches that might have worked on data that was small or mid-sized in structure may not cut it here. For a given company, harnessing the value in big data might mean using data that you already have, such as that from machine sensors or product RFID chips. It could also mean gathering the same data that your competitors are, or looking at social media data. Whatever the case, analytics are necessary to understand how to derive value for a business, and you need to have the computing power to run them.

With all of this in mind, many small or mi-sized enterprises may not have the capital they would need to do all of the necessary IT in-house, and many large companies may simply not want to. Because of this, and the desire to not lose out on all the potential value that big data holds, BDaaS may well be worth considering for a whole range of companies.