Enterprises will adopt a 'data lake' strategy in 2015

Big Data and analytics has come a long way from being just a buzz word to a core element in the business strategies of organisations of all sizes. I feel that 2015 will be the year when Big Data implementations will begin delivering value to the business in terms of improved customer satisfaction, new product development, and optimisation in operational efficiencies.

Here are some key trends that we are foreseeing for the year 2015:

Demand for extreme real-time Big Data analytics: In this time of insta-gratifications, businesses will increasingly rely on real-time insights to make decisions. The processes can even happen in milliseconds. However, this will be a far cry from 'hunches'. Extreme real-time insights will be based on complex algorithms and smart infrastructure backing them.

Growth of data lakes: Data lakes is a concept wherein all the data gets stored in one common massive storage, and users can work on that data source as and when needed. I feel that 2015 will see a lot of enterprises adopting a data lake strategy that will help them eliminate data silos, cut down on costs associated with multiple data stores, and help maintain a single source of truth.

Accelerated adoption of Big Data analytics at the point of decision: In a bid to improve customer experience, businesses need to be able to cut down turnaround times to their minimum. By bringing insights where they are needed most, organisations will be able to take quick actions at the point of decision. This could be in terms of loan sanctions at the point of application to determining real-time fraud.

Big Data analytics processing in the cloud: Big Data Analytics-as-a-service will garner focus and momentum in 2015, as enterprises will increasingly look at cloud-based solutions for salvation in a bid to manage costs. Security has always been a perceived challenge and show-stopper for most organisations. 2015 will see a lot of these fears dissipate on the back of highly-secure and sophisticated infrastructure provided by enterprise cloud players.

Development of deep learning capabilities: Not all data may be relevant for specific context. Deep learning enables algorithms to recognise items of interest in large quantities of data. Not only does this eliminate the need to parse through mountains of data, but will also power real time and at the point of decision analytics.

In addition to the above, we also see the impact of wearable technology and SQL on Hadoop as key shapers of 2015.

For more, please visit our blog.

(Pandit is head of marketing at Singapore-headquartered Big Data startup Aureus Analytics)