With the advent of the ‘internet of things’ and its data gathering abilities, our customers have ever-increasing amounts of information to analyse, parse, and process. I believe the term Big Data will soon lose its meaning, as extremely large datasets will become the rule, not the exception. A prime example of this is the Colorado School of Mines. In a mere six years, they outgrew their 23 Teraflop supercomputer and upgraded to a 155 Teraflop supercomputer (one Teraflop is equivalent to one trillion floating point operations performed in one second). The future is bright as wide ranging effects, trends, and correlations will become clearer as analysts look at bigger pictures.
The future and its challenges can also be daunting. Processing huge amounts of information takes a lot of power. Hardware and software must continue to improve, and I believe they will. The Bureau of Labor Statistics reported in 2012 there were over 1 million innovative and talented people working in software development. Developers have always had an incentive to optimise source code, to allow customers to utilise their hardware fully and work quickly and efficiently. Big Data makes this efficiency a necessity, and applications must be able to handle these large sets of data. Our customers are ambitious, and their demands for functionality must be met and exceeded to allow them to make the most of the data they acquire. As a software development company specialising in data visualisation, it is an exciting prospect.
Managing large datasets requires numerous considerations for our customers too. There is a lot to be learned from seeing the big picture, but what they gather from looking at the details will remain important. To be able to see the sum of forces acting on a system when examining these large datasets will allow customers to trade correlations for assumptions in their detailed analysis. It is therefore critical that customers are provided with the tools necessary to manage the data quickly and efficiently, regardless of the scope of the project.
One may argue the most important element in managing Big Data is multithreaded functionality, combined with an abundant supply of tools and features permitting users to manipulate and evaluate. I believe those come second to usability.
Usability is one of the most undervalued components of technical applications. It does not matter how optimised the program may be. The numerous pieces of functionality and features are rendered useless when a user is unable to access those tools effortlessly. The key here is to ensure workflows are streamlined for users as they transition from the raw data to the modelled end result. As these datasets continue to increase in size, customers will demand more intuitive products in the applications they use for work and research as they do not have time to battle with a poorly designed application. For this reason the user interface is the most important aspect of data visualisation.
The gains from increasing program usability are shared among all parties involved. The user will have a lessened disutility of labour and will find enjoyment in using the program. Frustrations will be lessened for both the user and their employer as increased productivity is a direct result of easy to use products. And finally, the company providing the software will receive increased brand loyalty and customer satisfaction, resulting in increased sales and revenue. An intuitive user experience is a requisite for any application’s continued success.
It is clear datasets are only increasing in size. To meet the demands, we will continue developing efficient visualisation software. In doing so, users will quickly transition from raw data to modelled results in order to make associations and improve their understanding of the data. Golden Software’s applications will continue to grow in scope and functionality, as we expect our customers to demand tools that will enable them to easily capture the value in the information they gather.
