Technologies collide to accelerate the value of data
A coincidence of transformational innovation is emerging in the coming years. Artificial intelligence (AI), the internet of things (IoT) and 5G are disruptive technologies in their own right, but they happen to be rising together and when deployed in conjunction they become something even more powerful to businesses wanting to leverage the full force of data.
Enterprises already have a huge amount of data at their disposal and have begun to manage that data more strategically in recent years. In industries including autonomous driving, life sciences, retail and finance, where the most valuable data sets are particularly large, companies are implementing their first AI production systems to bring new experiences and services to their consumers, and to realise the potential locked in that data.
With wider availability and capabilities for IoT and 5G technology on the horizon, however, AI use-cases are set to multiply rapidly. As creation rates of information spirals through 5G, and IoT brings data storage closer to the users, companies will need to take more advantage of streaming analytics and AI to manage data and deliver ever-richer and faster services to their consumers.
“The application of AI and the challenges of extracting more value from data will move on to steroids in the IoT and 5G era,” says James Coomer, senior vice president for products at DataDirect Networks (DDN), a leading supplier of high-performance data management solutions for customers at scale. “But when you need to apply advanced analytics just to cope with your data volumes and optimise your network streams, how do you build an environment scalable and sophisticated enough to ensure your business is managing its data?”
The application of AI and the challenges of extracting more value from data will move on to steroids in the IoT and 5G era
As data volumes continue to grow, companies face the challenge of extracting value at organisational scale. Managing data is now not just about operations and life cycles, it’s about ensuring data with value isn’t hidden, lost or underutilised.
DDN has been well focused on these challenges. Last year it acquired Tintri, which implements advanced analytics that can drive efficiencies that would be extremely difficult and time consuming for administrators. The technology detects anomalies, inefficiencies and poor policies, and delivers actionable, high-level advice on how to create a more efficient data environment.
So DDN doesn’t just store data; it is also building the mechanisms, tools and software to enable companies to optimise their end-to-end AI process. In the autonomous driving use-case, for example, neural network models are trained using supervised learning techniques and virtual training environments designed to prepare a model for production. Evidently, the development process for a self-driving car needs to be as close to ideal as possible to ensure it is optimally prepared for the road.
“There is a big area of oversight, audit and governance around the AI training and production cycle, and we work closely with a company called Dotscience, which is building this kind of software,” says Dr Coomer. “Within DDN, the focus is also on maximising the movement of that data at extreme scales. We have for 20 years worked at the upper echelons of data challenges of scale, movement, ingest and manipulation, all with predictable, low latencies. That storage engine is crucial for what’s coming.
“It’s tough working out which data has the most value, where it is, where it’s replicated, how you can move it and how best to apply analytics. To do that on an organisational scale with the sizes of data that AI is going to bring about requires new levels of automation in the storage environment.
“DDN delivers simple visibility into your storage and our technology allows us to move towards full automation of data management, which is a necessary step to overcome the challenges of competing in the world of IoT and AI.”
For more information please visit ddn.com