It’s difficult to overstate the impact AI is having on data management. “It’s the result of the sheer volume of available data, unprecedented computing power, and advances in data science,” said Narendra Mulani, chief analytics officer with Accenture Analytics. “AI-powered analytics is already helping organizations unlock value that was previously hidden or out of reach. Enterprises that are putting AI-powered analytics to work are seeing positive impact on revenue, margin, and the customer as well as the employee experience. They’re also outpacing non-adopters. What we expect to see in 2018, and onward, is rapid, large-scale adoption of analytics solutions that are powered by AI to help automate and optimize data management processes with speed and precision.”
Advanced analytics
Technologies such as AI may not even go far enough in helping companies sort through the jungle of data that is sprouting around them. “Many data management software solutions have already started to incorporate artificial intelligence technology into their offerings to identify data, tag it, and discover relationships. While this is a good first step to automating data management, it doesn’t go far enough,” said Ron Agresta, director, product management, for SAS. “Advanced analytics-based profiling can be used to identify problem areas in data, based on discovered context. The system learns to make suggestions and then quickly prepares data to meet business needs. Based on data selection patterns, user-accepted suggestions, and manual transformations, machine learning techniques can improve over time as the actions of groups of users and results from monitored data management processes are combined.”
Hybrid data clouds
There has been an unstoppable wave of data coming from outside the enterprise, much of it born on the cloud. To meet this challenge, there’s growing interest in “data infrastructure adapted to a hybrid model—one where some data may be on-premises, some may be in the cloud, and lots may be outside the enterprise,” said John Hagerty, VP of product management at Oracle. The elastic nature of this hybrid cloud provides organizations with the agility to address these new and voluminous sources of data. He pointed to enabling technologies for hybrid data environments, including streaming data services for data in motion; IoT services to capture external device data; data lakes and block storage for repository of raw, unrefined data; relational databases to store, process, and analyze structured data; data integration services to manipulate data; and analytic services, including advanced analytics and machine learning execution, to interrogate and understand what the data means.
Automation
Overwhelming amounts of data, combined with overwhelming business demand for insights, have made data management untenable. This means greater “data analytics and then operationalizing those insights,” said Dheeraj Remella, director of solutions architecture at VoltDB. “It starts with big data which is used for predictive and prescriptive analytics and unsupervised learning. From there, companies will focus on operationalizing this data and incorporating it into the front end, day-to-day decision making process.”
The ability to “make imperfect or incomplete data usable—and operate across legacy and modern data stores to produce the right insights that not only enable humans to do their job, but automatically feed into other applications, is the game changer,” said Georges Smine, VP of product marketing at Opera Solutions. “The capabilities to extract insights will also rely more on artificial intelligence models, reducing the dependency on data scientists as we know them today.”
Multi-cloud
Cloud has been a part of data environments for several years now, and the year ahead will see an expansion into multi-cloud constellations. However, this is exposing gaps in capabilities. “Organizations are realizing that cloud service providers alone don’t give them all the tools necessary to efficiently manage heterogeneous environments, across hybrid, multi-cloud systems,” said Alex Sakaguchi, senior director of global cloud solutions marketing for Veritas. “As such, they are proactively adopting data management tools that fill this gap, and we see this trend continuing to grow substantially in 2018. This gap will continue to spur companies to embrace their own digital transformations and put a higher value on their data.”