Overcoming dark data requires that companies shine a light on it, Raheja said. “The first step is to find where dark data resides and catalog it. Then it needs to be checked for redundancies and captured in a unified model. From there, the dark data’s lineage can be mapped back to its originating systems and assigned governance rules. Cataloging and connectivity tools will turn dark data from a liability into an asset in 2022.”
This will also include development of “single-pane-of-glass views for all sensitive data throughout the enterprise, with AI and robotic process automation—rather than manual efforts—for comprehensive data discovery,” said Sharma. “That means agentless, cloud-based solutions that can conduct intelligent scans with minimal configuration.”
DATA MIGRATIONS AHEAD
With the rise of cloud continuing, the coming year will be defined by an acceleration of data from on-premise systems to cloud, or even from one cloud to another. “We recently did a survey that showed data migration ranked as number one for IT leaders’ biggest hurdles to cloud adoption,” said Jozef de Vries, senior vice president with EDB. “To take advantage of the benefits of cloud, IT and data managers both need to focus on reskilling teams to learn these complex cloud environments. This often poses a challenge that database administrators used to on-prem may not be able to meet. Considerations like a cloud center of excellence can help support these processes across the business and guide implementation strategies for long-term success.”
WEAVING DATA FABRICS
Data fabrics—architectural app-roaches in which capabilities are evenly spread across multiple platforms, clouds, and on-premise environments, accessible via multiple endpoints—have been forming for several years and are expected to gain even greater traction in the year ahead. The need for a fabric approach is urgent, said Tapan Patel, senior manager for data management at SAS. With the growing complexity of data, organizations continue to struggle in delivering connected and consistent data across on-prem, public cloud, and multi-cloud sources, he said. “At the same time, they face the challenge of supporting different use cases or different types of users—users that need trusted, consistent, and real-time data to build apps or deliver insights without creating more silos.” Data fabrics “can deliver connected and consistent data for a variety of use cases around customer intelligence, IoT, streaming analytics, and machine learning,” said Patel.
To build out a well-functioning data fabric, data managers will need to focus on “ingesting, transforming, and cleansing data through pipelines in a governed manner, automating certain processes and steps in the DataOps cycle, and securely pushing down logic and processing to various data platforms,” said Patel. “Ensuring trust and transparency in data and analytics is dependent on having the right level of data governance, data catalog capabilities, and strong metadata.”
COMPOSABLE ANALYTICS
The year ahead will see more attention focused on developing agile analytics-driven applications that can be quickly adapted and configured for fast-changing business scenarios and environments, from edge systems to the enterprise cloud. Ultimately, business end users will see increasing capabilities to create these analytic applications on the fly in a self-service fashion. Composable analytics creates business and IT alignment, helping to break down data silos that hold back companies from innovating with data, said Roman Stanek, CEO of GoodData. “Composable data and analytics gives more people access to insights and empowers even those who are not versed in SQL to customize and utilize data insights for business growth and innovation. This creates a more agile and efficient operational process, allowing organizations to respond to change in a dynamic way.”
Such an architecture is built on a modular philosophy and can enable an organization to spin up new applications and connect to new ecosystems quickly and easily, said Stanek. “With composable analytics, enterprises can become more dynamic and agile, develop new applications, and quickly connect to new ecosystems.