Federal agencies cannot develop an effective strategy to migrate workloads to the cloud without having a solid data strategy in place, according to a leading cloud expert at Booz Allen Hamilton.
Cloud migration and implementation activity can’t really be effectively comprehended without a full understanding of an organization’s data architecture and future data extensibility and scalability needs, noted Dan Tucker, vice president at Booz Allen Hamilton, who leads the company’s Digital Platform capability team.
“You almost need the data strategy at the application and portfolio level before you can lay out your cloud strategy,” Tucker told MeriTalk during a discussion about how agency managers can become more “cloud smart.”
Agency managers will need a data strategy that lays out guidance for how to manage data seamlessly across mixed infrastructure as well as how to apply security and governance policies to all data, Tucker noted. The White House’s Cloud Smart strategy, unveiled in September 2018, references the need for agencies to consider data layer security as part of their information technology transformation initiatives.
“To implement a risk-based approach to cloud adoption, agencies should transition to security and protections at the data layer instead of the network and physical infrastructure layers, as well as improve the governance of systems,” the Cloud Smart policy document states.
Looking forward, though, it will be imperative for agency chief data officers to work closely with agency chief information officers and cloud management offices to aid in driving cloud transformation, Tucker said.
Booz Allen is one of the companies working with the Department of Agriculture (USDA) to stand up its IT Modernization Centers of Excellence, specifically in the area of cloud adoption and infrastructure optimization. Currently, the technology and consulting teams working with the USDA are focused on issues such as what should be in a transformation roadmap; what should be in a cloud service catalog, which allow organizations to create and manage catalogs of IT services; and what are the data needs of applications that have been rationalized. Application rationalization is the practice of strategically identifying business applications across an organization to determine which applications should be kept, replaced, retired, or consolidated.
All of this upfront work related to data will help managers make more informed decisions about how to move workloads to the cloud. “We would have to do rework if we looked at the cloud or application architecture before we truly understood the data architecture,” Tucker said.
Understanding the data needs of an organization starts with identifying the current and forecasted business needs of the agency. For example, the Food and Drug Administration’s Tobacco Products Division has seen a tenfold increase in submissions for products because of vaporizers and e-cigarettes. The increase in submissions has impacted the agency’s data storage need.
“When you think about your data strategy, you need to think about data governance, data sources, the extensibility of the data architecture, and data lineage,” Tucker said. The volume, variety, and velocity of data must be considered so managers can determine if an application supporting the data is better suited for a commercial cloud or for a more hybrid approach, Tucker explained.
It is important for agency managers to recognize that cloud transformation is part of digital transformation and that requires a perspective on culture and change management, Tucker advised. The Cloud Smart strategy addresses issues of change management and aims to help agencies adopt cloud solutions that streamline transformation and embrace modern capabilities, which is encouraging, Tucker noted.
Still, there “certainly is a lot of work ahead to flesh out that policy into a practical and actionable plan,” he said.