One of the overarching themes surrounding digital transformation is the need to get infrastructure out of the way of data and application productivity. Whether the mechanism is automation, devops or cloud computing, the idea is to give users greater control over both their data and the tools needed to make use of it.
A key element in this strategy is self-service. Too often, data-driven initiatives are stifled from the get-go because the small team of experts who know how to build and provision resources is swamped with projects. It takes a lot of clout to put infrastructure in place, and usually, by the time the process is complete, the opportunity it was meant to capitalize on has been lost.
Self-service puts the power of infrastructure into the hands of those who can make the best use of it, but it can lead to problems as well. Resources may be infinite in the cloud, but they still come at a cost, so too many people architecting their own parochial environments can quickly lead to duplication, resource isolation and poor data integration.
According to Damon Edwards, co-founder of automation developer Rundeck, self-service only works if it is part of a fundamental shift in IT provisioning and operations. In a Q&A with InfoQ at the recent DevOps Enterprise Summit in London, Edwards noted that it is heresy in most organizations for developers to define, let alone execute, operational procedures, but this is one of the founding principles behind self-service operations. In an ongoing continuous integration/continuous deployment (CI/CD) setting, the lines between definition, execution and control are blurred, so responsibility for these three steps must flow to where labor capacity and agility are greatest, or the entire process bogs down.
Indeed, companies that have empowered knowledge workers with greater control over data are already seeing dramatic improvements in productivity and revenue growth, says Oracle’s Dave Mommen. Data-dependent operations like marketing are among the leading candidates for self-service, with some early adopters seeing a 56 percent greater return on marketing investments and 10-fold jumps in revenue. By removing dependencies on IT, workers become empowered to leverage multiple data sources, enhance understanding through visualization and other means and demonstrate the real-world business impacts of operations.
Self-service in traditional data operations is certainly helpful, but it is essential when dealing with the volume and disparity of data generated by the Internet of Things (IoT). Cloudera recently launched the Data Science Workbench designed to give data scientists and others involved in large-scale analytics a means to accumulate and share the necessary data sets quickly and securely. The platform integrates into deep learning frameworks like Intel’s BigDL library for Apache Spark, and it embeds programming languages like Python and Scale directly into its browser-based interface. This gives scientists the ability to create highly customized project environments while maintaining support for Hadoop authentication and governance.
Still, with such large amounts of data in play, it can be difficult for self-service users to ensure that the data they are compiling is both accurate and germane to their objectives. Stephanie McReynolds, VP of marketing at data management firm Alation, advises organizations to incorporate data curation as a key element of any self-service data environment. While most analytics engines run data through multiple processes to establish context and meaning, curation goes a step further by tracking the social networking bonds that arise as data sets are re-used. This produces a high degree of trust in data that simple documentation cannot achieve, ultimately producing a one-stop shop for insights and data knowledge to drive better decision-making.
Despite what people say about dealing with impersonal, menu-driven systems, the fact is self-service is taking over many facets of our daily lives, including shopping, customer service and communications. So it’s no surprise that it would prove equally popular in the workplace.
As with all enterprise initiatives, however, self-service should be driven by outcomes, not technology. By focusing on improved performance, greater productivity and increased data values, the enterprise should be able to leverage self-service for the greater good, not simply give individuals greater autonomy over shared infrastructure.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.