It's difficult to suggest how data classification and its tools may evolve into the future. The actual process of data classification will likely remain a largely manual process, though more sophisticated, intelligent and versatile tools may help to automate much of the discovery and organization of corporate information. In addition, the experience gained in early data classification efforts, including the unavoidable mistakes and missteps, may help to foster a closer working relationship between IT and management, making larger and more complex classification projects successful. "I think it's an excellent opportunity for data classification to really bridge the gap between true business requirements and the way that IT works," says John Merryman, senior analyst at GlassHouse Technologies Inc. Eventually, data classification will become an integral part of storage technology.
Practices still building to critical mass
Early adopters are typically the largest organizations, such as defense contractors and life science companies, driven by the serious consequences of compliance and government regulation. But the tools are still evolving, and the practice may take another year or more before it truly emerges as a mandatory strategic focus in the enterprise. "In most environments, it's a 'nice to have,' " says Greg Schulz, senior analyst at the Evaluator Group. "The industry may get along for another six months or another year without it." Other analysts suggest that it may take up to three years before data classification becomes a mandatory practice. However, there is little doubt that business and IT both see data classification as a permanent addition to the enterprise. Even Storage Networking Industry Association (SNIA) has established a data classification task force to spearhead education and define practices. You can learn more about SNIA's efforts here.
Integration into standard products
Analysts generally agree that data classification software has a long way to go. The tools that exist today are typically quite good within their own particular niche, but there are currently no "killer apps" -- no single utility does everything and no single utility works in every environment. For data classification to truly become a mainstream technology, classification-centric features must eventually appear in more standardized data management tools. "Really the most promising space is the development of frameworks and applications that help support custom development of data classification," Merryman says. Some storage insiders suggest that data classification features may ultimately appear embedded in some storage devices. "I see the potential for it to become more integrated with general storage I/O and server processing," notes Schulz. He sees data classification products becoming a more in-line and event-based process, taking action whenever a file is created or changed -- perhaps even becoming part of the operating system itself.
Go to the first part of this article: Introduction
Or skip to the section of interest: