november 07

Keys to data management policies
Jim Damoulakis, CTO of GlassHouse Technologies, writes about "Moving beyond tiered storage"

http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9004829

Spending the past week at the Storage Networking World conference in Orlando has confirmed for me what we have begun to see among our customers: Companies are becoming more focused on data retention policy. There is a subtle shift beyond tiered storage -- does an application belong on Tier 1 or 2 -- to considering the significant challenge of managing and purging retained data, i.e., archival strategy. As companies begin to delve into these issues, there comes a point when they must reckon with the distinctions between archiving, records management and data retention and decide on how far they want to or need to proceed.

First, what is an archive? Well, at a minimum, there is a growing consensus of what it is not: It is not simply data that is stored away for a long time. Anyone who spends even a small amount of time planning an archiving strategy realizes very quickly that a traditional backup does not make an acceptable archive. To an archivist, the term traditionally implies long-term preservation and availability of documents based on their intrinsic historical interest. In business, we think of archiving of information required for regulatory or business purposes, e.g., records management. For years, records management departments have established and managed the handling and retention policies for paper business records, and it is entirely logical that this function be extended to records that exist in electronic form. The problem is that most applications create, manage and store data, not records. Mapping records to applications data is not trivial. It is a largely manual process that requires the cooperation of individuals with solid business, legal, regulatory, application and data storage knowledge. Then there is the challenge of the actual extraction of data and its operational impact.

However, regulatory compliance and the increasing risks of litigation are forcing organizations to come to grips with the problem. In the long run, applications must be designed to tag relevant data according to records management policy needs. Until then, it is critical to clearly define data retention policies and to begin the unavoidable process of culling through applications and data and applying those policies.

Jim Damoulakis is chief technology officer of GlassHouse Technologies Inc., a leading provider of independent storage services. He can be reached at jimd@glasshouse.com