May 31

Managing the data protection continuum
Dependence on specialized management tools exacerbates management issues
by Jim Damoulakis, GlassHouse CTO

http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9000828

We live in a time of growing specialization and compartmentalization, and the world of storage management certainly has not escaped this trend. Increased technological complexity and the need for differentiation have significant implications in areas like staffing – a single person can’t be an expert in everything – as well as management.

The dependence on specialized management tools for each vendor’s products (and often even among a single vendor’s products) exacerbates this problem. Effective management requires aggregating input from multiple data sources to understand trends and make decisions, and unfortunately this is stifled by the continued lack of interoperability among storage products.

Within the realm of data protection, the problem becomes even more evident as we see new technologies introduced. Long gone are the days when “data protection” equated to “backup.” Not only has the domain of data protection activities broadened to include functions like archiving and security, but the range of technical alternatives available to perform traditional functions like backup has expanded, as well.

For example, continuous data protection (CDP) – the concept of any-point-in-time recovery – is finally beginning to evolve beyond early adopter status. I think that this is good news, as this technology affords the opportunity to provide much higher degrees of recoverability across a broader collection of data than ever before. However, it also underscores the management compartmentalization problem.

There exists a data protection continuum that is currently managed entirely in a non-integrated fashion. Ranked from shortest to longest recovery point objective, these data protection options include: replication, CDP, split mirrors and snapshots, traditional backup to disk or tape, and archiving. Many environments employ several of these technologies, and from a data management policy perspective, there should be a common reporting and enforcement mechanism to ensure that data of a given category or class is being managed in accordance with business requirements.

Sadly, this isn’t the case. Each of these functions is handled discretely so there is no easy way for a manager to actually determine a given application’s recovery status. Addressing this problem typically requires customized tools that collect data from the various vendor-specific sources. What is really needed is the evolution of today’s backup reporting tools to application-aware data protection reporting and management tools.

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