Lifecycle management drives data management's evolution from art to science.

Data management, as we know it today, is evolving from art to science as technology is catching up to business demand and the value of data continues to increase. Currently, though, the elements of data management persist as flexible art forms.

Historically, the need for what came to be called data management started more than 30 years ago, as business demand for storing and recovering data from a disaster became more and more prevalent. At that time, businesses were focused on simply moving data off of mainframes and onto storage devices (primarily tape drives) in an effort to secure the information in the event of power loss or data corruption.

This movement marked the birth of the art of data management. Backing up data from one system to another, or even from one location to another, quickly became a challenge only attempted by skilled network architects and engineers. At the time, there was no single solution that would take the complexity out of managing the distribution of this data or the process itself.

[FIGURE 1 OMITTED]

As the value and quantity of this data continued to increase and data became a mission-critical asset, more stringent and complex data management requirements were created by businesses. Simply storing "old" data off site where it could be retrieved and available days after a disaster was no longer an option for many companies looking to remain competitive and respond rapidly to changing business conditions.

The evolution from the highly customized art to the science of complete data management really came to the top of the CIO's mind in the last few years. That evolution is embodied in the concept of information or data lifecycle management.

With the unfortunate events of September 11th and financial illegalities such as Enron, the need (and in many cases requirement) for more advanced management solutions surpassed what was available to the market. Many see a sophisticated information lifecycle management solution as the logical response to that need.

Information lifecycle management is the result of many different forms of data management (from basic back-up to business continuity to compliance) coming together under an umbrella solution for a business.

Today, data or information lifecycle management is at a critical point in its evolution, where complex management solutions are starting to be replicated and best practices or blueprints can be created and delivered to multiple businesses. What once was art is becoming hard science. But until this evolution is fully realized, data management persists as an art.

Data Management as Art

Today's businesses, almost regardless of the industry represented, are faced with very similar challenges. Organizations are seeing the quantity and value of their data increase while their resources to store and manage this data are decreasing or remaining flat. Skilled IT managers practice the art of data management, often using custom solutions that ultimately add to the complexity of their environments.

By replacing these artful custom techniques with a more scientific approach to data management, customers can enhance availability and security while reducing costs. As of now, there is no one technology that will solve all data management challenges and there is no one technology that can be implemented in any IT architecture. Data management continues to be primarily a highly customized and skilled art form as opposed to a science.

Looking at the evolution of data management demands, one can see that it started as a fundamental business need for data retention and more complex variations have been added, based on new or emerging requirements such as compliance and security.

However, understanding some of these primary building blocks of data management is a core necessity to understanding how to best maximize the business value gained from the management of this data and ensure that data is managed throughout its lifecycle.

The Art of Business Continuity

There is little argument that business continuity solutions are being implemented by the vast majority of companies today. This is an important shift in the way data is managed and stored as the traditional disaster recovery or backup processes begin to evolve into continuity solutions.

In its simplest form, disaster recovery solutions still create in down time and recovery time, as opposed to business continuity that focuses on maintaining operational capabilities when a disaster happens.

Traditionally, organizations approached disaster recovery in one of two unique ways:

Basic backup: Simply backing up data on additional storage resources with the hope that they will not have to ever fully lose their data and not be able to restore it. This approach still results in the business incurring downtime, potentially a loss of critical data and, in many cases, lost revenue.

Advanced Replication: This approach restores operational capabilities by restoring data and, ultimately, their applications' ability to get back up and running quickly--if they go down at all.

In the last several years, more and more organizations have been migrating to the advanced replication approach as the possibility of catastrophic loss becomes an unfortunate reality and the value of their data has reached mission-critical levels. This advanced approach, if designed and implemented well, also enables businesses to more effectively manage their backup environment, while realizing business benefits of reduced costs and increasing efficiencies in their data protection.

This type of solution will only continue to grow as technologies like SATA (Serial ATA)-based storage devices and complete virtualization of the complex physical attributes of IT infrastructures are incorporated and grow to maturity.

The Art of Compliance

Taking data management to a new level of artistic beauty is the new demand and requirement to become compliant with new and emerging industry regulations.

Many of these regulations cover four primary areas of data management that extend from data creation to disposal. They can be broken down into four primary categories:

* Creation & Use

* Retention & Management

* Control, Security & Privacy

* Reproduction & Disposition

One relevant example of the impact of these regulations is the varying requirements for businesses to retain and secure data for upwards of 20 years and, in some extreme cases, 50-100 years. For instance, some life insurance companies need to maintain data 99 years after the death of the insured. Adding to the complexity of the data management/compliance art form is the need for solutions that a business deploys to be consistent across the organization. This need will not permit pockets of non-compliance.

Take for example corporate-wide policies relating to data disposition. Such policies dictate that all areas of the corporation follow the same policy for the complete and auditable destruction of data. Any flaws or inconsistencies in the process would discredit all IT compliance efforts and potentially expose the corporation itself to significant financial and legal risk.

As a rule of thumb, the art of storage management for compliance calls for businesses to look for compliance management solutions that have heterogeneous management capabilities and enable flexibility for emerging requirements as well. The flexibility is essential, since requirements for compliance are still in development, and will infallibly face tests in a court of law.

Also, as data management is still an art form rather than hard science, any compliance vendor should also be able to offer strong professional service capabilities in order to integrate information technology with today's and tomorrow's business processes.

The Art of Consolidation

Consolidation is another component of a complete lifecycle management solution that most people consider a cost cutting measure. While this assumption is accurate as far as it goes, one of the primary reasons for consolidation is to manage the data through the four stages of compliance mentioned. Consolidation helps take some of the complexity out of compliance as it provides the IT staff with a stronger understanding of where their data is stored and how it can be best organized.

Additionally, increased redundancy, management capabilities, footprint and even security also benefit a complete consolidation initiative. These benefits, both financial and operational, reinforce the techno/business axiom that to maximize the business benefits of a consolidation solution, the solution must take the entire IT architecture into consideration.

While looking at the larger business goals and entire IT architecture, there are two primary approaches to consolidation:

Physical: Essentially, an organization decreasing the number of storage devices or management tools by consolidating them into larger or more flexible solutions.

Logical: Results from implementing a virtualization technology to help 'pool' data resources across an existing storage and IT infrastructure. The end result is to increase storage utilization and flexibility by tapping into unused storage capacity by creating a virtual storage pool of resources.

Today, the majority of consolidation efforts are focused on physical consolidation of storage resources, but that trend is shifting as technology evolves. Typically, organizations cannot afford to implement a forklift upgrade for physical consolidation, so the logical approach that leverages existing resources is growing in popularity. Technologies such as virtualization engines are entering the market and maturing, making logical consolidation a larger and larger segment of the IT strategy.

The Art of Tiered Value of Data

One of the advanced features of data management that is emerging and playing a very integral role in efficiently managing data through its full lifecycle is that of tiered data.

Data changes in value, availability and access requirements over time, and should be stored in the most appropriate way for both cost-efficiency and ready availability.

Tiered data is an approach that enables a business to prioritize its data to ensure that they are fully protecting mission-critical information but not wasting resources protecting data that is not critical to their day-to-day business.

Determining what data needs to be kept readily accessible versus what only needs to be stored for maintenance purposes is an art form more than science--as every business values (or is required by law to value) the data associated with particular applications differently. The most effective way to do this is by associating different tiers with data and applications, which may consist of:

Level 1: Information that needs instant recovery because your company will start to lose money the second it is unavailable or lost

Level 2: Data that could be recovered in a matter of hours without significant financial impact

Level 3: Data that could be recovered within a given number of days

Level 4: Information that does not require timely recovery and may not need to be readily available but must be maintained, for regulatory or other reasons.

From Art to Science

The move towards information lifecycle management (a buzzword du jour over the last year) is very much a byproduct of the realization that a complete data management solution means more than any one of the components of data management discussed earlier.

As critical management technologies continue to improve, businesses will see data management as a science with proven blueprints and methodologies as opposed to an arcane, creative art.

Today, however, businesses need to understand that lifecycle management is a holistic combination of many different data management art forms uniting as a business solution.

Chris Wood is director of technical sales and marketing at Sun Microsystems (Santa Clara, CA)

www.sun.com