How do you know when it's time to build a formal data archive? According to recent data, most enterprises are already long past due.

An interesting series of blogs from archive vendor Permabit shows that not only do most primary and near-line file servers hold vast amounts of unused data, but that most standard backup solutions, while suitable for disaster recovery and similar functions, are not the most effective means of storing aging or little-used data.

First, the company presents Jerome M. Wendt, president and lead analyst at DCIG, who sites new research from NetApp and the University of California that indicates the amount of untouched data in today's production storage systems could be approaching 90 percent. That means the high-performance disk drives for which you shelled out big money could be holding multiple terabytes of data that no one will ever touch again.

But once you've established that you need full-blown archiving rather than simple backup, where do you go from there? Permabit's Jered Floyd thinks it all depends on what kind of enterprise you are. On the one hand, there are those with largely disparate, independent applications that need to sort data using standard data-classification and life-cycle management. On the other hand are organizations consolidating applications onto archival tiers that sit right alongside primary storage to eliminate the mass data migration of old.

Coincidently, Permabit just joined up with a company called Atempo, which specializes in cross-platform archival systems, to develop a new generation of massively scalable archive systems designed to work across multiple NAS environments and operating systems. The system brings the policy-based management strengths of the Atempo Digital Archive (ADA) software to the Permabit Enterprise Archive platform that provides a drag-and-drop interface for files and file shares under both the NFS and CIFS standards.

Another major partnership in the archiving space is between COPAN Systems and Quantum Corp. The two have united the Quantum StorNext data-management software stack with the COPAN Enterprise MAID system to provide a file-based management approach for up to 896 TB of unstructured data. The system is targeted at pooling high-speed, high-capacity data, such as digital media and captured data.

Data archiving is not an endeavor to be undertaken lightly. The technologies involved are not plug-and-play, but involve a substantial amount of policy development and intricate system coordination. You might even say that archiving forces managers to ask themselves the most fundamental questions: Who are we? What do we value? Where are we going?

But if you can settle these issues to any meaningful degree of satisfaction, you'll find that an effective archiving system does wonders for your productivity and peace of mind.