Keepit.com Online Data Backup Expert Tips: The Wisdom Of Classifying Data For Backup

By Kresten Bach Sondergaard, Product Manager, keepit.com
July 30, 2010

Keepit.com Online Data Backup Expert Tips: The Wisdom Of Classifying Data For Backup

The reality is that most of us back-up far too much data. More often than not, we are so pressed for time to evaluate the quality and type of data that is backed-up, that we take the easy way out and instruct the Data Base Administrator to back-up everything there is. These may include financial reports, personnel files, research reports and customer records. Much of this data is not correlated with the business process of the enterprise and hence there is often a poor usage of information resources in some companies. This often leads to an inability to comply with audits or legal discovery challenges for the simple reason that they are unable to find the relevant information from the mass of data that has been backed-up into the storage repositories. This also implies that there is a huge waste of storage resources and an unwarranted load on expensive, high performance systems.

Therefore, the practice of classifying data before uploading it into backup servers will help align information to business processes and create a roadmap for retaining and storing relevant information.

The underpinning logic of data classification is management of information lifecycle. This requires a solid understanding of the concept of data classification. This begins with an analysis of the factors that drive data needs. Audit and legal compliance or availability of information to end-users could be drivers.

In the first instance, the responsiveness of the storage resources may or may not be a critical factor and the data could be backed up into storage systems that are often classified as secondary.

In the second instance, the response time is critical and the data may be required to be backed up into storage resources that provide quick response times. Therefore, an understanding of the drivers for storage will help them begin the process of classification of data for back up and storage.

Classification of data for back-up is a manual process. The process can be long-drawn and very involved. While there are a number of software tools that can help you evaluate and discover information assets, there are no tools that can tell you which information is vital to your business. Therefore, this high tech process of data classification really begins very low-tech. It is a collaborative effort that can be initiated with a discussion with a piece of paper and a whiteboard.

Data most often ends up being classified by application; by groups of users; by Meta data or by type. The actual classification for data back up is entirely business usage driven.

However, the most manageable set of classification of data for backup would be one that suits the needs of the entire organization. The approach would require classification of data in smaller units and building the units up to create a structure of classification for data backup that addresses the needs of different types of users and entities within the organization.

Data classification for backup should be followed-up by determining success metrics for the classification. This may be in terms of storage resources saved and monetary value of such savings. Such metrics will provide a visible incentive for units within the enterprise to identify and backup only such information as may be considered vital for the unit.

Data classification before backup of data is extremely useful in risk management. Mission critical data is identified and can be easily protected to meet compliance audits or legal discovery tasks.

Data classification also comes with a number of other seen and unseen benefits. Deduplication of data before backup can be initiated and redundancy of data can be reduced. Less important data can be migrated from expensive primary storage to cost effective secondary storage repositories. Data encryption technologies can be applied to mission critical relevant data and indexing of data for retrieval can be performed on data that requires best storage performance.

The key to data classification before back up is communication. Business needs and risk tolerance are the drivers. The consensus among the executives will be beneficial to the organization. Everyone involved in the exercise of data classification for backup needs determination must understand the logic of the classification; the need for the classification; awareness of the environment for classification and also the need to move aged primary data off the primary storage on to the secondary storage.

Keepit.com provides automated online data backup service with unlimited space and this means there is no need for data classification. Data classification also works to the user’s advantage when it comes to initial backups and/or instant onsite restores.

About the author: Kresten Bach Sondergaard is a Product Manager at Keepit.com, a Copenhagen, Denmark based online data backup company. Keepit.com is a subsidiary of Cohaesio, a leading web-hosting firm providing services since 1996.