Shifts in data protection

There's no doubt that the entire data protection environment is evolving; it used to involve backing up data to tape, as often as once an hour, once a day, once a week or even once a month.

Redundancy of data was typically checked at the file and block level for the purpose of backup, restore and replication. Single instance storage was implemented at an object level to reduce the physical storage requirements for applications.

If a given file or block changed, the complete file or block was backed up during an incremental backup, or in the case of replication solutions, transmitted to a replicated DR device.

To many IT managers, that is still the meaning of data protection. However, directors are becoming more responsible for corporate data backup and protection, and companies are beginning to feel pressure to safeguard their transactions from outside sources such as auditors. The need to guarantee the company's ability to restore its data is assuming paramount importance.

Key technologies have emerged to help organisations rapidly store and restore data. These developments are not unique to the data centre as there is increasing focus on the large amount of data in regional offices. The following points outline key industry developments in the data protection environment to achieve three main goals.

Key technologies have emerged to help organisations rapidly store and restore data.

Logan Hill, business unit executive for security and availability at Faritec

The first goal is to achieve faster backups to tape, faster restores from tape and more efficient tape utilisation. Virtual tape library (VTL) technology virtualises data storage used typically for archival storage purposes. It comprises software that enables disk to emulate a tape drive. Instead of the image being written directly to physical magnetic tape, it is first stored on a storage device and then, after a certain period of time, moved to magnetic tape.

VTL increases performance of both backup and recovery operations. Documented performance improvements include faster backups to tape (up to five times quicker), more efficient tape utilisation - 50% increase for Windows, Unix, Linux and a documented 900% increase for zOS.

The advantages of VTL include cost savings, reliability, greater performance and expandability.

The second goal is to make network backups to disk faster and more economical. De-duplication technology builds a catalogue of the data as it is written to the VTL. Instead of applying the principle of single instance storage purely at the object level, it is now done within objects at a binary level. The products index individual bits of data in a file or block of information.

This meta data is used to rebuild the file if it needs to be recovered. The benefit of this is that the catalogue is used on subsequent backups to identify which data elements are unique. Non-unique data elements are not backed up, unique ones are committed to disk. The result is that a lot less data is backed up.

De-duplication drastically reduces the amount of storage required - various players in the market claim data compression factors between 20:1 and 60:1. This technology also reduces the amount of data being replicated between sites for disaster recovery purposes.

The third goal is to control remote data protection process in geographically distributed branch offices. The excellent business drivers behind the need to protect distributed data include regulatory compliance, supporting litigation procedures, enhancing competitive advantage and improving operational efficiency.

In remote offices where there are no trained IT personnel, and the processes for backup and recovery are mostly manual, alternatives to tape storage need to be evaluated.

Emerging technologies offer the efficient copying and transferring of remote data over the WAN, enabling the data from many offices to be aggregated into a single centralised repository. Existing problems around backup are addressed by eliminating manual and distributed processes.

There are two key technologies which have emerged in this area: distributed remote office/back office backup to disk; and wide area file system. I will explore their strengths and weaknesses in next month's Industry Insight.

For now, suffice to say the development of all these technologies has been driven by regulatory compliance requirements. Still relatively new to the market, they will all become more robust, flexible and integrated in the coming months.