Enterprise storage: what are we talking about?
Enterprise storage - what is that, exactly? If you saw it as the title of an article in a magazine, you'd probably be justified in thinking it would be all about SATA drives, large disk arrays, virtualisation or massive tape libraries. But perhaps, to the storage administrator, IT director or compliance officer, the term `enterprise storage' should have a different interpretation - one to send a shudder down his or her spine.
This is because much corporate data sits on PCs and laptops. There is an average of 380 Megabytes of data on each device that isn't backed up regularly, according to one IDC analyst. And if that wasn't unpleasant enough, according to Gartner Group some 10 per cent of laptops are stolen yearly and 15 per cent of laptops suffer hardware failure annually. Go figure. That's a lot of unprotected data which, unless the IT department is on the ball, is likely to slip through the net at some point, with unpredictable consequences.

Arun Taneja, founder and consulting analyst at Taneja Group, puts this perspective on it: “IT departments cannot rely on users to protect data on their laptops and desktops, especially with today's strict regulations; for regulatory bodies, data is data, no matter where it resides. If you don't protect it or lose it, you're in violation.” So it gets worse: it's not just business continuity and disaster recovery that's at stake, it's compliance too. This need arises not just from US regulations but from those in force here, such as the Data Protection Act.

The upshot here is that any definition of, or discussion about, enterprise storage must include the distributed environment. Definition of and discussion of enterprise data protection must thus address the complex problems of how to back up the data held anywhere on the network. This is not, it seems, a problem with which businesses have properly got to grips. Strategic Research estimates that as many as half of remote back-ups fail. Often this is because IT specialists are not involved in what may be a less than bulletproof process. Also, the sheer volume of data growth means back-ups are not done to a schedule that can be met.

There is today a solution to the problem of remote client backup. Despite being a 2005 innovation, continuous data protection (CDP) is already well understood. It is a storage system in which all the data in an enterprise is backed up whenever any change is made. This is fairly straightforward to arrange in a data centre. In the networked environment, CDP is still the ideal solution, although the challenges are extended and fourfold:

To deliver comprehensive data protection. By definition, CDP must provide exclusive, up-to-the-moment data protection that guarantees the most current file and systems backup possible by tracking changes as they occur. Files must be protected throughout their entire life cycle. File and system restore need to be made using data as recent as the last save or any previously saved version. In a distributed environment this protection must also extend even to roaming laptops. This is achieved by caching file changes locally until a remote backup can be made automatically when a network connection is detected.
To provide transparent PC backup. As Strategic Research suggested, data protection software for remote devices must run without the need for intervention from users. It must also work independently of administrators and infrastructure. Nonetheless, users must be able to perform their own file recoveries to become productive again. The IT department might not be local.
To aid fast remote system restore. File and system recovery must be fast, easy and - unlike solutions responsible for the current statistics - reliable. It must be easy for administrators to control roll back a system to a good point. This limits the need for extensive troubleshooting and provides the ideal safety net for common difficulties, such as virus attacks.
To ease data management. Today's CDP-based solutions must also still meet the demands of minimizing downtime, of reducing PC total cost of ownership, of increasing IT and end-user efficiency, of simplifying data management - and of saving money.
Discussion between any two parties about enterprise storage needs to begin by making clear what you both understand by the term. This means researching your own environment and arming yourself with knowledge of where the data in your organisation resides. What is your risk and investment analysis of that data? Can you afford to lose it - fiscally, operationally or from a regulatory standpoint?
Next time someone starts to talk about their enterprise-class data protection solution, make sure you're both talking the same language. They could easily talking simply about replication from a primary data centre to a secondary. Make sure that their enterprise solution doesn't just look at the big things, but focuses on the little things, too. Like PCs and laptops.
***
François Gauthier, Chief Technology Office for Atempo, contributed this article. Our thanks go to him.
Atempo is exhibiting at Storage Expo, the UK's largest event dedicated to data storage. Now in its fifth year, the show features a comprehensive free education programme, and over 90 exhibitors. It takes place at the National Hall, Olympia, London from 12 - 13 October 2005.
|