Operating Systems

A tipping point, as Malcolm Gladwell explains in his bestseller from a couple of years ago, The Tipping Point: How Little Things Can Make a Big Difference, occurs when a series of changes cause organizations to behave in fundamentally different ways. After a tipping point occurs, it’s usually quite obvious that something significant has changed. But predicting a tipping point can be difficult. Nevertheless, I’m going to give it a go.

The last tipping point in IT occurred during the late ’80s and early ’90s. During this timeframe, the industry transitioned from the age of the batch window to non-stop availability. After the advent of the Internet, it no longer was acceptable for technology folks to tell the business that its applications and systems weren’t capable of supporting round-the-clock availability.

Oh, it took some time for this tipping point to become reality. At first, businesses clamored for continual up-time and IT struggled to squeeze more availability out of the current technology. The hardware and software took some time to catch up to the need. But the industry eventually adapted to catch up to the post-tipping point requirements. We added online reorganizations, autonomic management features, and hardware with rapid failover capabilities. And now 24x7 availability is an accepted business practice and many businesses will accept nothing less.

The next tipping point—regulatory compliance—is on the horizon. Legislation is being passed at breakneck speeds. There are more than 150 regulations tracked by the IT Compliance Institute (www.itcinstitute.com/ucp/index.aspx) in their Universal Compliance project. This project is the first independent initiative to exclusively support IT compliance management.

Slowly but surely, we’ve amassed an avalanche of regulations that dictate how we must treat corporate data. As organizations react to comply with these regulations, we will see this new tipping point. I believe it will manifest itself in the form of companies beginning to treat data as a valuable corporate asset.

The change that’s imminent is that we won’t just be saying that, but actually doing it. Just about every high-level executive mouths the platitude that they already treat data as a corporate asset. But do they? Think about how we treat other important assets. Our finances are treated much more rigorously than data. If our financial statement is one penny off, we won’t stop until we track it down and get it right. Do we do the same for data quality? What about human resources? Every company has an organizational chart that maps its personnel, department, and job. Do we have a corporate data model to accomplish the same task for our data? No, we don’t treat data like we treat other assets, at least not yet.

But the laws are ahead of reality, so once again, we’re in that period where the software must catch up to the requirement. What types of software innovation will be required to enable us to actually treat data as a corporate asset?

Well, we need advanced algorithms and techniques for ensuring data quality. Today we see data profiling and cleansing tools, but these still require too much manual intervention to be successful. Advances will be required to automate data quality through pattern detection and realtime data profile management.

Additionally, robust database archiving solutions are just now entering the market. These solutions enable data to be maintained in an authentic manner and queried over long durations—decades, and even centuries, in some cases.

Furthermore, the manner by which we protect the data in our enterprise databases needs to be improved. This includes, but isn’t limited to, better and more efficient encryption and decryption techniques, label-based access security to support more granular authorization, and, perhaps most important, improved database auditing. Knowing “who did what to which piece of data when” is a prime focus of many regulations, but today’s software offerings don’t yet provide the full range of capabilities required to be in compliance.

Finally, not every change will be technology focused. Organizations will need to adopt a data governance practice to ensure data is managed (governed) appropriately for the corporation and in compliance with the pertinent regulations. Data governance is the practice of managing the availability, usability, integrity, and security of the data in use within your organization. A sound data governance program includes a governing body or council, a defined set of procedures, and a plan to execute those procedures. Without a data governance practice, regulatory compliance is impractical, if not impossible.

Some people rue the onerous burden imposed by governmental regulations on their data, but I applaud them. After all, in most cases, all these regulations are doing is forcing businesses to do the things they should have been doing anyway. Too bad it takes legislation to make that happen. Z CRAIg s. MULLINs