Sensors: Bringing Static Relational Databases to Life

John Cullinane, Sr.
Founder, Cullinet Software, Inc.

Database management systems have always been about integration of tools with the database such as "active" data dictionaries, report generators, application development facilities, applications and decision support systems, etc. This is because building applications is much more than just how the data is stored. Consequently, integrated database management systems have had a profound impact on the creation of online real-time computing because, for the first time, online applications could be built that were reliable, and available in a reasonable period of time. This was possible because so much of the underpinning software architecture could be purchased, thus allowing developers to concentrate on what the application was supposed to do and including performance issues, rather than re-inventing this very complex database software technology.

These database systems worked very well, but had one great flaw. They support static databases where all information is inherently "old"; and therein lies the problem. Ironically, industry gurus debate that perhaps relational database systems have plateaued in technological innovation when the real opportunity, as in the past, is in integrating with other, complementary technologies such as sensors. One reason why this is so important is that decision makers want to know what is happening on a real time basis. This requirement often causes tension between decision makers and IT management. The reason is that most decision makers will tell you that they want to know if they have a problem in time to take corrective action, not after the fact when it is too late or the data is old. IT management could never give the decision makers what they wanted. It wasn't possible. However, this is all going to change because we are entering a new era that could be called the "sensor age."

Sensors, long the purview of the engineers, will become pervasive in corporate America in the years to come. This will happen for many reasons, not the least of which is security. However, thousands of sensors will create their own set of problems, including a whole new data flow, so huge that it will overwhelm traditional database management systems with useless data. What is needed is the ability to integrate sensor applications with static relational databses in such a way as to be able to interrogate the data flow, link to the relational database, if appropriate, and send the right message to the sensors. This is truly real-time data processing and affords management the opportunity to monitor critical operations "live" and take corrective actions if necessary.

This is a major breakthrough in information technology. However, it requires special software facility to do this. One example is the LiveData RTI, which was designed to interrogate sensor data on a two-way basis while linking it to legacy relational databases. Being able to work seamlessly with existing systems is a key requirement for any organization. The integration of this technology with relational databases brings these databases to life. Actually, it can do it quite easily and, as a result, will greatly facilitate the movement of sensors to mainstream corporate, security, medical, government, and other applications – the next major innovation in computing technology.

John J. Cullinane is president of the Cullinane Group, Inc., and was founder of Cullinet Software, Inc., one of the first companies to specialize in computer software products and the premier database software company of its era. Mr. Cullinane was named three times as The Wall Street Transcript's CEO of the Year in the computer software products industry. He was founding chairman of the Massachusetts Software and Internet Council, and was inducted into INFOMART's Information Processing Hall of Fame.



MainMessage