IBM debuts stream computing

After grid and cloud computing, welcome to the stream.

IBM today announced the commercial launch of its System S technology for analysing massive streams of data in real time - as much as 21,600GB per hour, the equivalent of all the web pages on the internet.

System S uses a new streaming architecture and mathematical algorithms to create a forward-looking analysis of data, a process called perpetual analytics. This can narrow down what people are looking for and continuously refine the answer as additional data is available.

For example, System S can analyse hundreds or thousands of simultaneous data streams from stock prices and retail sales to weather reports and medical data.

Although you probably couldn't find enough paper

As the world becomes increasingly interconnected, the amount of data is skyrocketing. By 2010, the amount of digital information is expected to reach 988 exabytes - the equivalent of a stack of books reaching from the Sun to Pluto and back again.

Trial stream computing systems are already in place in a Swedish university to study radio-frequency 'space weather', an Irish marine institute to track underwater acoustics, an automated options trading system and a monitoring unit for premature babies.

IBM hopes to sell the technology to government and law enforcement agencies, financial institutions, retailers, transportation companies and healthcare providers. It is currently making System S trial code available at no cost to help organisations understand the software's capabilities.

IBM also announced today the opening of the IBM European Stream Computing Centre, headquartered in Dublin.