By Matthias Kaiserswerth
Steve Jobs famously lured John Sculley from a soda pop company to Apple in 1983 by saying, “Do you want to sell sugared water for the rest of your life? Or do you want to come with me and change the world?” In today’s business environment, the comparable challenge to a young engineer or computer scientist would be: “Do you want to create the next mobile app that makes your friends look like zombies or do you want to help transform the world of computing?”
That, in fact, is the challenge that we’re issuing today. IBM and ASTRON, the Netherlands Institute for Radio Astronomy, have assembled what some call a dream team of scientists to create a next-generation computing system capable of handling the ultimate big data challenge. Our project, called DOME, is a system for handling the deluge of data that will be created by the Square Kilometre Array, a radio telescope made up of more than half a million individual antennas that are to be scattered across southern Africa and Australia. When the SKA is completed in 2024, it is expected to process 14 exabytes of raw data per day. The data collected by the SKA in a single day would take nearly two million years to play back on an iPod.
We’re in the process of recruiting more than a half-dozen PhD.-level students to help staff the project–and we’re staging a virtual job fair to engage prospective employees. If you’re interested and qualified, visit the job fair Web site on March 26 at 5 p.m. Central European Time (Noon US Eastern Time). Only top students with huge ambitions should apply.
What’s so special about DOME and SKA? Because the telescope is to be made from so many individual antennas, the antennas are to be so widely scattered, and such a large volume of data is being gathered, a novel computing system must be developed to manage the process of gathering, storing and analyzing data from end to end. The system must have computing power that is many times that of today’s fastest computers. It must take up much less space in data centers and use much less electrical power than today’s most efficient systems. And the system will have to learn and adapt as it interacts with data.
IBM has a vision of the future of computing that we call the era of cognitive systems. We believe that this new era of computing will be as distinct from the current era, that of programmable computers, as this one was from the one that preceded it, the tabulating era. Increasingly, because of the emergence of big data, we will need computers that can learn and adapt–thinking machines rather than machines that wait passively for human commands. The SKA is not only the ultimate big data challenge; it also lights the path towards cognitive computing. Most of the fundamental scientific and engineering advances that will be required to fulfill the promise of cognitive computing will also be required to handle the huge demands posed by the SKA. Ultimately, these advances will transform business, government and our personal lives, helping us make better decisions, which, in turn, will make the world work better.
Breakthroughs targeted by the DOME team range data storage and networking to supercomputing and data analytics. We’re even creating a computer program that will help the designers choose the optimal technologies and architectures for DOME. Think of it as an intelligent design system.
There are plenty of projects for bright young computer scientists to work on these days, but it is hard to imagine other ones with the potential of rewriting the rules of computing, like DOME does. So, if you want to change the world, please look at what we’re doing.
For more information about the SKA and DOME, read this white paper.