Researchers will be developing cutting-edge computer systems to process the 14 exabytes of data the SKA telescope will gather every day

Scientists sought to apply Big Data to Big Bang

Scientists are being sought to join a project to develop “big data” computer systems for the Square Kilometre Array (SKA).

A virtual recruiting event will be held tomorrow to fill 10 positions on the DOME project, a five-year collaboration between IBM and ASTRON, The Netherlands Institute for Radio Astronomy, to develop low-power exascale computer systems to process data from SKA.

The €1.5bn SKA will be the world’s largest and most sensitive radio telescope designed to address some of the fundamental unanswered questions about the Universe including carrying out extreme tests of general relativity and searching for evidence of dark matter and dark energy.

SKA will begin construction in 2016 and will make initial observations by 2019 but, with the telescope expected to gather 14 exabytes (14 billion gigabytes) of data a day, current computer systems will not have the power necessary to process the data.

To address these problems scientists from the two organisations will collaborate at the newly established ASTRON & IBM Center for Exascale Technology in Dwingeloo, the Netherlands, to investigate energy efficient exascale computing, data transport and storage processes and streaming analytics.

“The reason I am interested in doing this is that today we don’t know how to build this. That is what I find really motivating,” says IBM scientist and DOME project lead Ronald Luijten. “We will need breakthroughs on several fronts in order to get the system built in an affordable way and we need the best people on the planet to work on this.

“We need people who can work outside the box and have the creativity to generate their own vision of what the problems are and how to solve them in a different and new way.”

According to Luijten the major breakthroughs the team will have to make in order for the project to be energy efficient and affordable will be reducing data movement, which currently accounts for 98 per cent of energy usage, vastly reducing the cost of photonics and removing the need for rotating discs for memory storage.

With the processing power required to operate the telescope equal to several millions of today's fastest computers the challenge is enormous, but the team are confident of the project's success.

“I have been in research for 28 years and, as we say, need is the mother of invention,” says Luijten.

“We need people who have got an understanding of the whole system. They must be able to dig deep into particular problems trying to solve a particular issue, but they need a broad understanding of what the system is and what the requirements are.”

To attract the top applicants the project will hold a virtual recruitment session using IBM’s Smart Cloud technology at 4pm tomorrow to give a series of presentations on the project and roles available as well hosting a Q&A session for those interested.

The team is seeking both pre-doc and post-doc applicants who will be offered a salaried contract to work at the project’s base in the Netherlands for the remaining three-and-a-half years of the project.

Visit the IBM website for more information.

Recent articles

Info Message

Our sites use cookies to support some functionality, and to collect anonymous user data.

Learn more about IET cookies and how to control them