Square Kilometre Array (SKA) South Africa has partnered with IBM and the Netherlands Institute for Radio Astronomy (ASTRON) in the €34 million (US$44 million) DOME project, for the development of fast, low-power exascale computing systems to support the SKA telescope.
The announcement sees SKA SA join the project initiated by the two organisations last year, in a collaboration to investigate emerging technologies to support the telescope, which is to be the largest and most sensitive in the world to date.
The four-year project will involve scientists collaborating remotely and at a bespoke ASTRON and IBM Centre for Exascale Technology in the Netherlands, where they will conduct research and develop an extremely high-speed exascale computer which requires little power in order to help manage the substantial amounts of data that will be produced once the SKA is operational.
“The DOME research has implications far beyond astronomy. These scientific advances will help build the foundation for a new era of computing, providing technologies that learn and reason. Ultimately, these cognitive technologies will help to transform entire industries, including healthcare and finance,” said Dr Ton Engbersen, DOME project leader, IBM Research.
“For example, we are designing a system for storing information that learns from its interactions with the data and parcels it out in real time to the storage medium that’s most appropriate for each bit, which can also be applied to medical images.”
The SKA SA scientists will be responsible for three principal areas of research.
Firstly, they will work on signal processing and computing algorithms which will enable the capture and analysis of SKA data in the form of clear images.
Second, they will focus on “desert-proofing” the microserver equipment to withstand the conditions in the Karoo region where the SKA telescope will be located, with the team currently testing liquid cooled 3D stacked chips.
Finally, using experience gleaned from the 64 existing MeerKAT satellite dishes, SKA SA will work on developing a software infrastructure which will be able to optimally handle the 3,000 dishes of the SKA telescope.
