High-powered ‘microservers’ capable of handling harsh desert conditions are planned to be developed by scientists for the Square Kilometre Array (SKA) project.

SKA is a global science and engineering project to build the world’s largest radio telescope, comprising of 3,000 dish antennas in South Africa and Australia.

Upon its completion in 2024, SKA is planned to collect big data from deep space containing information dating back to the ‘Big Bang’ more than 13 billion years ago.

The dishes of the SKA project alone could produce ten times that of current global internet traffic levels, according to Skatelescope.org.

Power needed to then process all of this data is expected to far exceed the capabilities of current technology.

Moreover, temperatures at the SKA project site in South Africa’s Karoo can soar above 40 degrees celsius in summer, adding more challenges to the big data project.

In a bid to cope with these demands, SKA South Africa, a business unit of the country’s National Research Foundation, has joined the Netherlands Institute for Radio Astronomy ASTROM and IBM in a four-year collaboration to research what the organisations says is “extremely fast, but low-power exascale computer systems to handle the big data of the project.”

Last year, ASTRON and IBM kick-started this process by also launching a public-private partnership called DOME, which is tasked with developing an IT roadmap for the SKA.

The DOME team is researching and prototyping ‘microserver architectures’ based on ‘liquid-cooled 3D stacked chips’, while its team in South Africa is to extend this research to make the microservers rugged or “desert proof” to handle the extreme environmental conditions at the Karoo SKA site.

Dr. Ton Engbersen, DOME project leader, IBM Research, told ITWeb Africa that microservers are servers the size of a circuit card at about 13 centimetres wide and five centimetres high.

These microservers then are set to comprise of processor chips with 12 cores by the end of the year and a memory capacity of 16GB.

“We basically have a server on this little card,” Engbersen told ITWeb Africa.

“And then we put multiples of those cards - roughly 128 - in something which has the depth of a 19 inch rack,” added Engbersen.

Thus far only prototypes of microservers with water cooling tech are being developed by the likes of IBM. Yet, apart from being used for the SKA project, this tech could be used by even banks and telecommunications firms to handle big data more effectively, said Engbersen.

Other technological developments could spin off from the SKA microserver project as well, say officials.

For example, IBM says it is 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: a task that is key for handling the SKA project.

The IBM technology forms part of what the firm calls ‘cognitive computing’. IBM’s ‘Watson’ supercomputer is one such example of cognitive computing as it is capable of, for instance, taking a question in natural language and searching all of the data that has been fed into its system and find the correct answer through statistical ranking.

“The DOME research has implications far beyond astronomy,” said Engbersen in a press statement.

“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 Engbersen.

Other challenges that the four organisations are planning to collaborate over the next four years include research into signal processing and advanced computing algorithms for the capture, processing and analyses of the SKA data so clear images can be produced for astronomers to study.

Developing software analytics for the project is also another task for the scientists. They plan to use 64 dishes of the MeerKAT telescope in South Africa for the testing and development of a sophisticated program to aid in the design of the entire computing system.

“The DOME collaboration brings together a dream team of scientists and engineers in an exciting partnership of public and private institutions,” said Simon Ratcliffe, technical coordinator, DOME-South Africa.

This project lays the foundation to help the scientific community solve other data challenges such as climate change, genetic information and personal medical data,” added Ratcliffe.