Engineering design work has concluded for what will be the world’s fastest supercomputer, built to process the vast amount of data from the Square Kilometre Array (SKA) telescope in South Africa and Western Australia.
What’s called the SDP (science data processor) itself will be composed of two supercomputers, one in Cape Town, South Africa to process data from the SKA-mid array and one in Perth to process data from SKA-low array.
SKA is an international project consisting of thousands of antennas across the world, with central cores of operation in South Africa and in WA.
By combining signals from the large number of small antennas, SKA is in effect a single giant radio telescope capable of extremely high sensitivity and angular resolution, giving it the ability to determine where a signal originates.
SKA will feature a total collecting area of approximately a square kilometre, which will make it 50 times more sensitive than any existing radio telescopes, allowing astronomers to survey vast areas of the sky in parallel for the first time.
In Australia, the core site, comprising some 130,000 individual antennas, will be at the Murchison Radio-astronomy Observatory at Mileura Station near Boolardy in WA and operated by the CSIRO. Construction will begin next year.
Computer processing of data will be performed around the Pawsey Supercomputing Centre in Perth.
Last week, Labor promised $7.5 million additional funding for SKA should it win government.
The SKA SDP consortium, which includes the Pawsey Centre, has announced it has concluded five years of engineering design work on one of the two supercomputers to process the enormous amounts of data produced by the SKA’s telescopes.
The international consortium was led by the University of Cambridge in the UK and involved almost 40 institutions in 11 countries. SDP is the second stage of processing for SKA data and follows correlation and beamforming in the Central Signal Processor (CSP).
SKA’s SDP project manager Maurizio Miccolis said it had been a pleasure to work with such an international team of experts.
“We’ve worked with almost every SKA country to make this happen, which goes to show how hard what we’re trying to do is,’ he said.
“We estimate SDP’s total compute power to be around 250 PFlops – that’s 25 per cent faster than IBM’s Summit, the current fastest supercomputer in the world. In total, up to 600 PB of data will be distributed around the world every year from SDP – that’s enough to fill more than a million average laptops.”
The consortium role was to design computing hardware platforms, software, and algorithms needed to process data from CSP into science data products.
“SDP is where data becomes information,” said Rosie Bolton, SKA Data Centre scientist.
“This is where we start making sense of the data and produce detailed astronomical images of the sky.”
Because of the sheer quantity of data flowing into SDP – some five terabits per second or 100,000 times faster than the projected global average broadband speed in 2022 – it will need to make independent decisions in near real-time on what is noise and what’s worth keeping.
The team also designed SDP so that it can detect and remove man-made radio frequency interference (RFI) – for example from satellites and other sources – from the data.
Receive the latest developments and updates on Australia’s space industry direct to your inbox. Subscribe today to Space Connect here.