The Setonix supercomputer celebrated its entry into service by unveiling a superb image of a supernova remnant.
Australian astronomers have plenty to be happy about right now; the country of kangaroos has just secured the services of Setonix, a formidable brand new supercomputer that is just waiting to dissect data in industrial quantities. Less than 24 hours after its commissioning, astronomers have already used it to produce a breathtaking image of a supernova remnant.
The data in question comes from theASKAP. It is a somewhat special observatory since, like the Event Horizon Telescope, it is not an isolated device; instead, the signals are collected using a set of 36 radio antennas spread over an area of approximately one square kilometer.
This architecture makes it possible to identify particularly weak signals. Individually, they therefore do not weigh very heavily in terms of scientific research. But they all work at a slightly different wavelength; by processing all these elements in parallel, it is therefore possible to combine bits of information contained in each statement to obtain composite images. Specialists can then use them in their work.
A supercomputer entirely devoted to astronomy
The problem is that ASKAP and its equivalents also produce a phenomenal amount of data. It is far too large to be processed by a standard classical computer in a reasonable time; to produce the composite images that interest researchers, it is therefore necessary to equip themselves with veritable computer monsters. And that’s where Setonix comes in.
Since its construction in 2009, ASKAP has been connected to a neighboring research center (the Pawsey Supercomputing Research Center, or PSRC) by a set of specialized optical fibres. The resident researchers can thus collect and then use the data from the telescope as quickly as possible. And the arrival of Setonix in this institution is already changing the situation at this level.
On paper, this is a relatively modest supercomputer; it displays “only” 50 petaflops against the 1102 of Frontier, the current defending champion (see our article). But the fact that it is directly linked to ASKAP and specially designed to exploit its data still makes it a sensational tool for astronomers.
The craft is just emerging from the first phase of its deployment. Its operators therefore wanted to test the capabilities of their new toy. The researchers behind this work fed it with data from a spectacular supernova remnant called G261.9+5.5. The goal: to produce a usable image of the phenomenon from an incredibly dense dataset.
A magnificent supernova remnant
Supernovas are huge explosions that occur at the end of a star’s life cycle, when the latter reaches the end of the reserves that fuel its thermonuclear reactions. These apocalyptic events usually leave behind a large cloud of superheated, extremely bright and therefore easily spotted matter; we then speak of remanence. These remnants are very interesting for astronomers. They are usually packed with crucial information about the life cycle of stars — and by extension the overall dynamics of the universe.
If so, this image of G261.9+5.5 will probably only bring no revolutionary discovery; hasess all, this object has already been regularly scrutinized by specialists since its discovery in 1967. On the other hand, this test has made it possible to prove the ability of Setonix to ingest and digest a phenomenal amount of data. And incidentally, the swirls of superheated gas generated by this cosmic cataclysm are also a spectacular spectacle for space lovers.
The other good news is that this sensational image of G261.9+5.5 (see top of article) is probably just the first of a long series. The capabilities of the machine now validated, the PSRC teams will soon be able to complete the second phase of its deployment; a deadline that will further increase its processing capacity. We can therefore expect Australian astronomers to regularly provide us with breathtaking images, as the James Webb Space Telescope has been doing regularly since July 12 (see our article).