Square Kilometre Array Project

Posted by Eric Bastholm on December 17, 2009
Dec 172009
CSIRO's new ASKAP antennas at the Murchison Radio-astronomy Observatory (MRO) in Western Australia, 2010.  Credit: WA Department of Commerce.

CSIRO's new ASKAP antennas at the Murchison Radio-astronomy Observatory (MRO) in Western Australia, 2010. Credit: WA Department of Commerce.

I recently attended an industry brief on the Square Kilometre Array Radio Telescope (SKA) which, hopefully, will be built in Western Australia over the coming decade or so. I have attended several of these over the last few years and I always come away from them totally awe-struck. This is a big project and by big I mean really big, not just in cost, but in physical dimensions and the level of technology.

What is the SKA?

The SKA is to be the new way of doing radio astronomy; astronomy which images the sky in the radio waves part of the spectrum as opposed to the optical part or visible light part. Astronomers will be able to “look” further into space than ever before and in astronomy further in space also mean further back in time too. The distances that the light (or radio waves) have to travel are so vast that they take years to get here. In fact, they can take billions of years to get here. Some of the stars you see in the night sky may not actually be there anymore, it’s just that the light (or lack of it) telling us that fact hasn’t reached us yet. One of the reasons for building the SKA is to observe the time when the first stars were created.

A Square Kilometre?

The SKA is made up of thousands of individual dishes and other detectors spaced over an entire continent and joined together by optic fibre communications links. Together they add up to a square kilometre of receiving area and will quite easily be the biggest telescope ever. It will be 50 times more sensitive and 10,000 times faster at scanning the sky as the best telescope that we have now. The dishes are arranged in clusters which lie on the arms of five spirals emanating from a central core. If built in Australia, the spiral arms will stretch across the whole of the country and possibly into New Zealand as well. This design gives the telescope a simulated diameter, once the signals are digitally combined, equal to the distance between the two furthest detectors. Everything is wired up together and all of the data fed into a massive computing machine which, put next to your PC, would look like the human brain next to a small jellyfish.

An undertaking of this size and complexity will require much in the way of international cooperation, planning, design, testing, and exquisite coordination of all this to make it work. I am sure than many mistakes will be made along the way and the success of the project will depend very much on how nimble and flexible the various teams are in responding to the challenges and problems that arise.

Crunching the Numbers!

Astronomy is a field that is data intensive. Even modest telescopes produce tons of the stuff. Basically, billions of numbers which, when processed by clever software, will produce images of deep space in great detail. The SKA is going to take this to the extreme. So extreme, that the computing technology to pull it off hasn’t been invented yet. So it is a good thing that the project will take many years to build because it means that the computer technology has a chance to catch up. Moore’s law, which has computer technology doubling in power about every 2 years, will hopefully see the necessary data processing and storage capabilities being realised in time.

The project has a number of significant challenges to meet, particularly with respect to the data rates and processing power that is required for crunching through the numbers coming from all of the equipment. The numbers involved are so huge that I want to spend a moment trying to explain them to you, but first check out the Wikipedia page on SI prefixes which explains what the terms mean by basically showing you how many zeros you have to add to a number to get to the gigantic magnitudes that we are talking about.

The Nerdy Bit.

Storage or data rates are usually noted in terms of bytes, kilobytes, megabytes, etc. In computing these prefixes actually denote powers of 2 and not the simpler 1000s that they seem to imply, but for simplicity I will stick to the simpler interpretation. A byte is one character of information like each letter making up this sentence. A thousand of these is a kilobyte (KB). A thousand kilobytes (A million bytes) is a megabyte (MB). A thousand of these is a gigabyte (GB). A thousand gigabytes is a terabyte (TB). A thousand of these is a petabyte (PB). And finally (for now anyway) a thousand of these is an exabyte (EB). Jumping up in thousands of thousands of thousands like this quickly amounts to some very large numbers indeed.

Some Perspective.

The Apollo II spacecraft that went to the moon in 1969 had a computer with about 4 KB of memory. Around about a page or two of text.

A photo from your digital camera is between 2 to 20 MB, depending on the size of the sensor.

A music CD is about 500 MB.

A DVD is about 10 times this, around 5 GB.

You may have a hard drive in your computer of about 500 GB.

You may have a media centre serving your music and video around the home, maybe 1 TB.

That is sort of where we, in our normal lives, probably end with our familiarity of storage capacity. Even so this is pretty big, a TB will hold 2000 CDs which will take you about 12 weeks to listen to if you have enough heavy metal in the collection to keep you awake 24×7!

To the SKA though, this is nothing. Storage requirements will approach the Petabytes per day range and Exabytes data archiving. Data transmission rates of around 200 Terabytes per second will probably be required. A Petabyte of storage is about 2 million CDs – per day! Put another way if a CD were represented by the full stop at the end of this sentence, a Petabyte would be a disc 2 kilometres across.

Calculations are approximate, of course, but you get the idea – this thing is big.

Is it Worth It?

Sure. At the end of the day it is projects like this that invent the technologies that we enjoy today. The SKA will undoubtedly lead the way in huge data processing and archiving technology which we are constantly pushing the bounds of. It should result in improvements in data processing, communications, storage, archiving, retrieval, modelling and who knows what else? History is littered with examples of technology that came to be from projects that are first seemed “a waste of money”.

Links:

SKA – Square Kilometre Array

Australia and New Zealand SKA Project

SI Prefix – Wikipedia

Radio Astronomy – Wikipedia

Moore’s Law – Wikipedia

 

© 2011 phinetune Suffusion theme by Sayontan Sinha