The data processing needs of the SKA present a significant challenge and opportunity for the project.
The SKA Science Data Processor (SDP) is expected to process up to 1 terabyte per second of data and create around 200-300 petabytes of data products each year. The SDP element of the SKA focuses on the design of the computing platforms, software and algorithms needed to process data from the correlator into science data products.
The International Centre for Radio Astronomy Research (ICRAR) has been closely investigating Cloud computing as a means of addressing the SKA data storage and processing capabilities required by the international user community.
Cloud computing refers to the practice of using a network of remote servers hosted on the Internet to store, manage and process data, rather than a local server or personal computer. Due to the large amounts of data, it was acknowledged early in the design process that a single data centre with a limited power budget was insufficient to process and analyse the data needed.
Realising the huge potential offered by one of the largest public service Big Data projects in the world, Amazon Web Services (AWS) established a grant program to support international radio astronomy research. The program, Astrocompute in the Cloud, is administered by the SKA's project headquarters in the United Kingdom. Following a call for proposals in April 2015, ICRAR-led proposals were successfully awarded more than $50,000 under the program, which was around a quarter of the total funds available.
Cloud environments are changing the way software and workflows are designed. These usage patterns are closely monitored by Cloud providers who are eager to cater for Bid Data analytics. As such, the SKA, and in particular the SDP, is at the cutting edge of this rapidly increasing multi-billion dollar market.
Read about Australia’s involvement in the SKA project.