Square Kilometer Array Project

Lead Research Organisation: University of Cambridge
Department Name: Physics

Abstract

The Square Kilometre Array (SKA) is the next generation radio telescope. It is a global project and a truly transformational concept, providing a 50-fold improvement in sensitivity and a 100,000 increase in mapping speed over all existing instruments. It is an interferometer covering the frequency range 70-MHz to 10-GHz and consists of three collector types with common communications, control and processing technologies. Construction of the SKA will be phased: SKA Phase 1 (SKA1, ~10% of the final instrument) will be built 2016-2019. Science operations with SKA1, which itself will be the world's most powerful radio telescope, will continue during the build out to Phase 2, SKA2, with completion in ~2025.

The five key science projects (KSPs) defined by the astronomy community as driving the specifications of the SKA, together with the Exploration of the Unknown, are described in detail in "Science with the Square Kilometre Array" (eds. C. Carilli and S. Rawlings). SKA1 will enable revolutionary science, with a focus on pulsars (gravitational astronomy), and HI in the distant and the nearby Universe. SKA1 delivers breakthrough science early in the project, but truly transformational science return across all the KSPs needs the full capabilities of SKA2.

The cutting-edge technologies of the SKA have direct and relevant interest to industry especially in the areas of Information and Communications Technologies (ICT), High Performance "Exascale" Computing (HPC) requiring programming and operation techniques that support many thousands of individual processors; Digital signal processing (DSP); Electromagnetic design and modelling for high performance, low cost phased arrays, Synchronisation & Timing, and Signal & Data Transport. Recently, more than 100 UK and global high-tech commercial entities, including multi-nationals Intel, IBM, Selex Galileo, NVIDIA, CISCO, Microsoft and Nokia-Siemens, expressed interest in participating in the SKA; ~40 of the 100 companies were from the UK.

This proposal is for support of the work at Cambridge during the first two years of the preconstruction phase of the project. Cambridge will undertake two high-profile roles. Cambridge will lead the work package "Science Data Processor" which will deliver all the required elements associated with the software and computing of the SKA (excluding the real time control and monitor). The team will also make signficant contributions to the architecture, requirements definition and overall design of the SDP element. The second role is a major input to the low-frequency aperture array design Cambridge will lead the system engineering work and overall architecture definition as well as the design and delivery of proptotype low-frequency antenna elements.

Planned Impact

The cutting-edge technologies of the SKA have direct and relevant interest to industry especially in the areas of Information and Communications Technologies (ICT), High Performance "Exascale" Computing (HPC) requiring programming and operation techniques that support many thousands of individual processors; Digital signal processing (DSP); Electromagnetic design and modelling for high performance, low cost phased arrays, Synchronisation & Timing, and Signal & Data Transport. Recently, more than 100 UK and global high-tech commercial entities, including multi-nationals Intel, IBM, Selex Galileo, NVIDIA, CISCO, Microsoft and Nokia-Siemens, expressed interest in participating in the SKA; ~40 of the 100 companies were from the UK.

For the specific research proposed in this application in support of the Camrbridge SKA group the following specific interactions with industry partners are in place:

1. For the development of the low-frequency aperture array element the work is being undertaken in collaboration with Cambridge Consultants Ltd. to produce a design suitable for mass maufacture. The likely impact of this will be the ability of UK industry to bid for the manufacturing of this ke element. The close collaboration with CCL ensures the translation of EM design skills.

2. For the Science data processor string partnerships have been established with Intel, IBM, NVIDIA, Xyratex, Mellanox, Oracle, and Microsoft Research. The SKA represents one extreme example of big-data challenges in which extremely large quantities of streaming data must be processed in pseudo real-time. This is a challenge of direct and immediate importance to industry. We have structured the interaction with these leading global companies so that the SKA will act as an exemplar to drive development within these partner industries. Techniques developed through this collaborative project will be directly translatable to a wide range of other problems and through our industry partners into product. IP arrangements will be determined by agreements with each partner. A specific focus will be the development of generic architectures and implementation solutions for the processing of large data volumes with specific emphasis on optimal approaches to data routing to minimise software development costs as well as maximising throughput.

Publications


10 25 50