Breaking the Copper Bottleneck: Computer Architecture and Power Implications of Photonic Interconnect

The provision of future services in the digital economy is reliant on achieving more power efficient computers. Recent bandwidth improvements in electronic interconnects have only been achieved at the expense of dramatic increases in latency and power consumption [19]. Photonic technologies appear essential to make chip-to-chip communication sustainable for ever-higher data rates due to inherently lower power operation. Recent advances in silicon photonics, photonic printed circuit boards (PCB) and 3D integration technologies indicate great promise for short distance photonics. However, given the radical changes in computer design brought about by chip multiprocessors (CMP) and the fundamental differences between electronic and photonic communications, the design implications for complete computer systems are not clear. The proposed research will study the implications of upcoming photonic technologies on the power consumption and architecture of large computer systems such as high performance computers and data centres. The uniqueness of the proposal is its method and the holistic results that it should produce. I will start with a firm scientific foundation based on characterisation of emerging photonic devices leading to models of existing and predicted future components. FPGA-based emulation will enable investigation of complete multi-chip, multi-core computer systems and interconnect running at around 1/100th of real time. The outcome will be the knowledge to build large computer systems optimised for minimum power consumption. This multidisciplinary research therefore underpins several EPSRC themes: digital economy, next-generation healthcare and energy efficiency as well as responding to the EPSRC signposted Moore-for-Less microelectronic grand challenge.

The provision of future services in the digital economy is reliant on achieving more power efficient computers. The project will determine the performance and power consumption benefits that can be obtained in future multicore, multichip, high performance computers by the use of photonic chip-to-chip networks. The industrial beneficiaries of the research in the medium term will be manufacturers of computer equipment and designers/installers/users of high-end computer systems. The proposal addresses two broad areas of computer design: high performance computing (HPC) and data centre systems. In both areas there are applications which require higher performance computers with reduced or stabilised power consumption. For example: - Supercomputers used in a variety of UK sectors: meteorology, defence, academic scientific computing, financial services, food, and semiconductors; - Business intelligence and analysis - On-line transaction processing - Web search systems - Cloud computing (virtulisation) - Video distribution In the longer term, the results of this work could lead to new mass market products (i.e. photonic PCI) which will lead to new opportunites for computer equipment and photonic system manufacturers. We intend to leverage the commercial possibilities of the work through appointing advisors to the project from interested companies and consideration of consultancy and spin off company opportunities as described in the impact statement. In terms of the impact on wider society, the research will promote sustainable data centres which will allow continued growth of web-based products and services to be developed by government and industry. Ultimately all computer users will benefit from new mass-market products such as photonic PCI. The research will also enable reductions in carbon dioxide emmisions though more sustainable HPC and data centre systems. Improved sustainability in the provision of data centres enables the increased use of web-based services which should lead to increased efficiency and sustainability in other sectors (i.e. reduction in demand for transport). These impacts will be obtained as a by-product of the commercial activity described above with possible government intervention. Sustainability and power consumption reduction are major research and policy themes. The research will define the potential economic and environmental impact of future large computer systems and enable policy makers, research councils and other funding bodies make better informed funding decisions and research plans.