Role of Interconnects in the Future of Computing

Unprecedented transistor integration capacity will exist to make computing truly ubiquitous, but the energy consumption will be a major challenge. Compute energy can be reduced by employing near threshold voltage operation, and advances in memory architecture will reduce memory energy. However, energy consumed in data movement over interconnects will become prohibitive. Severe tapering of interconnect bandwidth brings interconnect energy within limit, but potentially hinders the system performance. Hence, a holistic hardware/software codesigned system approach is required.

[1]  Andrew A. Chien,et al.  The future of microprocessors , 2011, Commun. ACM.

[2]  S. Borkar,et al.  A 320 mV 56 μW 411 GOPS/Watt Ultra-Low Voltage Motion Estimation Accelerator in 65 nm CMOS , 2009, IEEE Journal of Solid-State Circuits.

[3]  Mark Anders,et al.  Near-threshold voltage (NTV) design — Opportunities and challenges , 2012, DAC Design Automation Conference 2012.

[4]  Saurabh Dighe,et al.  A 280mV-to-1.2V wide-operating-range IA-32 processor in 32nm CMOS , 2012, 2012 IEEE International Solid-State Circuits Conference.

[5]  S. Borkar,et al.  An 80-Tile Sub-100-W TeraFLOPS Processor in 65-nm CMOS , 2008, IEEE Journal of Solid-State Circuits.

[6]  K. Bergman,et al.  Silicon Photonic Microring Links for High-Bandwidth-Density, Low-Power Chip I/O , 2013, IEEE Micro.

[7]  James E. Jaussi,et al.  A Scalable 5–15 Gbps, 14–75 mW Low-Power I/O Transceiver in 65 nm CMOS , 2008, IEEE Journal of Solid-State Circuits.