Abstract:The expected exascale computer will comprise tens of thousands of computing nodes and nearly 5000 interconnected nodes in years to come. Such a large-scale system will represent a milestone in the progress of High-Performance Computing (HPC). The more efficient network hardware, like optoelectronic interconnection and configurable switches, is reforming the traditional architecture of supercomputers. However, the present architecture containing new hardware is not easy to adapt to the dynamically running condition, because the newly-developed hardware is normally unable to effectively improve overall performance. Here, we propose a new accelerated system called Software Defined Network Accelerator (sDNA) for the exascale computer. Inspired by edge forwarding index (EFI), the main contribution of our work is that it presents an extended EFI-based optical interconnection method with slow switching optical device. The optical link is connected by the evaluation of each optical link candidate's traffic offloading revenue. As the supporting method for optical interconnection, sDNA selects the most suitable routing configuration according to the job-schedule information and the prior-knowledge of HPC applications. We tested sDNA in a network simulator and a prototype system for the exascale computer, using both DOE application benchmarks and a real-world communication benchmark. From the result of verification of traffic offloading, we found that our optical interconnection method based on our extended EFI evaluation is not only able to offload the traffic from an electrical link to an optical link but is also able to avoid congestion inherent to electrical link. Furthermore, our experimental results show that sDNA maintains the throughput of more than 80% bandwidth and reduced the communication delay by 10% in our real prototype system and simulator. Together, our sDNA is an ideal candidate for accelerating communication performance of the exascale computer.

OeIM - An Optoelectronic Interconnection Middleware for the Exascale Computer.

Extending On-Chip Interconnects for Rack-Level Remote Resource Access

A New Traffic Offloading Method with Slow Switching Optical Device in Exascale Computer.

Hierarchical and Reconfigurable Optical/electrical Interconnection Network for High-Performance Computing

A New Optoelectronic Hybrid Network Based on Scheduling Optimization of Optical Links

Wormhole Optical Network: a New Architecture to Solve Long Diameter Problem in Exascale Computer

Simulation approach of hybrid optical-electrical on-chip interconnects for multicore systems

Demonstration of Optically Connected Disaggregated Memory with Hitless Wavelength-Selective Switch

Experimental demonstration of the optical multi-mesh hypercube: scaleable interconnection network for multiprocessors and multicomputers

High Performance Multiprocessor Computing System Using Optical Centralized Shared Bus

Design and Implementation of Reconfigurable Array Adaptive Optoelectronic Hybrid Interconnect Shunting Network

Energy-Efficient Hybrid Optical-Electronic Network-On-Chip for Future Many-Core Processors

OEHadoop: Accelerate Hadoop Applications by Co-Designing Hadoop with Data Center Network

A scalable silicon photonic chip-scale optical switch for high performance computing systems

Alleviating Thermal Constraints While Maintaining Performance Via Silicon- Based On-chip Optical Interconnects

Edge-guided inverse design of digital metamaterials for ultra-high-capacity on-chip multi-dimensional interconnect

Wavelength division multiplexers/demultiplexers for optical interconnects in massively parallel processing

All-optical scalable spatial coherent Ising machine

Traffic Pattern Adaptive Hybrid Electrical and Optical Switching Network for HPC System

On-Chip Optical Interconnect Roadmap: Challenges and Critical Directions

Electrical and optical on-chip interconnects in scaled microprocessors