Modeling and analysis of the performance of exascale photonic networks

Abstract

Photonics technology has become a promising and viable alternative for both on-chip and off-chip interconnection networks of future Exascale systems. Nevertheless, this technology is not mature enough yet in this context, so research efforts focusing on photonic networks are still required to achieve realistic suitable network implementations. In this regard, system-level photonic network simulators can help guide designers to assess the multiple design choices. Most current research is done on electrical network simulators, whose components work widely different from photonics components. In this work, we summarize and compare the working behavior of both technologies which includes the use of optical routers, wavelength-division multiplexing and circuit switching among others. After implementing them into a well-known simulation framework, an extensive simulation study has been carried out using realistic photonic network configurations with synthetic and realistic traffic. Experimental results show that, compared to electrical networks, optical networks can reduce the execution time of the studied real workloads in almost one order of magnitude. Our study also reveals that the photonic configuration highly impacts on the network performance, being the bandwidth per channel and the message length the most important parameters. KEYWORDS interconnection networks, photonic technology, simulation framework 1 INTRODUCTION The most powerful supercomputers in the world 1 are ranked by their computational power in terms of floating-point operations executed per second (FLOPS). The Sunway TaihuLight, the recent supercomputer leading the list at November 2017, realizes by 93 PetaFlops (10 15 ) with 10.5 million cores. The Top500 list tracks the computational power of supercomputers since 1971, and according to the current growing compu-tational trend, it is expected that supercomputers will break the ExaFlop (10 18 ) barrier by 2020. Reaching this target, however, is challenging and requires from multiple simultaneous solutions addressing, among others, computation at chip level (nodes of the system), data movement across the system, distributed storage, energy management, etc.From the aforementioned challenges, the data movement challenge is probably the most critical to be achieved, mainly due to the increasing number of computing nodes, and therefore, the increasing communication requirements. Exascale networks will count with thousands of computing nodes, so data transmission among them becomes a major design concern, and new requirements rise not only in terms of throughput but also in energy demands. In such systems, the underlying network technology 2,3 is a critical design choice, and this is the focus of the European ExaNeSt, project which is currently being developed (see Section 2.1 for more details).In this regard, photonics interconnects, both on-chip and off-chip, have emerged as a worth alternative technology addressing the key constraints of traditional electrical networks. This technology provides much m...

Segment Switching: A New Switching Strategy for Optical HPC Networks

“…This aggregated bandwidth is distributed among several channels, and each channel is used for a different transmission. A previous study [8] found that five 320-Gbps channels provide the best performance for an aggregated link bandwidth of 1.6 Tbps. Henceforth, this will be the number of channels assumed in this work.…”

“…Note, that if pure circuit switching is used then network resources are blocked for a longer time, which causes this switching technique to present a poor utilization of the network resources mainly due to the high number of reservation conflicts. We identified this problem in a previous work [8], where we showed that photonics can provide important performance improvements when applied to conventional topologies like torus and fat tree networks. Nevertheless, we found that photonic links were underused, which indicates that there is a high potential to improve performance.…”

Foreword to the Special Issue on Processors, Interconnects, Storage, and Caches for Exascale Systems

“…Also focused on the interconnection network of an exascale system, Duro et al conduct an extensive simulation study using realistic photonic network configurations with synthetic and realistic traffic and show that, compared to electrical networks, optical networks can reduce the execution time of the studied real workloads in almost one order of magnitude. The study is performed from an architectural perspective, and the authors state that the photonic configuration highly impacts on the network performance, being the bandwidth per channel and the message length the most important parameters.…”