A class of highly scalable optical crossbar-connected interconnection networks (SOCNs) for parallel computing systems

Abstract

AbstractÐA class of highly scalable interconnect topologies called the Scalable Optical Crossbar-Connected Interconnection Networks (SOCNs) is proposed. This proposed class of networks combines the use of tunable Vertical Cavity Surface Emitting Lasers (VCSEL's), Wavelength Division Multiplexing (WDM) and a scalable, hierarchical network architecture to implement large-scale optical crossbar based networks. A free-space and optical waveguide-based crossbar interconnect utilizing tunable VCSEL arrays is proposed for interconnecting processor elements within a local cluster. A similar WDM optical crossbar using optical fibers is proposed for implementing intercluster crossbar links. The combination of the two technologies produces large-scale optical fan-out switches that could be used to implement relatively low cost, large scale, high bandwidth, low latency, fully connected crossbar clusters supporting up to hundreds of processors. An extension of the crossbar network architecture is also proposed that implements a hybrid network architecture that is much more scalable. This could be used to connect thousands of processors in a multiprocessor configuration while maintaining a low latency and high bandwidth. Such an architecture could be very suitable for constructing relatively inexpensive, highly scalable, high bandwidth, and fault-tolerant interconnects for large-scale, massively parallel computer systems. This paper presents a thorough analysis of two example topologies, including a comparison of the two topologies to other popular networks. In addition, an overview of a proposed optical implementation and power budget is presented, along with analysis of proposed media access control protocols and corresponding optical implementation.

An optical interconnection network and a modified snooping protocol for the design of large-scale symmetric multiprocessors (SMPs)

“…The snoop response also takes similar number of cycles after the address is snooped. The data network considered for SYMNET is the SOCN [17] network. SOCN is an optical crossbar constructed using VCSEL/PD arrays and diffraction grating.…”

“…The address and data subnetworks are separated, reducing the design complexity and enabling the design of large scalable SMPs. Scalable data subnetworks have been studied elsewhere [17], therefore this paper focuses only on the address subnetwork. The address subnetwork consists of two components, a transport part capable of transmitting multiple address requests and a control part which ensures collisionless transmission of these address requests.…”

RAPID: Reconfigurable and Scalable All-Photonic Interconnect for Distributed Shared Memory Multiprocessors

“…RAPID R( ) is compared with several well-known network topologies such as a traditional crossbar network (CB), the Binary Hypercube, the Ring network, the Torus, two-dimensional (2-D) Mesh, and Scalable Optical Crossbar Network (SOCN) [24]. Each of these networks will be compared with respect to degree, diameter, number of links and bisection width.…”

“…The multiplexer in our proposed architecture is designed as a tree of 1 2 couplers constructed using optical polymers to comine the various signals from different modes [24].…”

A scalable architecture for distributed shared memory multiprocessors using optical interconnects

“…RAPID R(d,g) is compared with several well-known network topologies such as a traditional crossbar network (CB), the Binary Hypercube, the Ring network, the Torus, 2-D Mesh and Scalable Optical Crossbar Network (SOCN) [23]. Each of these networks will be compared with respect to degree, diameter, number of links and bisection width.…”