Non-blocking Switch Research Articles

Spatial and mode selective switch for orbital angular momentum mode division multiplexing.

In analogy to a wavelength selective switch in wavelength-division multiplexing (WDM) optical fiber communication systems, a spatial and optical mode selective switch (SMSS) would be an important component in future ultrahigh capacity optical fiber communication systems based on space and mode division multiplexing. In this work, a free-space SMSS for orbital angular momentum (OAM) mode-division multiplexing (MDM) is proposed and experimentally demonstrated. The SMSS consists of a separating part for transforming OAM modes to spatial modes and a recombining part for selecting and recombining the modes to any spatial channel. The SMSS is able to implement strictly non-blocking switching between a total of 36 SDM/MDM channels configured as four spatial channels each supporting nine OAM mode channels.

Multicast Space-Conversion-Space Strict-Sense Nonblocking Switching Fabrics with Multicast Opportunity in the Last Stage for Continuous Multislot Connections

This paper introduces two nonblocking switching fabric architectures designed for multicast connections. These connections are defined using the multislot connection approach, mainly applied in elastic optical networks. In contrast to earlier solutions, this approach assumes that multislot connections are consistently established in adjacent continuous slots. This implies that previously established solutions could not be applied. Our study presents a comprehensive theoretical framework applicable to the general case of three-stage switching fabrics. These fabrics feature external stages equipped with space switches, while the middle stage incorporates conversion switches that operate in the wavelength, time, or frequency domain. In addition, multicast capabilities are deliberately confined to the output-stage switch or switches. A fundamental contribution of this work lies in the formulation of the worst-case scenario, which serves as the foundational basis for deriving strict-sense nonblocking conditions governing such multicast switching fabrics. Our analysis formally demonstrates that the fundamental structure of the multicast nonblocking switching fabric aligns closely with that of the previously examined point-to-point fabric. The only difference is related to the ability to multicast within the output stage of the switching fabric.

A Universal Method for Constructing N-Port Reconfigurable Non-Blocking Optical Switches on a Silicon Chip

A Universal Method for Constructing <i>N</i>-Port Reconfigurable Non-Blocking Optical Switches on a Silicon Chip

O-Star: An Optical Switching Architecture Featuring Mode and Wavelength-Division Multiplexing for On-Chip Many-Core Systems

In this paper, we propose O-Star, a scalable optical switching architecture for on-chip many-core systems, employing hybrid mode and wavelength division multiplexing technology. The O-Star uses the Benes topology as the core switching module to realize non-blocking switching. Besides, we design a wavelength and mode allocation module to enable each processor to transmit data in parallel. To quantitatively analyze the minimum hardware cost required, we establish a mathematical model for the different number of processors. As a proof of concept, a 64-core optical switching architecture featuring 8-port Benes topology, 2-mode, and 4-wavelength channels is simulated with single channel 25 Gbps data rate. The O-Star is flexible to scale, both for the number of supported processors and the switching capacity. O-Star holds promise for realizing large-scale optical switching networks to address the incoming challenges of high-performance computing (HPC) systems and data center networks (DCNs).

Optimizing meta-slots for nonblocking elastic optical switching networks

Elastic optical network is a promising technology for building flexible and wideband communication systems. This technology features the frequency slot unit (FSU) concept, which defines the bandwidth unit in the frequency domain. The utilization of a specified number of consecutive FSUs allows the bandwidth to be flexibly assigned to a data stream. For a successful elastic optical network operation, the use of nonblocking optical switching networks is indispensable. This paper focuses on two previously proposed W-S-W switching network architectures, namely, WSW1 and WSW2. With regard to WSW1, a previous study elucidated that the hardware complexity, which is evaluated by the number of FSUs, decreases by employing the meta-slot scheme. A meta-slot is a frequency range containing one or more FSUs. Although its effectiveness depends on the sizes of meta-slot classes, the previous study did not present how to optimize the meta-slot class sizes. Moreover, the employment of meta-slots was not considered for WSW2 in the previous study. This paper investigates the optimization of meta-slot class sizes and demonstrates that such optimization is modeled as the shortest path problem. For WSW1, the meta-slot scheme optimized by the shortest path model is compared with the previously reported nonblocking conditions. The result confirms the superiority of the optimized meta-slot scheme. For WSW2, the assignment of meta-slots among S-switches is also essential. The paper models the assignment of meta-slots as a bin-packing problem. Thus, the near-optimal assignment can be obtained by a known bin-packing heuristic. For WSW2, the number of S-switches is evaluated for the optimized meta-slot scheme and the previously known nonblocking conditions. The result confirms that the meta-slot scheme is advantageous for WSW2 as well as for WSW1.

Scheduling coflows of multi-stage jobs under network resource constraints

As an emerging network abstraction, coflow greatly improves the communication performance of data-parallel computing jobs. Many existing studies have focused on the design of coflow scheduling to minimize the completion time of jobs. However, they treat the underlying network as a large non-blocking switch without considering the constraints of network resources, which may increase network bottlenecks, reduce link capacity utilization, and extend job completion time. In this paper, we take network resource constraints into account and study how to schedule coflows in multi-stage jobs with the objective of minimizing the total weighted job completion time. We first formalize this multi-stage job scheduling problem as nonlinear programming and prove its NP-hardness. By introducing a priority order of jobs using a linear programming relaxation-based approach, we propose a polynomial-time algorithm with a performance guarantee, which can achieve a constant approximation ratio in many typical scenarios. Simulation results based on a Facebook trace show that, compared with a state-of-the-art approach, our algorithm can shorten the average weighted job completion time by up to 48.46% and run faster by up to 19.12×.

Advanced Multi-Functional Integrated Photonic Filters Based on Coupled Sagnac Loop Reflectors

We theoretically investigate integrated photonic resonators formed by two mutually coupled Sagnac loop reflectors (MC-SLRs). Mode interference in the MC-SLR resonators is tailored to achieve versatile filter shapes with high performance, which enable flexible spectral engineering for diverse applications. By adjusting the reflectivity of the Sagnac loop reflectors (SLRs) as well as the coupling strength between different SLRs, we achieve optical analogues of Fano resonance with ultrahigh spectral slope rates, wavelength interleaving / non-blocking switching functions with significantly enhanced filtering flatness, and compact bandpass filters with improved roll-off. In our designs the requirements for practical applications are considered, together with detailed analyses of the impact of structural parameters and fabrication tolerances. These results highlight the strong potential of MC-SLR resonators as advanced multi-functional integrated photonic filters for flexible spectral engineering in optical communications systems.

On Scheduling Coflows

Applications designed for data-parallel computation frameworks such as MapReduce usually alternate between computation and communication stages. Coflow scheduling is a recent popular networking abstraction introduced to capture such application-level communication patterns in datacenters. In this framework, a datacenter is modeled as a single non-blocking switch with m input ports and m output ports. A coflow j is a collection of flow demands \(\{d^j_{io}\}_{i \in m, o \in m}\) that is said to be complete once all of its requisite flows have been scheduled.

The Construction of High-Performance Optoelectronic Interconnection Network and the Realization of Cloud Computing for Traffic Data Recognition

Due to the problems of congestion, insufficient scalability, and high cost in the network core layer link, the network interconnection architecture needs to be optimized. The photoelectric interconnection technology has attracted the attention of many researchers with its high bandwidth, low latency, and low power consumption. In this study, after fully considering the requirements of optoelectronic switching on network node degree and processing delay and corresponding to different transmission characteristics of circuit switching and optical circuit switching, an optoelectronic interconnection architecture supporting heterogeneous network topology, namely THtop, is proposed. According to the optical circuit switching/circuit switching characteristics of the Internet, the corresponding topology structure is designed respectively in this architecture to enhance the system integration and reduce the cost. At the same time, in order to meet the requirements of wavelength and number of hops of optical circuit switching, distributed control is designed in THtop, and the design method of heterogeneous optical switching unit is introduced in the switching equipment, so that the optical switching node can further reduce the cost while playing its role. The use of high-speed optoelectronic networks will cause a large amount of traffic on the channel. In this study, the traffic identification is analyzed on the proposed THtop architecture. The Map Reduce framework in cloud computing is used as the identification model, and the K-means algorithm is taken as the identification algorithm. According to its characteristics, the W_K-means algorithm is proposed for the type identification of traffic. The results show that under the OpenNet Internet simulation environment, when the traffic is set to a certain level, more than 90% of the non-blocking network switching can be realized under the THtop architecture, and the completion time of the mouse flow and elephant flow can be significantly improved. Therefore, under the Map Reduce model, the proposed W_K -means algorithm can quickly identify different traffic types.

Wide-sense nonblocking and blocking converting-space-converting switching node architecture under XsVarSLOT algorithm

In this study, we evaluate the operation of the Converting-Space-Converting switching fabric architecture. The investigated switching fabric is suitable for all kinds of slot-operating switches, no matters the slots are wavelengths, time, or others. The switching fabric is evaluated in terms of the number of slots in the interstage links and the blocking probability. In the evaluation process, we used the proposed class of routing algorithms based on the functional decomposition of slots into subsets. The routing algorithm decomposes the slots in the interstage links into X subsets. The number of slots in each subset is evaluated by simulation, in order to obtain a blocking probability at the value of about 10−8, and analytically, in case the blocking probability is reduced to 0. The proposed algorithm can reduce the required number of slots in the interstage links by almost 84% in the nonblocking switching fabrics. However; when the blocking is acceptable, the proposed routing algorithm can also reduce this number of slots, but only when the values of blocking probability are very low.

NDP

The key goals of datacenter networks are to simultaneously provide wire-level latency for RPC-style applications and high-throughput for network-bound applications such as storage. Folded Clos networks [1, 11] are used in datacenters worldwide; such networks use many cheap commodity switches to provide the illusion of a big non-blocking switch to all hosts in a datacenter, offering many paths between any two pairs of hosts.

Multi-stage Interconnection networks CLOS/BENES Parallel Routing Algorithms for circuit switching system

This article approaches the design of parallel routing Clos and Benes switching networks in Communication Technology. In communication, the transmission of data with less traffic and low latency are the biggest challenges. The conventional packet switching circuits takes the more power and high area to overcome this problem parallel routing algorithms are proposed. Clos and Benes networks are designed for the circuit switching systems where the switching configuration will be rearranged and it’s relatively low speed. Most of the existing parallel routing algorithms are not practical those are fail to interconnects the inputs with the matched outputs with less traffic. In this article, we designed Clos and Benes network. Clos and Benes networks are the Non-blocking switching Networks. Clos Switching network provides the better results like low area and less delay when compare with the Benes Switching Network. Clos and Benes non-blocking switching circuits are designed by Verilog HDL, Synthesized and simulated by XILINX 12.1 tool

Design and Implementation of Rearrangable Non-Blocking Switching Network in VLSI

The main goal of this article is to implement an effective Non-Blocking Benes switching Network. Benes Switching Network is designed with the uncomplicated switch modules &amp; it’s have so many advantages, small latency, less traffic and it’s required number of switch modules. Clos and Benes networks are play a key role in the class of multistage interconnection network because of their extensibility and mortality. Benes network provides a low latency when compare with the other networks. 8x8 Benes non blocking switching network is designed and synthesized with the using of Xilinx tool 12.1.

The control system elements of the new generation optical switching cell

In this paper calculation of new optical switch parameters that allows us to create next generation all-optical non-blocking switching system without external control devices is carried out for the first time. In particular, switching cell control device is studied in detail. The presented one includes Bragg filter, frequency detector, optical isolator, and former of a control signal. Here we also present the detail description and the numerical calculations of these devices for third transparent window (1550nm). The reflection and transmission coefficients are obtained, the passbands of the Bragg filter are presented, and the amplitude characteristic of the frequency detector is calculated.

Rearrangeable 2 times 2 elastic optical switch with two connection rates and spectrum conversion capability

In this study, we consider simultaneous connection routing in the three-stage switching fabric of the wavelength-space-wavelength architecture for elastic optical switches, which serve connections that can occupy different spectrum widths. Recently, the upper bound of the rearrangeability conditions was derived and proved for a switching fabric serving a limited number of connection rates. A control algorithm based on matrix decomposition was also proposed. In addition, the necessary and sufficient conditions were derived and proved for a switching fabric with a size of 2 times 2 serving only two connection rates, but only when the ratios between the connection rates and link capacity are integers. In this study, we extend these results to an arbitrary ratio between the connection rates and link capacity. The number of frequency slot units required in the interstage links is much lower than that in the strict-sense nonblocking switching fabrics. We also propose modifications to the control algorithm, which further reduces the number of frequency slot units required in many cases. Finally, we extend the results for 2 times 2 switching fabrics to those with a size of r times r.

Layered OXC With Intermode Switching Bridge for Optical SDM-WDM Networks

Recently, spatial-division-multiplexing based on weakly coupled multicore or few-mode fibers has been proposed for short-reach switching scenarios, in which spatial mode can be used as an independent switching dimension. Fully connected cross-connections (OXC) converting all spatial modes and wavelengths into one-dimensional switching ports and followed by huge switching array can provide nonblocking switching, but the hardware implementation is very complex and expensive. In this paper, we propose a layered OXC structure consisting of intramode sub-OXCs and intermode switching bridges. Then, a scattered-spectrum-scan intramode-first routing and spectrum allocation (SSS-IMF-RSA) scheme is designed, in which only a few discrete spectrum segments instead of the whole spectrum should be scanned for spectrum allocation. The switching will be performed preferentially for the wavelength dimension because of mature switching capability and less searching time. We show by simulation that the proposed layered OXC with the SSS-IMF-RSA scheme can achieve similar blocking performance to fully connected OXCs with up to 76.1% hardware scale reduction and 93% searching time reduction for network accommodation. The proposed OXC and RSA scheme greatly decrease the requirements for the port number of WSS and enable components reuse to achieve smooth evolution.

Study of Parallel Algorithms for IP Switches

Abstract Owing to the increase in the internet traffic, any calculation that requires more than linear time would be considered unreasonably moderate for constant applications. One cure is to utilize numerous processors to build up associations in parallel and the other is to construct low cost, high speed, large capacity non-blocking switching architecture. In this paper, our focus is on developing parallel algorithms for routing which will ensure high-speed internet connectivity and at the same time making the system to be cost effective.

Asymmetrical Space-Conversion-Space SCS1 Strict-Sense and Wide-Sense Nonblocking Switching Fabrics for Continuous Multislot Connections

Recently, Danilewicz et al. have proposed two strict-sense nonblocking structures of switching fabrics, called Space-Wavelength-Space 1 and Space-Wavelength-Space 2, for elastic optical network nodes. This paper considers one of the two previously presented structures. The node uses a three-stage switching fabric that applies space switching in the first and third stages and wavelength switching in the second stage (an S-W-S switching fabric). In elastic optical networks, an optical path can use a frequency slot spread over m adjacent frequency slot units. Such a connection is called an m-slot connection. In this paper, a more general case is presented in which conversion in the middle-stage switches can be made for one of the different domains (frequency, time, etc.). However, the m-slot connections are a basis for the presented theory. This theory concerns space-conversion-space (S-C-S) switching fabrics in which conversion can be performed for different domains. Danilewicz et al. considered S-W-S switching fabrics in which 1 ≤ m ≤ m max . For the S-C-S switching fabrics discussed in this paper, we derive and prove strict-sense nonblocking conditions when m-slot connections are set up, in which m min ≤ m ≤ m max . Strict-sense nonblocking conditions are derived for asymmetrical S-C-S switching fabrics. In addition, wide-sense nonblocking conditions for switching fabrics with functional decomposition of center-stage switches are presented. It is shown that the wide-sense nonblocking switching fabrics may require less than a half of the switches in the middle-stage compared with the strict-sense nonblocking switching fabrics.

Rearrangeability of Wavelength-Space-Wavelength Switching Fabric Architecture for Elastic Optical Switches

In this paper, we consider the necessary and sufficient rearrangeability conditions for the wavelength-space-wavelength switching fabric, called the WSW2 switching fabric, for elastic optical network nodes. We derive and prove the lower bound of the necessary conditions for such switching fabrics to be rearrangeable. We also show the upper bound of the sufficient conditions for the rearrangeability of the switching fabric with two input and two output switches. For simultaneous connection routing, we propose a routing algorithm that always ends with success in switching fabrics, which fulfill the sufficient conditions. Finally, we extend the proposed algorithm to the switching fabric with $r$ input and $r$ output switches, and we derive the conditions under which this routing algorithm always ends in success. The required number of center-stage switches is significantly lower than that in the strict-sense nonblocking switching fabrics. To our knowledge, the proposed routing algorithm is the first one which ends with success for any connection set, within the provided sufficient conditions.

An Algorithmic Framework to Construct Optical Switch via Scaling From N-to-2N Ports for Optical Network on Chip

Optical network-on-chip (ONoC), designed with non-blocking optical switches, is gaining a significant research attention to meet the upcoming requirements of higher throughput, larger bandwidth, lower latency, and reduced power consumption for manycore processors. In this paper, we propose an algorithmic framework to construct non-blocking optical switches with multiple interconnection possibilities. The design is accomplished by scaling the optical switch from N-to-2N ports using two intermediate mapping matrices each of size 2N × 2N. These mapping matrices are the row and column permutations of identity matrices, where 1's represent passive interconnections among the optical switching units (OSUs). The proposed framework provides validation to the design by identifying all the non-blocking permutations of the mapping matrices, thereby, providing a flexibility to adopt any permutation as an interconnection scheme. Furthermore, it has the ability to quantify the redundancy in switching combinations which exists for any input-to-output routing, authenticates the non-blocking feature of the optical switch and to reduce the number of optical switching units while preserving the non-blocking characteristic. For the scaled 4 × 4 and 6×6 optical switches, we respectively identified 16 and 192 different interconnections to build multiple non-blocking switches. Moreover, a 20% reduction in OSUs is achieved by optimizing a 6 × 6 switch. The influence of the insertion loss, power consumption, and crosstalk noise on various scaled optical switch networks are also analyzed and compared with several existing optical switch topologies.