A Hybrid Packet/Circuit Optical Transport Architecture for DCN

IEEE Network, 2000

ata centers (DCs) are currently the largest closedloop systems in the information technology (IT) and networking worlds, continuously growing toward multi-million-node clouds . DC operators manage and control converged IT and network infrastructures in order to offer a broad range of services and applications to their customers. Typical services and applications provided by current DCs range from traditional IT resource outsourcing (storage, remote desktop, disaster recovery, etc.) to a plethora of web applications (e.g., browsers, social networks, online gaming). Innovative applications and services are also gaining momentum to the point that they will become main representatives of future DC workloads. Among them, we can find high-performance computing (HPC) and big data applications . HPC encompasses a broad set of computationally intensive scientific applications, aiming to solve highly complex problems in the areas of quantum mechanics, molecular modeling, oil and gas exploration, and so on. Big data applications target the analysis of massive amounts of data collected from people on the Internet to analyze and predict their behavior.

Electronics, 2015

As cloud computing data centers grow larger and networking devices proliferate; many complex issues arise in the network management architecture. We propose a framework for multi-layer; multi-vendor optical network management using open standards-based software defined networking (SDN). Experimental results are demonstrated in a test bed consisting of three data centers interconnected by a 125 km metropolitan area network; running OpenStack with KVM and VMW are components. Use cases include inter-data center connectivity via a packet-optical metropolitan area network; intra-data center connectivity using an optical mesh network; and SDN coordination of networking equipment within and between multiple data centers. We create and demonstrate original software to implement virtual network slicing and affinity policy-as-a-service offerings. Enhancements to synchronous storage backup; cloud exchanges; and Fibre Channel over Ethernet topologies are also discussed.

Optics Express, 2015

We present a converged EPS and OCS switching fabric for data center networks (DCNs) based on a distributed optical switching architecture leveraging both WDM & SDM technologies. The architecture is topology adaptive, well suited to dynamic and diverse *-cast traffic patterns. Compared to a typical folded-Clos network, the new architecture is more readily scalable to future multi-Petabyte data centers with 1000 + racks while providing a higher link bandwidth, reducing transceiver count by 50%, and improving cabling efficiency by more than 90%.

—The rising popularity of cloud computing and big data applications has led to massive volumes of rack-to-rack traffic featuring diverse communication patterns denoted as *-cast that combines unicast, multicast, incast and all-to-all cast. Effective support of these traffic patterns in data center networks is still an open problem. We propose a hybrid (optical and electrical) approach that leverages physical layer optics to accelerate traffic delivery for each pattern. Our design consists of an application-driven control plane compatible with software defined networking (SDN) to dynamically configure the optics. We present the network architecture and control plane design and results on the multicast case.

2021

The optical interconnection and switching technology has recently gained ground in the data centers (DCs) globally. In this paper, we propose an optical network configuration to interconnect the servers of the same rack in a DC, based on passive optical functionalities. Thus, we focus in the intra-rack passive optical network and we propose the use of a number of dedicated wavelengths for the data communication among the servers through a passive optical coupler in the data plane. Meanwhile, we adopt a parallel control plane which is implemented in the optical domain too, in opposition to the majority of previous studies that performs the control communication over electrical links. Based on the software-defined networking (SDN) paradigm, we adopt at the Top-of-Rack (ToR) switch level, a rack controller which is responsible for the operational functionalities in the intra-rack network, such as medium access control (MAC) etc. Moreover, we introduce an intra-rack MAC protocol in conj...

The adoption of optical switching technologies in Data Centre Networks (DCNs) offers a solution for high speed traffic and energy efficiency in Data Centre (DC) operational management, enabling an easy scaling of DC infrastructures. Flexible, slotted allocation of optical resources is fundamental to efficiently support the dynamicity of DC traffic. In this context, the NEPHELE project proposes a Time Division Multiple Access approach for optical resource allocation, orchestrated through a Software Defined Networking controller which coordinates the DCN configuration based on real-time cloud application requests.

Journal of Optical Communications and Networking

Cloud services are fundamentally supported by data center networks (DCNs). With the fast growth of cloud services, the scale of DCNs is increasing rapidly, leading to great concern about system scalability due to multiple constraints. This paper proposes a scalable DCN architecture based on optical switching and transmission, with the distributed placement of optical switches and server racks at different nodes in a given optical network. This solves the scalability issue by relaxing power and cooling constraints and by reducing the number of (electronic) switches using high-capacity optical switches, as well as by simplifying DCN internal connections using wavelengths in the optical network. Moreover, the distributed optical switches provide service access interfaces to meet demand within areas, and thus reduce the transmission cost of the external traffic. The major concern is the additional delay and cost for remote transmissions of the DCN internal traffic. To this end, we study...

2018

The soaring traffic demands in datacenter networks (DCNs) are outpacing progresses in CMOS technology, challenging the bandwidth and energy scalability of currently established technologies. Optical switching is gaining traction as a promising path for sustaining the explosive growth of DCNs; however, its practical deployment necessitates extensive modifications to the network architecture and operation, tailored to the technological particularities of optical switches (i.e. no buffering, limitations in radix size and speed). European project NEPHELE is developing an optical network infrastructure that leverages optical switching within a software-defined networking (SDN) framework to overcome the bandwidth and energy scaling challenges of datacenter networks. An experimental validation of the NEPHELE data plane is reported based on commercial off-the-shelf optical components controlled by FPGA boards. To facilitate dynamic allocation of the network resources and perform collision-f...

2012 35th IEEE Sarnoff Symposium, 2012

Tremendous growth in data-intensive cloud applications have resulted in an increased demand for highly scalable data center network (DCN) architectures with high throughput and low network complexity. In this paper, we propose HyScaleII to improve the performance of HyScale [17]. HyScaleII is a switchcentric high performance hybrid optical network based DCN architecture that has most of the desirable properties of a data center, e.g. high scalability, low diameter, high bisection width, fault-tolerance, and low network complexity. We also present an efficient and simple routing scheme called HySII routing, which exploits the structural properties of HyScaleII. In our experiments, HyScaleII has lower packet loss ratio and higher average aggregate throughput by an average of 50% and 13.8% respectively as compared to HyScale [17].

2021

DCN architecture design that integrates fast-switching optics and space switches to allow dynamic reconfiguration and separation of intra- and inter-cluster connections. Simulation results show 20% improvement in average Flow Completion Time compared to traditional leaf-spine architectures.