1
Architectures of Data Center Networks: Overview

Boutheina Dab1, 2, Ilhem Fajjari3, Dallal Belabed4, and Nadjib Aitsaadi5

1 VMware, Hauts-de-Seine, La Defense, France

2 LiSSi Lab, UPEC, Val de Marne, Vitry sur Seine, France

3 Orange Labs, Orange, Hauts-de-Seine, Chatillon, France

4 Airbus Defense and Space, Airbus, Saint-Quentin en Yvelines, Elancourt, France

5 Universités Paris-Saclay, UVSQ, DAVID, F-78035, Versailles, France

Abstract

To deal with the widespread use of cloud services and the unprecedented traffic growth, the scale of the Data Center has importantly increased. Therefore, it is crucial to design novel efficient network architectures able to satisfy the requirements on bandwidth. As a key physical infrastructure, Data Center Network (DCN) designing has widely been a hot research focus.

This chapter reviews the main DCN architectures propounded in the literature. To do so, a taxonomy of DCN designs will be proposed, while analyzing in depth each structure of the given classification. Then, we will provide a qualitative comparison between these different DCN groups. Finally, we will present hybrid DCN architecture based on wired and wireless architecture.

1.1 Taxonomy of DCN Architectures

In this section, we present a taxonomy of the existent Data Center Network (DCN) architectures with a detailed review of each drawn class. In general, several criteria have to be considered to design robust DCNs, namely, high network performance, efficient resource utilization, full available bandwidth, high scalability, easy cabling, etc. To deal with the aforementioned challenges, a panoply of solutions have been designed. Mainly, we can distinguish two research directions. In the first one, wired DCN architectures have been upgraded to build advanced cost-effective topologies able to scale up data centers. The second approach has resorted to deploying new network techniques within the existing DCN so as to handle the challenges encountered in the prior architectures. Hereafter, we will give a detailed taxonomy of these techniques.

1.1.1 Classification of DCN Architectures

With regard to the aforementioned research directions, we can identify three main groups of DCN architectures, namely, switch-centric DCN, server-centric DCN, and enhanced DCN. Each group includes a variety of categories that we will detail hereafter.

• Switch-centric DCN architecture: switches are, mostly, responsible for network-related functions, whereas the servers handle processing tasks. The focus of such a design is to improve the topology so as to increase network scale, reduce oversubscription, and speed up flow transmission. Switch-centric architectures can be classified into three main categories according to their structural properties:
  1. Traditional tree-based DCN architecture: represents a specific kind of switch-centric architecture, where switches are linked in a multirooted form.
  2. Hierarchic DCN architecture: is a switch-centric DCN, where network components are arranged in multiple layers. Each layer characterizes traffic differently.
  3. Flat DCN architecture: compresses the three switch layers into only one or two switch layers, in order to simplify the management and maintenance of the DCN.
• Server-centric DCN architecture: servers are enhanced to handle networking functions, whereas switches are used only to forward packets. Basically, servers are simultaneously end-hosts and relaying nodes for multihop communications. Usually, server-centric DCN are recursively defined multilevel topologies.
• Enhanced DCN architecture: is a specific DCN which is tailored for future Cloud computing services. Indeed, the future research direction attempts to deploy networking techniques so as to deal with wired DCN designs limitations. Recently, a variety of technologies have been used in this context, namely, optical switching and wireless communications. Accordingly, we distinguish two main classes of enhanced DCN architectures:

  Figure 1.1 Taxonomy of DCN architectures.

  1. Optical DCN: makes use of optical devices to speed up communications. It can be either: (i) all-optical DCN (i.e. with completely optical devices) or (ii) hybrid optical DCN (i.e. both optical and Ethernet switches).
  2. Wireless DCN: deploys wireless infrastructure in order to enhance network performance, and may be: (i) fully wireless DCN (i.e. only wireless devices) or (ii) Hybrid DCN (i.e. both wireless and wired devices).

Figure 1.1 illustrates the taxonomy of current DCN architectures. In the following, we will detail each category and discuss their impact on Cloud computing performance.

1.1.2 Switch-Centric DCN Architectures Overview

1.1.2.1 Tree-Based DCN

The traditional DCN is typically based on a multiroot tree architecture. The latter is a three-tier topology composed by three layers of switches. The top level (i.e. root) represents the core layer, the middle level is the aggregation layer, while the bottom level is known as the access layer. The core devices are characterized by high capacities compared with aggregation and access switches. Typically, the core switches' uplinks connect the data center to the Internet. On the other hand, the access layer switches commonly use 1 Gbps downlink interfaces and 10 Gbps uplink interfaces, while aggregation switches provide 10 Gbps links. Access switches (i.e. top of rack, ToRs) interconnect servers in the same rack. Aggregation layer allows the connection between access switches and the data forwarding. It is worth noting that the above values of network interface cards throughput are continuously increasing. For instance, nowadays it is easy and not really expensive to deploy interfaces with 25 and 100 Gbps. An illustration of tree-based DCN architecture is depicted in Figure 1.2.

Figure 1.2 Traditional tree-based DCN architecture.

Unfortunately, traditional DCNs struggle to resist to the increasing traffic demand. First, core switches are prone to bottlenecks issues as soon as the workloads reach the peak. Moreover, in such a DCN, several downlinks of a ToR switch share the same uplink which limits the available bandwidth. Second, DCN scalability strongly depends on the number of switch ports. Therefore, the unique way to scale this topology is to increase the number of network devices. However, these solutions results in high construction costs and energy consumption. Third, tree-based DCN suffers from serious resiliency problems. For instance, if a failure happens on some of the aggregation switches, then servers are likely to lose connection with others. In addition, resource utilization is not efficiently balanced. For all the aforementioned reasons, researchers put forward alternative DCN topologies.

1.1.2.2 Hierarchical DCN Architecture

Hierarchical topology arranges the DCN components in multiple layers. The key insight behind this model is to reduce the congestion by minimizing the oversubscription in lower-layer switches using the upper-layer devices. In the literature, we find several hierarchic DCN examples, namely, CLOS, FatTree, and VL2. Hereafter, we will describe each one of them.

CLOS-Based DCN Is an advanced tree-based network architecture. It was, first, introduced by Charles Clos, from Bell Labs, in 1953 to create nonblocking multistage topologies, able to provide higher bandwidth than a single switch. Typically, CLOS-based DCNs come with three layers of switches: (i) access layer (ingress), composed of the ToRs switches, directly connected to servers in the rack; (ii) aggregation layer (middle), formed by aggregation switches referred as spines and connected to the ToRs; and (iii) core layer (egress), formed by core switches serving as edges to manage traffic in and out the DCN (Chen et al., 2016).

The CLOS network has been widely used to build modern IP fabrics, generally referred to as spine and leaf topologies. Accordingly, in this kind of DCN, commonly named folded-CLOS topology, the spine layer represents the aggregation switches (i.e. spines), while the leaf layer is composed of the ToR switches (i.e. leaves). In other words, in CLOS topology, (i) leaf layer is composed of ToR switches and (ii) spine layer is composed of aggregation switches. The spine layer is responsible for interconnecting leafs. CLOS inhibits the transition of traffic through horizontal links (i.e. inside the same layer). Moreover, CLOS topology scales up the number of ports and makes possible huge connection using only a small number of switches. Indeed, augmenting the switches ports enhances the spine layer width and, hence, alleviates the network congestion. In general, each leaf switch is connected to all spines. In other words, the number of up (respectively, down) ports of each ToR is equal to the number of spines (respectively, leaves). Accordingly, in a DCN of leaves and spines, there are wired links. The main reason behind this link redundancy is to enable...