A Hierarchical WDM-Based Scalable Data Center Network Architecture

Abstract

Massive data centers are at the heart of the Internet. The rapid growth of Internet traffic and the abundance of rich data-driven applications have raised the need for enormous network bandwidth. Towards meeting this growing traffic demand, optical interconnects have gained significant attention, as they can provide high throughput, low latency, and scalability. In particular, optical Wavelength Division Multiplexing (WDM) provides the possibility to build data centers comprising of millions of servers, while providing hundreds of terabits per second bandwidth.In this paper, we propose a WDM-based Reconfigurable Hierarchical Optical Data Center Architecture (RHODA) that can satisfy future Internet traffic demands. To improve scalability, our DCN architecture is hierarchical, as it groups server racks into clusters. Cluster membership is reconfigurable through the use of optical switches. Each cluster enables heavy-traffic communication among the racks within. To support varying traffic patterns, the inter-cluster network topology and link capacities are also reconfigurable, which is achieved through the use of optical space switches and Wavelength Selective Switches (WSSs). Our simulation results demonstrate that in terms of average hop distance, RHODA outperforms OSA, FatTree and WaveCube by up to 81%, 66% and 60%, respectively.

RHODA Topology Configuration Using Bayesian Optimization

“…In this section, we briefly describe the RHODA architecture. The reader is referred to [24] for more design details of RHODA.…”

“…Moreover, large-scale optical switches consume less power per bit/s, making the network architecture scalable and energy-efficient. RHODA [24] is a WDM-based reconfigurable hierarchical optical DCN architecture. The architecture can scale up to 10+ million servers and support various traffic patterns.…”

“…The focus of this paper is on the design of the inter-cluster network topology to minimize the average number of hops (weighted by traffic demands). In [24], the inter-cluster topology is constructed using a Hungarian-based method [12]. The method first builds a circle/ring among clusters to first make the inter-cluster topology connected, and then the Hungarian algorithm is iteratively used to find a perfect bipartite matching to maximize the total traffic over the directly connected clusters (i.e., single-hop traffic).…”

“…Given the pair-wise cluster traffic, our objective for the cluster topology is to minimize the average hop length. In [24], a simple variant of the Hungarian algorithm that maximizes the one-hop or direct traffic among clusters was used. In this paper, we leverage the Bayesian Optimization (BO) framework and propose a fast algorithm to minimize the average number of hops in the inter-cluster network of RHODA.…”

A Survey on Optical Technologies for IoT, Smart Industry, and Smart Infrastructures

“…Not only submarine transmission systems but also optical transmission and switching elements within various network areas, namely access, aggregation, and core network areas, together with sophisticated network concepts and energy management solutions, have the potential to greatly improve the energy efficiency of the network infrastructure [55,71,72]. Similar technologies and concepts can also be used for internal interconnection networks in large data centers in order to improve both their energy efficiency and scalability [73][74][75].…”