Criso: An Incremental Scalable and Cost-Effective Network Architecture for Data Centers

Abstract

With the explosive data growth, an enormous number of computing and networking components (e.g., servers, switches, and wires) are continuously being augmented to data centers. Data center networks (DCNs) - maintaining a high network capacity - must be cost efficient, incrementally scalable, and fault-tolerant. To address these challenges, we propose in this study a new type of DCN architecture referred to as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> . Different from the existing network architectures, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> is designed hierarchically and recursively by employing two ports servers and commodity switches. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> is constructed based on numerous isomorphic <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">pod</i> s, each of which leverages external interfaces supplied by switches to connect with neighboring pods. Additionally, a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">pod</i> -based and fault-tolerant routing algorithm is designed to handle multiple failures. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> has an array of promising features, including being cost-efficient and delivering a high-network capacity that can be extended to millions of nodes. The analytic results demonstrate that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> is significantly superior to the four state-of-the-art data center structures in terms of network capacity, scalability, cost, power consumption, and other static characteristics. Furthermore, the experimental results unveil that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> satisfies the fault-tolerant demands of modern data centers. Compared to the four existing topologies (i.e., <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DCell</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">BCube</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FiConn</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Fat-Tree</i> ) that have been widely investigated, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Criso</i> is adroit at maintaining a balanced performance in terms of throughput and latency.