Server-centric Network Structure Research Articles

The vertex-pancyclicity of data center networks

The k-dimensional data center network with n-port switches, denoted by Dk,n, has been proposed for data centers as a server centric network structure. Wang et al. had proved that Dk,n is Hamiltonian-connected for k≥0 and n≥2 except for (k,n)=(1,2). In this paper, we prove that for any edge (u,v) in a complete subgraph Kn of Dk,n, there is a cycle of every length from 3 to |V(Dk,n)| passing through the edge (u,v) in Dk,n for n≥5. As corollaries, we obtained that Dk,n is vertex-pancyclic and pancyclic.

BCube and DCell Topology Data Center Infrastructures Performance

Data Center (DC) quality is if it can provide optimal service. Optimal service means that the access provided should be free from interference. Interference can come from outside the system or from within the system itself. One of the variables used to measure the performance of data center architecture is a routing protocol. Routing Information Protocol is a variable to measure performance on BCube and DCell topologies. Testing is done using a simulator that can produce information architecture that has the best performance when communicating from the server to the client. BCube is a server-centric network structure, while DCell is a recursively defined architecture. This study was conducted to compare the two data center topologies with experimental methods. The results show that DCell has a better speed in data transmission. However, when viewed from the side of security and integrity of BCube delivery much better, this is because BCube is able to send data completely without any failure or inhibition in the delivery.

Conditional edge-fault-tolerant Hamiltonicity of the data center network

The k-dimensional data center network with n port switches, denoted by Dk,n, has been proposed for data centers as a server centric network structure. Wang et al. (2015) had shown that Dk,n is (n+k−3)-fault-tolerant Hamiltonian. In this paper, we consider more faulty edges and prove that Dk,n is conditional (2n+2k−9)-edge-fault-tolerant Hamiltonian for any k≥0 and n≥2 except k=1 and n≥6. Moreover, the upper bound 2n+2k−9 of |F| is optimal.

A Novel Network Structure with Power Efficiency and High Availability for Data Centers

Designing a cost-effective network for data centers that can deliver sufficient bandwidth and provide high availability has drawn tremendous attentions recently. In this paper, we propose a novel server-centric network structure called RCube, which is energy efficient and can deploy a redundancy scheme to improve the availability of data centers. Moreover, RCube shares many good properties with BCube, a well known server-centric network structure, yet its network size can be adjusted more conveniently. We also present a routing algorithm to find paths in RCube and an algorithm to find multiple parallel paths between any pair of source and destination servers. In addition, we theoretically analyze the power efficiency of the network and availability of RCube under server failure. Our comprehensive simulations demonstrate that RCube provides higher availability and flexibility to make trade-off among many factors, such as power consumption and aggregate throughput, than BCube, while delivering similar performance to BCube in many critical metrics, such as average path length, path distribution and graceful degradation, which makes RCube a very promising empirical structure for an enterprise data center network product.

BCube

This paper presents BCube, a new network architecture specifically designed for shipping-container based, modular data centers. At the core of the BCube architecture is its server-centric network structure, where servers with multiple network ports connect to multiple layers of COTS (commodity off-the-shelf) mini-switches. Servers act as not only end hosts, but also relay nodes for each other. BCube supports various bandwidth-intensive applications by speeding-up one-to-one, one-to-several, and one-to-all traffic patterns, and by providing high network capacity for all-to-all traffic. BCube exhibits graceful performance degradation as the server and/or switch failure rate increases. This property is of special importance for shipping-container data centers, since once the container is sealed and operational, it becomes very difficult to repair or replace its components. Our implementation experiences show that BCube can be seamlessly integrated with the TCP/IP protocol stack and BCube packet forwarding can be efficiently implemented in both hardware and software. Experiments in our testbed demonstrate that BCube is fault tolerant and load balancing and it significantly accelerates representative bandwidth-intensive applications.