P-Cube: A New Two-Layer Topology for Data Center Networks Exploiting Dual-Port Servers

Moeen AL-MAKHLAFI; Huaxi GU; Xiaoshan YU; Yunfeng LU

doi:10.1587/transcom.2019EBP3219

IEICE TRANSACTIONS on Communications

Open Access
P-Cube: A New Two-Layer Topology for Data Center Networks Exploiting Dual-Port Servers

Moeen AL-MAKHLAFI, Huaxi GU, Xiaoshan YU, Yunfeng LU

Full Text Views

53

Cite this

Free PDF (2.1MB)

Summary :

Connecting a large number of servers with high bandwidth links is one of the most crucial and challenging tasks that the Data Center Network (DCN) must fulfill. DCN faces a lot of difficulties like the effective exploitation of DC components that, if highlighted, can aid in constructing high performance, scalable, reliable, and cost-effective DCN. In this paper, we investigate the server-centric structure. We observe that current DCs use servers that mostly come with dual ports. Effective exploitation of the ports of interest for building the topology structure can help in realizing the potentialities of reducing expensive topology. Our new network topology, named “Parallel Cubes” (PCube), is a duplicate defined structure that utilizes the ports in the servers and mini-switches to form a highly effective, scalable, and efficient network structure. P-Cube provides high performance in network latency and throughput and fault tolerance. Additionally, P-Cube is highly scalable to encompass hundreds of thousands of servers with a low stable diameter and high bisection width. We design a routing algorithm for P-Cube network that utilizes the P-Cube structure to strike a balance among the numerous links in the network. Finally, numerical results are provided to show that our proposed topology is a promising structure as it outperforms other topologies and it is superior to Fat-tree, BCube and DCell by approximately 24%, 16%, 8% respectively in terms of network throughput and latency. Moreover, P-Cube extremely outperforms Fat-tree, and BCube structures in terms of total cost, complexity of cabling and power consumption.

Publication: IEICE TRANSACTIONS on Communications Vol.E103-B No.9 pp.940-950

Publication Date: 2020/09/01

Publicized: 2020/03/03

Online ISSN: 1745-1345

DOI: 10.1587/transcom.2019EBP3219

Type of Manuscript: PAPER

Category: Network

Authors

Moeen AL-MAKHLAFI
  Xidian University
Huaxi GU
  Xidian University
Xiaoshan YU
  Xidian University
Yunfeng LU
  Xidian University

Keyword

data center networking, P-Cube topology, server-centric network, routing algorithm

Cite this

Copy

Moeen AL-MAKHLAFI, Huaxi GU, Xiaoshan YU, Yunfeng LU, "P-Cube: A New Two-Layer Topology for Data Center Networks Exploiting Dual-Port Servers" in IEICE TRANSACTIONS on Communications, vol. E103-B, no. 9, pp. 940-950, September 2020, doi: 10.1587/transcom.2019EBP3219.
Abstract: Connecting a large number of servers with high bandwidth links is one of the most crucial and challenging tasks that the Data Center Network (DCN) must fulfill. DCN faces a lot of difficulties like the effective exploitation of DC components that, if highlighted, can aid in constructing high performance, scalable, reliable, and cost-effective DCN. In this paper, we investigate the server-centric structure. We observe that current DCs use servers that mostly come with dual ports. Effective exploitation of the ports of interest for building the topology structure can help in realizing the potentialities of reducing expensive topology. Our new network topology, named “Parallel Cubes” (PCube), is a duplicate defined structure that utilizes the ports in the servers and mini-switches to form a highly effective, scalable, and efficient network structure. P-Cube provides high performance in network latency and throughput and fault tolerance. Additionally, P-Cube is highly scalable to encompass hundreds of thousands of servers with a low stable diameter and high bisection width. We design a routing algorithm for P-Cube network that utilizes the P-Cube structure to strike a balance among the numerous links in the network. Finally, numerical results are provided to show that our proposed topology is a promising structure as it outperforms other topologies and it is superior to Fat-tree, BCube and DCell by approximately 24%, 16%, 8% respectively in terms of network throughput and latency. Moreover, P-Cube extremely outperforms Fat-tree, and BCube structures in terms of total cost, complexity of cabling and power consumption.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2019EBP3219/_p

Copy

@ARTICLE{e103-b_9_940,
author={Moeen AL-MAKHLAFI, Huaxi GU, Xiaoshan YU, Yunfeng LU, },
journal={IEICE TRANSACTIONS on Communications},
title={P-Cube: A New Two-Layer Topology for Data Center Networks Exploiting Dual-Port Servers},
year={2020},
volume={E103-B},
number={9},
pages={940-950},
abstract={Connecting a large number of servers with high bandwidth links is one of the most crucial and challenging tasks that the Data Center Network (DCN) must fulfill. DCN faces a lot of difficulties like the effective exploitation of DC components that, if highlighted, can aid in constructing high performance, scalable, reliable, and cost-effective DCN. In this paper, we investigate the server-centric structure. We observe that current DCs use servers that mostly come with dual ports. Effective exploitation of the ports of interest for building the topology structure can help in realizing the potentialities of reducing expensive topology. Our new network topology, named “Parallel Cubes” (PCube), is a duplicate defined structure that utilizes the ports in the servers and mini-switches to form a highly effective, scalable, and efficient network structure. P-Cube provides high performance in network latency and throughput and fault tolerance. Additionally, P-Cube is highly scalable to encompass hundreds of thousands of servers with a low stable diameter and high bisection width. We design a routing algorithm for P-Cube network that utilizes the P-Cube structure to strike a balance among the numerous links in the network. Finally, numerical results are provided to show that our proposed topology is a promising structure as it outperforms other topologies and it is superior to Fat-tree, BCube and DCell by approximately 24%, 16%, 8% respectively in terms of network throughput and latency. Moreover, P-Cube extremely outperforms Fat-tree, and BCube structures in terms of total cost, complexity of cabling and power consumption.},
keywords={},
doi={10.1587/transcom.2019EBP3219},
ISSN={1745-1345},
month={September},}

Copy

TY - JOUR
TI - P-Cube: A New Two-Layer Topology for Data Center Networks Exploiting Dual-Port Servers
T2 - IEICE TRANSACTIONS on Communications
SP - 940
EP - 950
AU - Moeen AL-MAKHLAFI
AU - Huaxi GU
AU - Xiaoshan YU
AU - Yunfeng LU
PY - 2020
DO - 10.1587/transcom.2019EBP3219
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E103-B
IS - 9
JA - IEICE TRANSACTIONS on Communications
Y1 - September 2020
AB - Connecting a large number of servers with high bandwidth links is one of the most crucial and challenging tasks that the Data Center Network (DCN) must fulfill. DCN faces a lot of difficulties like the effective exploitation of DC components that, if highlighted, can aid in constructing high performance, scalable, reliable, and cost-effective DCN. In this paper, we investigate the server-centric structure. We observe that current DCs use servers that mostly come with dual ports. Effective exploitation of the ports of interest for building the topology structure can help in realizing the potentialities of reducing expensive topology. Our new network topology, named “Parallel Cubes” (PCube), is a duplicate defined structure that utilizes the ports in the servers and mini-switches to form a highly effective, scalable, and efficient network structure. P-Cube provides high performance in network latency and throughput and fault tolerance. Additionally, P-Cube is highly scalable to encompass hundreds of thousands of servers with a low stable diameter and high bisection width. We design a routing algorithm for P-Cube network that utilizes the P-Cube structure to strike a balance among the numerous links in the network. Finally, numerical results are provided to show that our proposed topology is a promising structure as it outperforms other topologies and it is superior to Fat-tree, BCube and DCell by approximately 24%, 16%, 8% respectively in terms of network throughput and latency. Moreover, P-Cube extremely outperforms Fat-tree, and BCube structures in terms of total cost, complexity of cabling and power consumption.
ER -

IEICE TRANSACTIONS on Communications