Comparison of Virtualization Algorithms and Topologies for Data Center Networks

Yedder, Hanene Ben; Ding, Qingye; Zakia, Umme; Li, Zhida; Haeri, Soroush; Trajković, Ljiljana

doi:10.1109/icccn.2017.8038524

Cited by 3 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DCN has been built over the years with various kind of topologies. The research work of [3] [5] [8] [10] [11] [83], has shown how vital the topology used in building a DCN is essential to energy savings, as well as its performance. Hence, based on these works and Cisco's recommendation [4] [91], the topology used in these thesis work, was the leafspine topology.…”

Section: Motivationmentioning

confidence: 99%

“…From various surveys carried out on DCN topologies, we can say that there are two major architectures used, the switch-centric, and the server-centric topologies [5] [84] [85] [86] [87]. It can be summarized, that the most preferred topology used for building DCNs which integrates the use of virtualization technologies, with better network performance as to throughput and congestions, scalability, and latency is the switch-centric topology (a DCN topology such that the routing technique is centered on the switches) [83][84][85][86][87]. In this thesis, the clos topology (A multistage circuit switching systems used for data centers architecture) is used interchangeably for the switch-centric topology based on the original design for a clos topology used for DCN (although not all clos topologies are switchcentric).…”

Section: Energy Distribution In Dcnsmentioning

confidence: 99%

“…A recent study in [83], focused on the performance of various DCN topologies in the switch-centric family, shows that for building a DCN consisting of the same number of hosts, VMs, and leaf switches, the leaf-spine infrastructure, which is a two-tier topology, consumes less energy compared to the traditional 3-tier clos topologies, such as Fat-tree, F 2 -Tree, Diamond, etc. A further explanation the switch-centric topologies is provided in Section 2.2 and 2.3 of this thesis.…”

Section: Energy Distribution In Dcnsmentioning

confidence: 99%

“…Of the clos family DCN, the switch-centric topology is widely used by many data centers of recent, e.g., Google (google Fat-tree), Facebook (leaf-spine, two-tier), Microsoft [83] [87], etc. The reason is its improved performance in integrating virtualization technology for servers into the DCN, compared to the server-centric topologies.…”

Section: Dcn Switch-centric Topologiesmentioning

confidence: 99%

“…It should support { } servers, which results in high scalability [4] [56] (where k is the number of ports on the switches). Tables 2.1 -2.3 show essential metrics adapted from both [3] & [83] from the experiments in comparing these DCN topologies. summarized the efficiency of leaf-spine topology in a DCN over other topologies in the clos family for deploying virtualization technology.…”

Section: Comparing the Leaf-spine Over Some Switch-centric Topology Imentioning

confidence: 99%

See 4 more Smart Citations

Segment Routing Green Spine Switch Management System for DCN

Osamudiamen¹

View full text Add to dashboard Cite

It is without a doubt that the data center is the core of most IT industries, providing services to billions of users today. The preferred data center network (DCN) architecture used for this platform is the 2-tier leaf-spine topology, which is not only energy efficient but has a good performance metrics. Furthermore, due to the exponential increase of information and services from datacenters, it is projected that a 14.1 zeta bits of bandwidth would be needed to meet the current demands of data by the end of 2020 in USA data centers alone. This would require a tremendous amount of energy to run those data centers.Hence, in recent years there is more focus on reducing the energy consumption in a data center. In trying to reduce the energy consumption of a data center, this thesis approaches the problem by utilizing Segment Routing-Traffic Engineering, and other power algorithms for an efficient DCN. The proposed approach makes it possible to deactivate spine switches and links on the data center networks, which results in energy savings. Our experimental results yielded an approximate of 80% in energy savings of the spine switches consumption whilst maintaining an average similar traffic performance.iii

show abstract

Section: Motivationmentioning

confidence: 99%

Section: Energy Distribution In Dcnsmentioning

confidence: 99%

Section: Energy Distribution In Dcnsmentioning

confidence: 99%

Section: Dcn Switch-centric Topologiesmentioning

confidence: 99%

Section: Comparing the Leaf-spine Over Some Switch-centric Topology Imentioning

confidence: 99%

See 3 more Smart Citations

Segment Routing Green Spine Switch Management System for DCN

Osamudiamen¹

View full text Add to dashboard Cite

show abstract

Fast and isolation guaranteed coflow scheduling via traffic forecasting in multi-tenant environment

Li,

Zhang,

Yang

et al. 2024

J Supercomput

View full text Add to dashboard Cite

Virtual Network Embedding for Switch-Centric Data Center Networks

Rios

Bekshentayeva

Singh

et al. 2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

Self Cite

View full text Add to dashboard Cite

Advances in software defined and data center networks have enabled network virtualization. Virtual network embedding increases resources utilization and reduces cost of network deployment. Its performance depends on embedding algorithms and data center network topologies. In this paper, we evaluate performance of virtual network embedding algorithms based on acceptance ratio, revenue to cost ratio, and node and link utilizations by simulating virtual network embeddings on Spine-Leaf, Three-Tier, and Collapsed Core data center network topologies.

show abstract

Comparison of Virtualization Algorithms and Topologies for Data Center Networks

Cited by 3 publications

References 25 publications

Segment Routing Green Spine Switch Management System for DCN

Segment Routing Green Spine Switch Management System for DCN

Fast and isolation guaranteed coflow scheduling via traffic forecasting in multi-tenant environment

Virtual Network Embedding for Switch-Centric Data Center Networks

Contact Info

Product

Resources

About