Publication: Load Balancing in Software-Defined Data Centre with Fat Tree Architecture
Date
2022
Authors
Hacham S.
Din N.M.
Balasubramanian N.
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers Inc.
Abstract
Software-defined data centres have emerged as an automated and virtualized platform for the management of networked computer servers in providing services to organizations and customers. Given the fast increase in demand by different applications on storage, the challenge is to keep up with performance requirements by avoiding bottlenecks in the network. Some examples of these applications are Google Drive, YouTube, and Megaupload. Data centre systems are intended to fulfill the information transmission request of tightly interrelated hosts with the network topology and routing devices playing a part in influencing the latency and other network performances. At present, the fat-Tree network is among the most generally utilized topologies for data centre systems. As traffic demand increases in the data centres, the traffic load needs to be shared between servers and inevitably load balancing is required. In software-defined data centres with software-defined networks, the key to load balancing is in the form of smart congestion-conscious routing. It is an important element in promoting the accessibility and adaptability that further prompts to accomplish a minimal reaction application time. This paper reviews load balancing techniques in software-defined data centres with the fat-Tree architecture. Load balancing techniques have become smarter and more effective over the years with reductions on overhead statistical collection and improvement in the quality of service. Nevertheless, focus on optimal energy usage is another dimension that can be investigated in promoting green computing. � 2022 IEEE.
Description
Digital storage; Forestry; Information management; Network architecture; Network routing; Software defined networking; Traffic congestion; Trees (mathematics); Computer servers; Data center systems; Datacenter; Fat trees; Load balancing technique; Load-Balancing; Networked computers; Quality-of-service; Software-defined data center; Tree architectures; Quality of service