Bandwidth Reservation Approach to Improve Quality of Service in Software-Defined Networking: A Performance Analysis
Autor: | Yu Song, Michael J. Watts, Amirhossein Moravejosharieh |
---|---|
Rok vydání: | 2018 |
Předmět: |
business.industry
computer.internet_protocol Computer science Quality of service Control (management) 020206 networking & telecommunications 02 engineering and technology Vertical integration Traffic flow (computer networking) Internet protocol suite Control system 0202 electrical engineering electronic engineering information engineering Bandwidth (computing) 020201 artificial intelligence & image processing business Software-defined networking computer Computer network |
Zdroj: | JCSSE |
DOI: | 10.1109/jcsse.2018.8457339 |
Popis: | Software-Defined Networking (SDN) is a new networking paradigm designed to resolve traditional IP network shortcomings by breaking the vertical integration of control and data planes. SDN separates the network control logic from underlying routers and switches and introduces the ability to program the network. Bandwidth reservation is an approach offered in SDN-enabled networks to guarantee relatively high Quality of Service for different types of media, e.g., video, audio or data. Although, this approach has been proven to be worthy of considering in SDN, there are still some concerns regarding its applicability in a relatively large networks. In this paper, we have evaluated the performance of bandwidth reservation approach in a relatively large-scaled SDN-enabled network in terms of its suitability when the number of users demanding for reserved bandwidth becomes larger. The obtained results from our simulation study show that bandwidth reservation can be beneficial only when the number of users asking for guaranteed bandwidth is relatively smaller than other users. Moreover, higher end-to-end QoS delivery can be achieved as an immediate outcome of deploying bandwidth reservation approach for a particular type of traffic flow, however, at the cost of incurring negative impact on other types of traffic flow in terms of achievable network throughput. |
Databáze: | OpenAIRE |
Externí odkaz: |