Popular Internet applications such as web browsing, web video download or variable-rate voice suffer from standard Transport Control Protocol (TCP) behaviour because their transmission rate and pattern are different from conventional bulk transfer applications. Previous works have analysed the interaction of these applications with the congestion control algorithms in TCP and proposed Congestion Window Validation (CWV) as a solution. However, this method was incomplete and has been shown to present drawbacks. This paper focuses on the ‘newCWV’ which was proposed to address these drawbacks. newCWV depicts a practical mechanism to estimate the available path capacity and suggests a more appropriate congestion control behaviour. These new modifications benefit variable-rate applications that are bursty in nature, with shorter transfer durations. In this paper, this algorithm was implemented in the Linux TCP/IP stack and tested by experiments, where results indicate that, with newCWV, the browsing can get 50% faster in an uncongested network.
Popular Internet applications such as web browsing, and web video download use HTTP protocol as application over the standard Transport Control Protocol (TCP). Traditional TCP behavior is unsuitable for this style of application because their transmission rate and traffic pattern are different from conventional bulk transfer applications. Previous works have analyzed the interaction of these applications with the congestion control algorithms in TCP and the proposed Congestion Window Validation (CWV) as a solution. However, this method was incomplete and has been shown to present drawbacks. This paper focuses on the ‘newCWV’ which was designed to address these drawbacks. NewCWV provides a practical mechanism to estimate the available path capacity and suggests a more appropriate congestion control behavior. This paper describes how this algorithm was implemented in the Linux TCP/IP stack and tested by experiments, where results indicate that, with newCWV, the browsing can get 50% faster in an uncongested network.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.