Web-caches play an important role in the architecture of a Dynamic Adaptive Streaming over HTTP (DASH) system, which can bring video files near to the clients. In such systems, the placement of the caches affects the performance of the video streaming applications. In this paper we select the optimal network nodes for cache placement by taking into account characteristics such as the bandwidth of the paths, the places, and the number of the online clients, for a given network graph. For this purpose, we use a placement algorithm which is enhanced version of the Pressure algorithm proposed by Clegg which we call PressureCache, in an implementation of virtualized DASH Aware Network Elements (vDANEs). The results show that the proposed approach gives better performance in terms of received quality than random cache placement.
The demand for playing the video with the highest quality possible leads video streaming companies to utilize network-based solutions and server assistance. As being one of the popular technology in video streaming area, HTTP adaptive video streaming systems can benefit from the emerging network technologies such as SDN and NFV. Such multimedia systems require the management of network resources efficiently to be able to provide acceptable service. In this paper, we propose a multimedia service management architecture that focuses on migration one of the most crucial components of such systems, virtualized caches. SDN controller gathers underlay network statistics and based on real-time network condition determines the optimal location for virtual cache and/or migrate it when network conditions change. We focus on hand-off affects the perceptual quality when clients migrate from one cache to another. The proposed approach was evaluated using the Mininet Emulator and its performance was given comparatively. The performance results show that the combination of SAND (Server and Network Assisted DASH) features with the advantage of the NFV-SDN technology can provide good service to its users.
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.