Parameter analysis for a push/pull-based P2P live video streaming application

Abstract

Özetçe-Bu bildiride, tampon harita değişim periyodunun ve düğümlerin sahip olduğu ortak sayısının görevdeş ağlarda bir itme/çekme tabanlı canlı video akışlandırma uygulaması üzerindeki etkisi ortaya konmuştur. Oluşturulan uygulama, network simulator-3 simülasyon ortamı kullanılarak gerçeklenmiş ve ilgili parametrelerin video akışlandırma performansı üzerindeki etkisi gerçekleştirilen deneylerle gösterilmiştir. Sonuçlar göstermektedir ki, tampon harita değişim periyodu sabit tutulduğunda düğümlerin sahip olduğu ortak sayısının artışıyla birlikte izlenen video kalitesi ve kontrol mesajlarının maliyeti artmaktadır. Öte yandan, düğümlerin sahip olduğu ortak sayısı sabit tutulduğunda kontrol mesaj maliyetinin ve video kalitesinin optimale yakın olduğu bir tampon harita değişim periyodu bulunabilmektedir. Anahtar Kelimeler-video akışlandırma; görevdeş ağlar; itme/çekme tabanlı video akışlandırma uygulamaları; ortak sayısı; tampon haritası değişim periyoduAbstract-In this paper, the impact of both the buffermap exchange period and partnership size of peers has been presented in a push/pull-based peer-to-peer (P2P) live video streaming application. The application has been implemented on network simulator-3 and the impact of aforementioned parameters on video streaming performance has been introduced via the simulation test results. The test results show that if the buffermap exchange period is fixed, received video quality and the control overhead increase with the increment in partnership size. On the other hand, if the partnership size of the nodes is fixed, a buffermap exchange period can be found for near-optimal video quality and control overhead. Keywords-video streaming; peer-to-peer networks; push/pull-based video streaming applications; number of partners; buffermap exchange period

Engineering a multi-agent system for peer-to-peer video streaming

“…Considering the complementary studies, the inference mechanisms of the agents described in this paper are given in our previous work [13]. Recommendations for the selection of some parameters related to mesh based P2P system can be found in our another previous work [12].…”

“…All the simulations were run over 1800 seconds. As also discussed in [12], the partnership List size is set to 15 and Membership List size is set to 30. There is one server which is the source of the video and one tracker server which is assumed to have all network information.…”

A path selection approach with genetic algorithm for P2P video streaming systems

“…The average bandwidth distribution of links in the system is in accordance with the values given in [19]. The values of the parameters used in the simulations are given in are selected according to the optimal values given in [18] and [17]. The available bandwidth of internal links and the available bandwidth of external links represent the available bandwidth values remaining from the other traffic on the internal and external links, respectively.…”

“…The partnership is achieved mutually and buffermap exchange is realized within 2 s' periods between the partners. (1)) 500 chunks T p (used in formula (2)) 500 chunks Generation count [10][11][12][13][14][15][16][17][18][19][20] The peer joining the system plays the video after it buffers for up to 8 s. During the streaming, if a peer decides to choose a new parent and if this selection is not successful, i.e., the received video bitrate is not adequate to play the video properly, and candidate parents cannot be found, some partners are deleted from the list and new partners are selected from the membership list. If this problem continues for 20 s, the peer resets itself and joins the system again.…”