Stability in heterogeneous dynamic multimedia networks

1Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Internet and other multimedia packet-switched networks are heterogeneous due to the simultaneous running (composition) of different contention-resolution protocols over different network hosts and the existence of various types of network links. Also, real networks are dynamic in their nature due to intentional or unintentional changes on network link service rates or tra nsient link failures. Our interest is focused on FIFO compositions with other contention-resolution protocols due to the FIFO popularity for offering best-effort services in packet-switched networks. A packet-switched network is stable, if the number of packets in the network remains bounded at all times against any adversary. We use an enhanced adversarial framework that is based on an adversary that controls packet injection rates, along with packet paths, and manipulates link slowdowns or capacities. Within this framework, we study the impact of specific compositions of FIFO with other protocols on the network stability using as a test-bed specific network topologies which have been proved forbidden for stability for a single protocol, fixed link slowdowns/ capacities and packet paths without repeated links/edges. Our results suggest that the instability behavior of a network using FIFO compositions under adversarial attacks, that dynamically change link slowdowns/ capacities, is not only maintained, but, also, may become worse than in the case of attacks that do not change slowdowns/capacities or when a single protocol, like FIFO, is employed on all network queues for contention-resolution. We believe that this study can advance the research for the provision of trustworthy heterogeneous networks.

Cite

CITATION STYLE

APA

Koukopoulos, D. (2015). Stability in heterogeneous dynamic multimedia networks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9295, pp. 261–280). Springer Verlag. https://doi.org/10.1007/978-3-319-24024-4_16

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free