Abstract
End-to-end (e2e) quality of service (QoS) is the key target to be achieved in order to successfully support high quality real-time services. While ISDN, SDH and ATM provide service classes that natively support constant bit-rate transmission tunnels, i.e. provide circuit switching that offers constant bandwidth for the entire duration of a service, packet switched technologies natively do not provide that option. The economically irreversible migration to the Internet protocol (IP) as transport plane for all kinds of services necessitates sophisticated traffic engineering to fulfil the basic service demands of real-time services being low latency, fast round-trip times (RTT), negligible jitter and minimal loss-rate. Most of the current approaches to optimise IP routing for real-time services target firstly at admission control and secondly on node internal priority scheduling. In this study we focus on the second part and try to evaluate the problems that arise when such node internal mechanisms get cascaded along a path, i.e. study the problems that need to be solved in order to derive e2e QoS from node performance in dynamic network scenarios. From the results mechanisms for dynamic admission control and priority scheduling in relation to current network and node state could be derived.
Original document
The different versions of the original document can be found in:
Document information
Published on 31/12/07
Accepted on 31/12/07
Submitted on 31/12/07
Volume 2008, 2008
DOI: 10.1109/icton.2008.4598686
Licence: CC BY-NC-SA license
Share this document
claim authorship
Are you one of the authors of this document?