Connection caching to reduce signaling loads with applications to softswitch telephony
Computer Networks: The International Journal of Computer and Telecommunications Networking
Exploiting the temporal dimension in QoS-aware multicast routing
Computer Communications
| Hi-index | 0.00 |
Quality of Service (QoS) routing algorithms have become thefocus of recent research due to their potential for increasing theutilization of an Integrated Services Packet Network (ISPN) servingrequests with QoS requirements. While heuristics for determiningpaths for such requests have been formulated for a variety of QoSmodels, little attention has been given to the overall processingcomplexity of the QoS routing protocol. This paper deals with theprocessing complexity of determining QoS paths in link state basedrouting architectures. Although on-demand path computation is veryattractive due to its simplicity, many believe that its processingcost will be prohibitive in environments with high request rates.In this work, we first characterize the processing cost of QoSrouting algorithms that use the widest-shortest path heuristic.Then we study alternatives to on-demand path computation that canreduce this processing overhead. In addition to the well knownsolution of path pre-computation, we introduce and study pathcaching, an incremental modification of on-demand path computation.By simulating realistic topologies and traffic conditions weinvestigate the performance of both alternatives. Our results showthat caching is an effective alternative to path pre-computationand that both path caching and pre-computation can achievesignificant processing cost savings without severely compromisingrouting performance.