Traffic matrices describe the volume of traffic between a set of sources and destinations within a network. These matrices are used in a variety of tasks in network planning and traffic engineering, such as the design of network topologies. Traffic matrices naturally possess complex spatiotemporal characteristics, but their proprietary nature means that little data about them is available publicly, and this situation is unlikely to change. Our goal is to develop techniques to synthesize traffic matrices for researchers who wish to test new network applications or protocols. The paucity of available data, and the desire to build a general framework for synthesis that could work in various settings requires a new look at this problem. We show how the principle of maximum entropy can be used to generate a wide variety of traffic matrices constrained by the needs of a particular task, and the available information, but otherwise avoiding hidden assumptions about the data. We demonstrate how the framework encompasses existing models and measurements, and we apply it in a simple case study to illustrate the value.
Traffic matrices are used in many network engineering tasks, for instance optimal network design. Unfortunately, measurements of these matrices are error-prone, a problem that is exacerbated when they are extrapolated to provide the predictions used in planning. Practical network design and management should consider sensitivity to such errors, but although robust optimisation techniques exist, it seems they are rarely used, at least in part because of the difficulty in generating an ensemble of admissible traffic matrices with a controllable error level. We address this problem in our paper by presenting a fast and flexible technique of generating synthetic traffic matrices. We demonstrate the utility of the method by presenting a methodology for robust network design based on adaptation of the mean-risk analysis concept from finance.
Most present day switching systems, in Internet routers and data-center switches, employ a single input-queued crossbar to interconnect input ports with output ports. Such switches need to compute a matching, between input and output ports, for each switching cycle (time slot). The main challenge in designing such matching algorithms is to deal with the unfortunate tradeoff between the quality of the computed matching and the computational complexity of the algorithm. In this paper, we propose a general approach that can significantly boost the performance of both SERENA and iSLIP, yet incurs only O(1) additional computational complexity at each input/output port. Our approach is a novel proposing strategy, called Queue-Proportional Sampling (QPS) , that generates an excellent starter matching . We show, through rigorous simulations, that when starting with this starter matching, iSLIP and SERENA can output much better final matching decisions, as measured by the resulting throughput and delay performance, than they otherwise can.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.