بایگانی برچسب برای: Queueing model

Modeling-and-Evaluation-of-Multisource.[taliem.ir]

Modeling and Evaluation of Multisource Streaming Strategies in P2P VoD Systems

In recent years, multimedia content distribution has largely been moved to the Internet, inducing broadcasters, operators and service providers to upgrade with large expenses their infrastructures. In this context, streaming solutions that rely on user devices such as set-top boxes (STBs) to offload dedicated streaming servers are particularly appropriate. In these systems, contents are usually replicated and scattered over the network established by STBs placed at users’ home, and the video-on-demand (VoD) service is provisioned through streaming sessions established among neighboring STBs following a Peer-to- Peer fashion. Up to now the majority of research works have focused on the design and optimization of content replicas mechanisms to minimize server costs. The optimization of replicas mechanisms has been typically performed either considering very crude system performance indicators or analyzing asymptotic behavior. In this work, instead, we propose an analytical model that complements previous works providing fairly accurate predictions of system performance (i.e., blocking probability). Our model turns out to be a highly scalable, flexible, and extensible tool that may be helpful both for designers and developers to efficiently predict the effect of system design choices in large scale STB-VoD systems1.
A DFO technique to calibrate queueing models[taliem.ir]

A DFO technique to calibrate queueing models

A crucial step in the modeling of a system is to determine the values of the parameters to use in the model. In this paper we assume that we have a set of measurements collected from an operational system, and that an appropriate model of the system (e.g., based on queueing theory) has been developed Not infrequently proper values for certain parameters of this model may be difficult to estimate from available data (because the corresponding parameters have unclear physical meaning or because they cannot be directly obtained from available measurements, etc.). Hence, we need a technique to determine the missing parameter values, i.e., to calibrate the model. As an alternative to unscalable “brute force” technique, we propose to view model calibration as a nonlinear optimization problem with constraints. The resulting method is conceptually simple and easy to implement. Our contribution is twofold. First, we propose improved definitions of the “objective function” to quantify the “distance” between performance indices produced by the model and the values obtained from measurements. Second, we develop a customized derivative-free optimization (DFO) technique hose original feature is the ability to allow temporary constraint violations. This technique allows us to solve this optimization problem accurately, thereby providing the “right” parameter values. We illustrate our method using two simple real-life case studies.