We suggest a method of optimizing resource allocation for real time protocol traffic in general, and VoIP in particular, within an H.323 environment. There are two options in the packet network to allocate resources: aggregate peak demand and statistical multiplexing. Statistical multiplexing, our choice for this case, allows the efficient use of the network resources but however exhibits greater packet delay variation and packet transfer delay. These delays are often the result of correlations or time dependency experienced by the system's queue due to the variations observed in different point processes that occur at a point of time. To address these issues, we suggest a queuing method based on the diffusion process approximated by OrsteinUlenbeck and the non-validated results of Ren and Kobayashi.

The complexity of big data structures and networks demands more research in terms of analysing and representing data for a better comprehension and usage. In this regard, there are several types of model to represent a structure. The aim of this article is to use a social network topology to analyse road network for the following States in the United States (US): California, Pennsylvania and Texas. Our approach mainly focuses on clustering of road network data in order to create "communities".

A method for modeling aggregated heavy Markov bursty Ethernet traffic from different sources is proposed in this paper, particularly that prevailing between gateway services and internet routing devices, with the aim of achieving rate accommodation. In other words, to accommodate different rates while filtering out delays in the queue, to achieve access network convergence. Although gateway functions can be used to achieve this by adapting service rates, as many gateways as services are required. Instead of considering the distributed gateway services method, statistical multiplexing is chosen for this study for cost efficiency in network resources. Unfortunately, statistical multiplexing exhibits greater packet variation (jitter) and transfer delay. These delays, basically resulting from positive correlations or time dependency in the queue system, are addressed through infinitesimal queue modeling, based on the diffusion process approximated by Ornstein-Uhlenbeck, which deals with infinitesimal changes in the Markov queue. The related analysis has resulted in an exponential queueing model for univariate and/or multivariate servers obtained through Markov Gaussian approximation. An experiment based on two different voice algorithms shows rate accommodation, and a fluid solution, which is dynamically outputted according to the transmission link availability during each transition time, without any significant delay. Hence, better transfer delay and rate control is obtained through the proposed two multiplexing levels within an Ethernet LAN

In this paper, a parametric prediction model is proposed for a queuing system driven by the Markov process. The aim of the model is to minimize the packet loss caused by time dependency characterized by the queue tail being too long, resulting in a time-out during the transfer of a large dataset. The proposed technique requires the key parameters influencing the queue content to be determined prior to its occupation regardless of the server capacity definition, using Bayesian inference. The subsequent time elapsing between the arrival and departure of a packet in the system is given, as well as the system load. This queue content planning is considered as the Markov birth-death chain; a type of discretization that characterizes almost all queuing systems, leading to an exponential queue, and captured herein using beta distribution. The inference results obtained using this exponential queue indicate that the queue with predictive parameters employing beta distribution, even when dealing with a loss system queue, involves less transition time and a greater load than a queue with coarse parameters; hence, preventing a long tail in the queue which is the cause of packet loss.

No abstract