Abdullah AlWehaibi

Packet loss probability for DiffServ over IP and MPLS reliable homogeneous multicast networks

Multicasting has become increasingly important with the emergence of Internet-based applications such as IP telephony, audio/video conferencing, distributed databases and software upgrading. IP multicasting is an efficient way to distribute information from a single source to multiple destinations at different locations. In practice IP is considered as a layer 3 protocol. Multiprotocol Label Switching (MPLS) replaces the IP forwarding by a simple label lookup. MPLS combines the flexibility of layer 3 routing and layer 2 switching.In order to provide QoS in group communications for real time applications such as video conferencing, reliable multicasting is used. Miscellaneous efforts have been undertaken to provide reliability on top of IP multicast. Two error control strategies have been popular in practice. These are the FEC (Forward Error Correction) strategy, which uses error correction alone, and the ARQ (Automatic Repeat Request) strategy, which uses error detection, combined with retransmission of data.In this paper, we present a new fair share policy (FSP) that utilizes Differentiated Services to solve the problems of QoS and congestion control when reliable ARQ multicast is used. The results should provide insight into the comparisons of the residual packet loss probability between IP multicast in MPLS networks using FSP and plain IP multicasting using the same policy when DiffServ are adopted and when reliable ARQ multicast is considered.

Performance behavior evaluation of Internet congestion control policing mechanisms

Performance behavior is an important issue in the design and implementation of an efficient Internet congestion control policing mechanism. The effectiveness of such a mechanism can be measured by packet loss probability, bandwidth allocation, packet delay, throughput or other quality of service measures. In this paper, we carry out a comprehensive study to investigate the performance behavior of four selected policing mechanisms for the Internet namely: token bucket (TB), jumping window (JW), triggered jumping window (TJW) and exponentially weighted moving average (EWMA). Three types of bursty sources modeled as On/Off, Poisson and batched Poisson processes are utilized. Three criteria are used to evaluate the performance behavior of the selected policing mechanisms. These are the average packet delay, the average packet loss probability and the average number of lost credits. Computer simulations were used to arrive at various conclusions regarding the dependence of performance on source traffic characteristics and policing mechanism parameters. Furthermore, a comparison of the performance behavior of the selected policing mechanisms was carried for different input traffic characteristics.

Computation of the residual packet loss probability in a binary multicast tree

In order to achieve a better quality of service (QoS), the use of reliable multicasting has become increasingly important especially with the emergence of Internet-based applications such IP telephony, audio/video conferencing. In this paper, the residual packet loss probability in a complete binary multicast tree which consists of N routers with a given probability of successful delivery to the next router is evaluated, when automatic repeat request (ARQ) multicast repairs is employed. In this paper, we also derive and compare two other mathematical expressions, which can be used to calculate the final packet loss probability in a binary tree where reliable ARQ multicasting is used. These expressions can be used in the case of IP or MPLS multicasting. The first expression, which is called the average packet loss probability for ARQ deals only with the number of routers that should have correct transmissions (no loss and no errors) during the repair trial of one previous loss depending on the location of this previous loss. The second one, which is called the accurate expression, takes into account the number of trials, the number of errors and the position of each error (at which level the error occurred).

Packet loss probability for DiffServ over heterogeneous MPLS multicast networks: a simulation study

In this paper, we present a fair share policy (FSP) that utilizes Differentiated Services to solve the problems of QoS and congestion control when reliable FEC/ARQ multicast is adopted. Simulation programs are used to evaluate the fair share policy. The results should provide insights into the comparisons between homogeneous IP multicast networks, homogeneous MPLS multicast networks and heterogeneous MPLS multicast networks using the same FSP when DiffServ is adopted and when reliable FEC/ARQ multicast is considered. This comparison is based on the residual packet loss probability.

Average multicast delay for FEC/DIFFSERV over IP and MPLS homogeneous networks

In order to provide QoS in group communications for real time applications such as video conferencing, reliable multicasting is used. Miscellaneous efforts have been undertaken to provide reliability on top of IP multicast. In this paper, we present a new fair share policy (FSP) that utilizes Differentiated Services to solve the problems of QoS and congestion control when reliable FEC multicast is adopted. The results should provide insight into the comparisons between IP multicast in MPLS networks using FSP and plain IP multicasting using the same policy when DiffServ are adopted and when reliable FEC multicast is considered.