Fast Recovery

In this paper we consider the problem of reconfiguring processor arrays subject to computational loads that alternate between two modes. A strict mode is characterized by a heavy computational load and severe constraints on response time while a relaxed mode is characterized by a relatively light computational load and relaxed constraints on response time. In the strict mode, reconfiguration is performed by a distributed local algorithm in order to achieve fast recovery from faults. In the relaxed mode, a global reconfiguration algorithm is used to restore the system to a state that maximizes the probability that future faults occurring in subsequent strict modes will be repairable. Several new results are given for this problem. Efficient reconfiguration algorithms are described for a number of general classes of architectures. These general algorithms obviate the need for architecture-specific algorithms for architectures in these classes. We show that it is unlikely that similar algorithms can be obtained for related classes of architectures since the reconflguration problem for these classes is NP-complete. Finally, a general approximation algorithm is described that can be used for any architecture. Experimental results are given, suggesting that our algorithms are very effective.

The phenomenon of traffic burstiness is a big challenge in the switching equipments of the core of computer network, because it leads to more delay and jitter to be tolerated by users. Therefore, we should use all of the available facilities to improve the traffic behavior. In this paper, we present an idea based on delay management and QoS considerations to decrease the variances in short-time scales. Using our approach the impact of short-term burstiness will be decreased. Also, this idea can be implemented in MPLS ingress nodes or by delay management of paths in MPLS traffic engineering when we have congestion in the network.

This paper studies the performance of MPLS networks with TE signal protocols in relation with voice codecs. Simulation were performed and compared for a multisite network with PCM and GSM based VoIP. Simulation results show that the MPLS network with CR-LDP TE signal protocol outperforms the MPLS network with RSVP TE signal protocol in terms of both the total amount of received voice packets and the number of maintained calls for both voice codecs.

In this paper, we compare the throughputs of two different TCP NewReno variants, namely Slow-but-Steady and Impatient. We develop analytic throughput models of these variants as a function of round-trip time, loss event rate, and the burstiness of packet drops within a loss event. Our models build upon prior work on TCP Reno throughput modeling, but extend this work to provide an analytical characterization of the NewReno fast recovery algorithms. We validated our models using the ns-2 simulator. Our models accurately predict the steady-state NewReno throughput for a wide range of loss rates. Based on these models, we analytically determine the preferred operating regions for each TCP variant. Our results show that the Slow-but-Steady variant is comparable to or superior to the Impatient variant in all but the most extreme scenarios for network packet loss.

The hierarchical token bucket (HTB) algorithm allows to specify per-flow bitrate guarantees and enables excess bandwidth sharing between flows of the same class. Additionally, it provides capabilities to prioritize the traffic of specific flows, potentially considering their delay demands. HTB hence constitutes a powerful mechanism to enforce QoS requirements hierarchically and on a fine granular per-flow level, making it an appropriate choice in numerous use-cases. In this paper, we present HTBQueue, our implementation of a compound module for HTB support in the discrete event simulator OMNeT++. We validate HTBQueue’s functionality in terms of rate conformance and fair bandwidth sharing behavior between competing flows. We furthermore demonstrate its support for flow prioritization.

Voice communication is an emerging technology and has great importance in our routine life. Perceptual, Voice over Internet Protocol quality is an important issue for VoIP Apps services because VoIP Apps require real-time support. Many network factors (packet loss, packet delay, and jitter) affect to VoIP quality, to achieve this objective we used an approach based on Fuzzy Logic. We configure Resource Reservation Protocol application to control Token Bucket Algorithm and the simulation experiments are carried out with Opnet. In addition, compare Token Bucket with and without Quality of Service for measure network factors. In this paper, building Fuzzy Token Bucket System consists of three variables (Bandwidth Rate, Buffer Size, and New Token) in order to improve Token Bucket Shaper output variable (New Token) by Fuzzy Stability model for Voice over IP quality maintaining.

IP/MPLS communication networks are experiencing an increase in real-time multimedia applications like voice over IP (VOIP) and video. Real-time multimedia applications have stringent quality of service (QoS) requirements with regard to delay, loss, bandwidth and availability. These applications must therefore be provided with prompt recovery from network failure. MPLS fast reroute provides recovery of about 50ms which makes it attractive for the recovery of voice traffic. Voice traffic must be recovered within 50ms so that call quality is not compromised. Bandwidth reservation and bandwidth allocation are needed to guarantee the protection of voice traffic during network failure. Since voice has a time constraint of 50ms within which traffic must be recovered, real-time bandwidth allocation is required. A bandwidth allocation scheme that prioritises voice traffic will ensure that voice is guaranteed bandwidth during network failure. A mechanism is also required to provide bandwidth to voice traffic when the reserved bandwidth is insufficient to accommodate voice traffic. This mechanism must be able to utilise the working bandwidth or bandwidth reserved for lower priority applications and allocate it to the voice traffic when a network failure occurs. This research therefore proposes a real-time bandwidth encapsulation mechanism to guarantee Quality of Protection (QoP) to voice traffic during single link and node failures. The mechanism uses label switched path (LSP) preemption and the Russian dolls bandwidth constraint model to guarantee bandwidth to voice. LSP preemption is used to free up bandwidth for protection and the Russian dolls model is used for bandwidth allocation. The metrics used to evaluate the performance of the proposed mechanism are LSP reroute time and packet loss. Simulations conducted in Optimised Network Engineering Tool (OPNET) modeler showed that this solution achieved faster LSP reroute time and minimal packet loss to voice.

Multiprotocol Label Switching (MPLS) is the latest technology in the evolution of routing and forwarding mechanisms for the core of the Internet. It ensures the reliable delivery of the Internet services with high transmission speed and lower delays. Traffic Engineering (TE) in MPLS is used for effectively managing the networks for efficient utilization of network resources. MPLS technology makes it more suitable for implementing real-time applications such as Voice and video due to lower network delay, efficient forwarding mechanism, scalability and predictable performance of the services. In this paper the performance of Voice over Internet Protocol (VoIP) application is compared in MPLS network and conventional Internet Protocol (IP) network. The comparison is based on the performance metrics such as Voice jitter, Voice packet end-to-end delay, voice delay variation, voice packet send and received. The simulation results are analyzed and it shows that MPLS based solution provides...

Efficient operation of multipath routing is considered as one of the important aspects in MPLS traffic engineering. Multipath routing has an advantage over the traffic routing based on single shortest paths because single path routing may lead to unbalanced traffic situations and degraded QoS (Quality of Service). This paper proposes a novel model for the performance evaluation of multipath routing in MPLS networks. The paper applies the MM K k=1 CPP k /GE/c/L G-queue to model the operation of ingress nodes and derives equations to obtain their performance measures.

Efficient operation of multipath routing is considered as one of the important aspects in MPLS traffic engineering. Multipath routing has an advantage over the traffic routing based on single shortest path because single path routing may lead to unbalanced traffic situations and degraded QoS (Quality of Service). The adequate strategy for the order of repairing failed links can greatly influence performance characteristics of an MPLS network. In this paper we propose a new repair strategy for MPLS multipath routing. We compare our repair strategy with several existing repair strategies, by way of comparing the associated performance measures. Numerical results show that our scheme performs better than many other repair schemes.

Efficient operation of multipath routing is considered as one of the important aspects in MPLS traffic engineering. Multipath routing has an advantage over the traffic routing based on single shortest paths because single path routing may lead to unbalanced traffic situations and degraded QoS (Quality of Service). This paper proposes a novel model for the performance evaluation of multipath routing in MPLS networks. The paper applies the MM K k=1 CPP k /GE/c/L G-queue to model the operation of ingress nodes and derives equations to obtain their performance measures.

This paper presents a new discrete event-based network simulator named TARVOS-Computer Networks Simulator, being designed as part of the first Author's Masters research and will provide support to simulating MPLS architecture, several RSVP-TE protocol functionalities and fast recovery in case of link failure. The tool is used in a case study, where the impact of a link failure on a VoIP application, within an MPLS domain network, is analyzed. The paper displays a preliminary research of six already available simulators and reasons why they were not adopted as tools for the Masters research. Then, it follows to describe the basics of TARVOS implementation and exhibits the case study simulated by this new tool.

In recent years, Voice over IP (VoIP) has impacted global telecommunications and networking tremendously. Traffic engineering and Quality of Service (QoS) guarantees for VoIP services pose a challenge for network researchers and designers. The repeated use of Internet Protocol shortest path towards the same destination may lead to unbalanced traffic situations and degraded network performance. Therefore, load balancing and link utilization become the critical functions in Internet Protocol routing for providing Quality of Service assurance for VoIP application. The aim of this work is to employ Multiprotocol Label Switching Network as a traffic engineering tool to enhance the QoS for VoIP applications. To achieve this, an effective Multiprotocol Label Switching Network load balancing architecture is developed that classifies the Internet traffic flows, routes the flows into multiple paths. Flow arrival rate, packet loss rate and delay are measured and taken as the input parameters and compared with the threshold values to identify the VoIP flow. Network load status is calculated by estimating the average buffer occupancy value and multipath routing is triggered when the network load is high to enhance the QoS. The investigated performance measures like throughput, delay and packet loss are reported to show the efficiency of the proposed technique for effective VoIP flows.

This paper presents a discrete-event systems approach to the modeling of a packet-switched communication link shared by multiple users for the purpose of end-to-end congestion control. It introduces a discrete-event system model for the interaction between a shared communication link and the users, which captures the behavior that the link users receive acknowledgments for successfully transmitted packets with a delay that is proportional to the level of congestion in the link. An end-to-end congestion control scheme for this system model is presented that uses the concept of an observer from supervisory control theory. The link model and controllers are implemented in both a Java programming language simulation and the NS2 network simulation software for analysis.

The ease of administration and its reduced costs make the MPLS (Multiprotocol Label Switching) technology attractive to those who want to deploy a performant, reliable and scalable private network. Connecting enterprises to remote locations, customers, and vendors via the MPLS is a very flexible solution that is being taken into account by many communication providers. VPN (Virtual Private Networks) is one of the most demanding services for the MPLS technology. The service providing of MPLS Differentiated Services (MPLS DiffServ) allows the customization of each client's traffic. This paper briefly outlines the theoretical aspects of the MPLS technology. The theoretical analysis is summed up in a case study called MPLS_DiffServ, developed with the OPNET Modeler simulator. This simulator allowed us to create a network model and configure the network equipment to provide MPLS DiffServ services. The results that are obtained from the simulation run are suggestive for networking education and a good starting point for the networking research area, as services for new business environment.

This paper studies the performance of MPLS networks with TE signal protocols in relation with voice codecs. Simulation were performed and compared for a multisite network with PCM and GSM based VoIP. Simulation results show that the MPLS network with CR-LDP TE signal protocol outperforms the MPLS network with RSVP TE signal protocol in terms of both the total amount of received voice packets and the number of maintained calls for both voice codecs.

Multiprotocol Label Switching (MPLS) is an evolving standards-approved technology to increase the speed of the network traffic flow and making the network easier to manage. In MPLS a specific path is set up for a given sequence of packets, identified by a label put in each packet, thus saving the time needed for a router to look up the address to the next node to forward the packet. MPLS is called multiprotocol because it works with the Internet Protocol (IP),Asynchronous Transport Mode(ATM), frame relay network protocols. It ensures the efficient and reliable delivery of the Internet services with high transmission speed and lower delays. The main feature of MPLS is the concept of Traffic Engineering (TE), which is used for effectively managing the networks for efficient utilization of network resources. This technology is more suitable for implementing real-time applications such as voice and video due to lower network delay, efficient forwarding mechanism, scalability and predictable performance of the services provided by MPLS. In this paper the performance of voice over internet protocol in MPLS and non MPLS network is compared. Here we consider non-MPLS network as conventional IP network. In order to simulate the network we use Optimized Network Engineering Tool 14.5 and comparison is made on various parameters such as Ethernet delay, voice jitter, voice packet end to end delay.

Sweden OPNET simulation of voice over MPLS with considering Traffic Engineering ABSTRACT Multiprotocol Label Switching (MPLS) is an emerging technology which ensures the reliable delivery of the Internet services with high transmission speed and lower delays. The key feature of MPLS is its Traffic Engineering (TE) which is used for effectively managing the networks for efficient utilization of network resources. Due to lower network delay, efficient forwarding mechanism, scalability and predictable performance of the services provided by MPLS technology makes it more suitable for implementing real-time applications such as Voice and video. In this thesis performance of Voice over Internet Protocol (VoIP) application is compared in MPLS network and conventional Internet Protocol (IP) network. OPNET modeler 14.5 is used to simulate the both networks and the comparison is made based on the metrics such as Voice jitter, Voice packet end-to-end delay, voice delay variation, voice packet send and received. The simulation results are analyzed and it shows that MPLS based solution provides better performance in implementing the VoIP application.

The complexity (number of entities, interactions between entities, and resulting emergent dynamic behavior) of large communication environments which contain hundreds of nodes and links make simulation an important tool for the study of such systems. Given the di culties associated with complete analytical treatment of complex dynamical systems, it is often the only practical tool that is available. This paper presents an example of a exible, modular, object-oriented toolbox designed to support modeling and experimental analysis of a large family of heuristic knowledge representation and decision functions for adaptive self-managing communication networks with particular emphasis on routing strategies. It discusses in detail, the design, implementation, and validation of the toolbox using the discrete event simulation paradigm. It addresses several major practical design challenges presented by potentially large computational time and memory requirements of simulation of large communication networks and describes novel solutions adapted from the simulation and performance modeling and analysis literature to meet such challenges. It also examines some issues that arise in the validation of the resulting object-oriented simulation and modeling toolbox.

The complexity (number of entities, interactions between entities, and resulting emergent dynamic behavior) of large communication environments which contain hundreds of nodes and links make simulation an important tool for the study of such systems. Given the di culties associated with complete analytical treatment of complex dynamical systems, it is often the only practical tool that is available. This paper presents an example of a exible, modular, object-oriented toolbox designed to support modeling and experimental analysis ...

This paper addresses the problem of resource fragmentation (RF) in QoS controlled networks. Resources are said to be fragmented when they are available in non-contiguous blocks and hence cannot be utilized by incoming calls with high resource demands. This paper shows the effect of resource fragmentation on QoS controlled networks and presents the Dynamic Resource Redistribution (DRR) algorithm to counteract RF. The DRR algorithm reduces the effects of RF by attempting to redistribute resources in different paths to make resources to incoming calls. A variety of simulation experiments were conducted to study the performance of the DRR algorithm on different network topologies with varying traffic characteristics. The DRR algorithm, when used, increased the number of calls accommodated in the network as well as the overall resource allocation in the network.

This documents provides a quick introduction to OMNeT++, with a focus on queueing applications. The aim is to get the interested user up and running with a simple M/M/1 fifo queueing demo. After this the user can start exploiting more of OMNeT++'s functionality based on the extensive user manual and the sample directories.

This paper focuses on the protection of virtual circuits (Label Switched Paths, LSPs) in a (G)MPLS (Generalised Multi-Protocol Label Switching) network. The proposed algorithm is designed to protect traffic with strong delay requirements such as EF (Expedited Forwarding) ordered aggregates in a DiffServ domain. Indeed, for this type of application, we need fast restoration in case of failure. The duplication of all the packets in a 1+1 end-toend restoration scheme consumes a large amount of bandwidth. Furthermore, end-to-end recovery with bandwidth sharing schemes are usually considered to be far too slow. Local fast-rerouting is a solution which can compete with restoration times and bandwidth consumption offered by SONET self-healing rings. Our scheme includes a sophisticated resource aggregation mechanism based on the concepts of "backup-backup aggregation" and "backup-primary aggregation". The path selection algorithm is also designed to efficiently reduce the resource usage. Moreover, when considering LSPs at different preemption levels, our algorithm is able to correctly calculate the amount of bandwidth that can be preempted despite the sharing of resource. We show that our approach, though local, can compete with the state-of-the-art end-to-end recovery schemes in terms of resource consumption. The major contribution of our scheme, the "backup-primary aggregation", was then also used in the context of end-to-end recovery and improved its performance substantially. To be able to save a maximum amount of bandwidth in a decentralised implementation, the nodes that compute backup LSPs need to obtain a certain amount of link-state information. We propose a solution where the nodes learn almost all the information they need with RSVP messages. This drastically reduces the information that needs to be flooded in the whole network and is the first scalable decentralised solution capable of sharing a large amount of bandwidth.

This paper presents a new discrete event-based network simulator named TARVOS -Computer Networks Simulator, being designed as part of the first Author's Masters research and will provide support to simulating MPLS architecture, several RSVP-TE protocol functionalities and fast recovery in case of link failure. The tool is used in a case study, where the impact of a link failure on a VoIP application, within an MPLS domain network, is analyzed.

Web-based simulation is a powerful skill for e-learning. Simulation gives concrete answers to the students questions. Having a Virtual Simulation L aboratory is possible when using web-based simulation. This paper presents JDESK, a web-based discrete event simulator written in Java. It is fast, versatile and powerful enough to manage whatever system. This simulator allows to describe and to change a model easily. JDESK may be used as a prototyper. Although it allows simulating any system behavior, it includes libraries containing predefined behaviors to allow students define simple models easily. This simulator allows changing dynamically system topology. Any continuous or discrete behavior may be included in the system. JDESK can manage models with high complexity and size. The programmer only focuses on the model description, not in events or programming bugs. The effort to describe a behavior can be reused (in the same model or others). JDESK includes an assistant to guide the user in the model definition process.

This paper presents a new discrete event-based network simulator named TARVOS -Computer Networks Simulator, being designed as part of the first Author's Masters research and will provide support to simulating MPLS architecture, several RSVP-TE protocol functionalities and fast recovery in case of link failure. The tool is used in a case study, where the impact of a link failure on a VoIP application, within an MPLS domain network, is analyzed. The paper displays a preliminary research of six already available simulators and reasons why they were not adopted as tools for the Masters research. Then, it follows to describe the basics of TARVOS implementation and exhibits the case study simulated by this new tool.

Abstract—Corporates and multisite organizations are now applying VOIP usage all over their branches, this made offices with no boundaries and reduced a huge amount of cost for their infrastructure; facilitated exchanging for voice, video and Data. Growing demand for such usage has pushed the wheel for improving and applying more techniques to make this service more reliable, efficient and scalable. In this paper a simulation were performed and compared for a multisite office network for G.

This paper presents capacity planning rules for the control plane of all-optical networks featuring GMPLS and RSVP-TE as a connection setup protocol. As per RSVP standard, a refresh message mechanism is incorporated to RSVP such that the state is periodically refreshed on a link per link basis. We provide analytical expressions for the bandwidth and buffer sizes to be provided such that no flows are torn down due to lack of refresh messages. Our findings show that small buffers (several KBytes) suffice to sustain the signaling load for as much as 400 RSVP flows per link, with the simplest RSVP refresh mechanism (neither using link bundling nor acknowledgments). On the other hand, we also find the packet drop probability per link for a given network topology for the case that the flow survival probability is larger than a given threshold. We provide numerical examples based on the COST 239 european network topology and real RSVP traffic traces from early-commercial switching equipment.

This paper presents a new discrete event-based network simulator named TARVOS – Computer Networks Simulator, being designed as part of the first Author's Masters research and will provide support to simulating MPLS architecture, several RSVP-TE protocol functionalities and fast recovery in case of link failure. The tool is used in a case study, where the impact of a link failure on a VoIP application, within an MPLS domain network, is analyzed. The paper displays a preliminary research of six already available simulators and reasons why they were not adopted as tools for the Masters research. Then, it follows to describe the basics of TARVOS implementation and exhibits the case study simulated by this new tool.