QOS control in an ATM network considering virtual path concept

Deterministic algorithm for VP assignment in ATM networks

The virtual path (VP) concept in ATM networks simplifies network structure, traffic control and resource management. For VP formulation, a VP can carry traffic of the same type (the separate scheme) or of different types (the unified scheme). For VP adjustment, a certain amount of bandwidth can be dynamically assigned (reserved) to VPs, where the amount (the bandwidth incremental/decremental size) is a predetermined system parameter. In this paper, we study least loaded path-based dynamic routing schemes with various residual bandwidth definitions under different bandwidth allocation (VP formulation and adjustment) schemes. In particular, we evaluate the call blocking probability and VP set-up processing load with varying (bandwidth) incremental sizes. Also, we investigate numerically how the use of VP trades the blocking probability with the processing load. It is found that the unified scheme could outperform the separate scheme in certain incremental sizes. Moreover, we propose two ways to reduce the processing load without increasing the blocking probability. Using these methods, the separate scheme always outperforms the unified scheme.

A challenge in B-ISDN is effective provision of QOS guarantees. This paper proposes a scheme for guaranteeing QOS in the B-ISDN/ATM framework while utilizing resources. It addresses the key open issue involved in this challenge and poses a solution different from the current approaches. The fundamental idea of the scheme is to extend the virtual path concept to include a reservation of time slots at switching nodes by each virtual path. Cells for a session are divided into two classes at the access node: those that are guaranteed transmission within the reserved time slots and those that are discardable. Through this global and active control along virtual paths. Data transfer in a network resembles writing to input ports of a personal computer with deterministic delay and zero loss. By implementing an architecture as layers, resources are utilized without sacrificing processing time on fly. >

The ITU-T defined next generation network (NGN) architecture designates the resource admission control function to perform the application-driven Quality of Service (QoS) control across access and core networks. The Generalized Multiprotocol Label Switching (GMPLS) was introduced by the Internet Engineering Task Force (IETF) to cope with new traffic engineering challenges in fast optical networks and provide them with reliable end-to-end QoS mechanisms. However, an actual Connection Admission Control (CAC) implementation inside the resource reservation protocol - traffic engineering extension (RSVP-TE) in GMPLS networks does not provide the ability of effective decision making, since the applied threshold CAC lacks of the capability to consider QoS policies on GMPLS network nodes. This prevents an effective end-to-end QoS control in a fully dynamic, application driven Label Switched Path (LSP) setup scenario. This work presents a specific implementation of fuzzy-CAC operating over an RSVP-TE agent in GMPLS network domain. This fuzzy-CAC implementation is applied to a testbed where a client application requests a real-time data transfer through a GMPLS network, which results in dynamic LSP setup and exclusion. The admission control is performed upon service request based on QoS class requirements and network resource availability. The differentiated traffic treatment on per-flow basis is realized through employment of IF-THEN rule based expert knowledge. Effective traffic differentiation is achieved in a multi-service network scenario and thus it validates fuzzy-CAC as candidate for RSVP-TE protocol enhancement for application driven QoS provisioning in GMPLS networks.

Real-time communication with performance guarantees is expected to become an important feature of future computer networks. Given an ATM network topology, its virtual path (VP) layout, and its traffic demands, we consider in this paper the problem of selecting for each virtual circuit (VC) with timing requirements a route (i.e., a sequence of VPs) along which sufficient resources are available to meet the user-specified end-to-end delay requirements. Our objective is to (i) provide the timing guarantee for the VC, while not jeopardizing the QoS guarantees to other existing VCs. We adopt the real-time channel model to characterize the traffic characteristics and the timing requirements of a VC. We then propose a distributed VC routing scheme based on the distributed Bellman-Ford algorithm to identify an "appropriate" route through the network. By "appropriate", we mean that the route traverses a minimum number of VPs among all possible routes that have sufficient resources over to fulfill the end-to-end timing requirement of the VC. To ensure that there are sufficient bandwidths over all the VPs along the selected route, we incorporate in our proposed scheme a priority assignment method to calculate the minimum worst-case traversal time which messages of a VC will experience on a VP along which the VC is routed. We also comment on the performance of, and the message overhead incurred in the proposed scheme.

The advantages of applying the virtual path (VP) concept to asynchronous transfer mode (ATM) networks are summarized. The implementation techniques of VPs with ATM cross-connect systems and some VP applications are presented. It is demonstrated that network capability can be effectively increased by applying VP techniques. >

Performing QoS (Quality of Service) control in large computing systems requires an on line metric that is representative of the real state of the system. The Tardiness Quantile Metric (TQM) introduced earlier allows control of QoS by measuring efficiently how close to the specified QoS the system is, assuming specific distributions. In this paper we generalize this idea and propose the Generalized Tardiness Quantile Metric (GTQM). By using an online convergent sequential process, defined from a Markov chain, we derive quantile estimations that do not depend on the shape of the workload probability distribution. We then use GTQM to keep QoS controlled in a fine grain manner, saving energy in soft real-time web clusters. To evaluate the new metric, we show practical results in a real web cluster running Linux, Apache, and MySQL, with our QoS control and for both a deterministic workload and an e-commerce workload. The results show that the GTQM method has excellent workload prediction capabilities, which immediately translates in more accurate QoS control, allowing for slower speeds and larger energy savings than the state-of-the-art in soft real-time web cluster systems.

Scalable Quality of Service (QoS) control is of paramount importance to effectively enable a seamless convergence of the rapidly evolving Peer-to-Peer (P2P) overlay communications over the Internet since the latter only supports best-effort service paradigm. For example, the European Union (EU) funded ROMEO project is focusing on a joint use of DVB-T2 and P2P overlay networks for live multimedia content sharing and collaboration among multiple users. This raises a strong need that the media packets transmitted through the P2P overlay delivery system must arrive earlier enough at the end users to assure a proper synchronization of the multiple views that may be received via the hybrid network. For this purpose, the P2P network must assure a certain QoS guarantee in terms of bandwidth, delay, jitter, and loss. More importantly, the control must be scalable to prevent excessive signalling and the related processing overhead, usually suffered in the traditional per-flow QoS control approaches. In this view, recent research effort claimed that the Internet resources can be efficiently over-provisioned (booking more resources in-advance) in such a way to allow differentiation of QoS control with reduced signalling overhead and increased resource utilization. This approach, however, needs further investigations for proper integration into innovative networking architectural designs to achieve performance. In addition to that, and in order to provide an end-to-end QoS, a mechanism to enforce prioritization policies within the customer’s access network is also needed. Hence, this chapter proposes a cross-layer control architecture that takes advantage of the Internet resources over-provisioning and the QoS policy enforcement within access networks to facilitate rapid development of P2P applications. The design aims to alleviate the requirements of buffers and the need for adaptation on end users’ devices, thus allowing for cost-effective and rapid development of attractive services in similar hybrid content delivery networks.

In a distributed multimedia integration environment, there is a need for transmission control methods based on real-time communication protocols, that provide a relative quality of service (QOS) to allow a variety of receiving capacities for group members and their requests and that provide a dynamic QOS to cope with temporary CPU overload and network congestion. Therefore, this study proposes a cooperative control method for end-to-end QOS control at a transport layer and for point-to-point control at a network layer. The former is mainly a set of flow-monitoring and flow-adjusting functions, and the latter a set of flow-control functions. The evaluation results show that dynamic QOS control with the two-level monitoring is preferable to conventional one in terms of the stabilization of average packet-loss rate and jitters.

Quality of Service (QoS) and Inter-domain traffic control are the current research topics in the fields of computer communication and IP Telephony in Local Area Network (LAN), Wide Area Network (WAN) and Wi-Fi network. By appropriate design and implementation of QoS and traffic control in a corporate network a significant amount of bandwidth and thereby money can be saved. Performance and reliability are also enhanced in this process. In the current QoS framework emphasis has only been put on layer 2 and layer 3, using traffic classification, scheduling and provisioning. Also, there is no effective means of measuring an agency's QoS implementation level. In this paper we have proposed a new QoS framework which includes QoS at all 7 layers of OSI reference model. We have also introduced a new technique, Quality of Service Quotient (QoSQ) for measuring an agency's QoS posture. We have implemented QoS in a NSW government agency following this new QoS framework and have provided ‘bandwidth’ usage data for both pre and post QoS implementation. The data shows considerable subscribed bandwidth savings (55%) and increased average bandwidth usage (30%).

An improved scheme for self-healing in ATM networks

A ring architecture that uses the asynchronous transfer mode (ATM) virtual path (VP) concept to reduce the SONET ring cost in terms of bandwidth management is discussed. The add-drop multiplexer (ADM) proposed for VP ring architectures can evolve from existing SONET ADMs by replacing the STS-3 termination cards by the ATM STS-3c line cards. Existing standard self-healing schemes and protocol SONET rings can be applied to proposed ATM/SONET VP rings. A case study based on a BCC (Bellcore client company) ring network and the sensitivity analysis suggests that the proposed ATM VP ring architecture may be a cost-effective option for implementing the distributed ring grooming system at the DS 1 level. The proposed VP self-healing rings are not only used to carry existing DS 1 and DS 3 service, but can also be used to more cost-effectively consolidate switches for public-switched services and for SMDS, frame relay, and FDDI in metropolitan areas.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

In the paper we describe a framework for resource management and traffic control which is based on the virtual network concept. In this context the virtual networks are used as a toot for customization of networks management functions and for virtual separation of network resources. We identify three main categories of virtual network applications (service, user and management oriented virtual networks). We discuss generic and application oriented problems, which have to be solved to take full advantage of the proposed framework. We also study the relation between the virtual network concept and the virtual path concept in the context of bandwidth management. The study indicates that using virtual paths for bandwidth management involves some inherent contradictions.

A crucial component of the interaction between ATM's and the Internet's Quality of Service (QoS) architectures is the efficient mapping of RSVP (Resource reSerVation Protocol) as the Internet's signalling protocol onto the according ATM mechanisms. In particular, this article focuses on one of the most contrary characteristics of RSVP and ATM signalling. This is the support for heterogeneous reservations by RSVP over the ATM subnetwork, taking into account that ATM only allows for a homogeneous QoS within a single Virtual Circuit (VC). We present previous approaches to the solution of this problem and argue for more sophisticated and efficient approaches to manage ATM VCs taking into consideration ATM tariffs and resource consumption.

In this paper, we examine the effect of QoS (Quality of Service) control for cooperation between two remote robot systems with force feedback. Previously, we applied the adaptive Δ-causality control as QoS control and the stabilization control with filters to cooperative work of carrying an object together by using the systems and demonstrated the effectiveness of the two types of control. However, when changing the movement direction, somewhat large force was applied to the object. We may be able to solve the problem by performing the robot position control using force information, which we previously proposed as QoS control. This paper applies the control to the work, and then the effect of the control is investigated by experiment.