Adaptive provisioning of differentiated services networks based on reinforcement learning

The issue of bandwidth provisioning for Per Hop Behavior (PHB) aggregates in Differentiated Services (DiffServ) networks has received a lot of attention from researchers. However, most proposed methods need to determine the amount of bandwidth to provision at the time of connection admission. This assumes that traffic in admitted flows always conforms to predefined specifications, which would need some form of traffic shaping or admission control before reaching the ingress of the domain. This paper proposes an adaptive provisioning mechanism based on reinforcement-learning principles, which determines at regular intervals the amount of bandwidth to provision to each PHB aggregate. The mechanism adjusts to maximize the amount of revenue earned from a usage-based pricing model. The novel use of a continuous-space, gradient-based learning algorithm, enables the mechanism to require neither accurate traffic specifications nor rigid admission control. Using ns-2 simulations, we demonstrate using Weighted Fair Queuing, how our mechanism can be implemented in a DiffServ network.     

Trustworthy Web services provisioning for differentiated customer services

With the number of e-Business applications dramatically increasing, a service level agreement (SLA) will play an important part in Web services. The SLA is a combination of several quality of services (QoS), such as security, performance, and availability, agreed between a customer and a service provider. Most existing research addresses only one of these QoS metrics. Furthermore, in the case of the response time defined as one of QoS metrics for performance, only the average time to process and complete a job is typically used. Moreover, customer requests often need to be distinguished, with different request characteristics and customer’s different service requirements. In this paper, we consider a set of computer resources used by a service broker to host enterprise applications for two classes of differentiated customer services subject to a service level agreement. We study three QoS metrics, namely, trustworthiness, a percentile response time, and availability. The percentile response time metric defines a value below which the end-to-end response time has to be for a given percent of time. We present an approach for resource optimization in such an environment that minimizes the total cost of computer resources while satisfying all these three QoS metrics in a trust-based resource provisioning problem which typically arises in Web services. We formulate the trust-based resource provisioning problem as an optimization problem under SLA constraints, and then solve it using an efficient numerical procedure.     

QoS provisioning for multimedia wireless networks

As we move towards IP-based multimedia wireless networks with voice, video and data convergence, quality of service (QoS) provisioning will become an increasingly challenging task. One implication is that greater emphasis on managing the call admission and overall network resources will be needed. This paper presents a conservative and adaptive quality of service (CAQoS) framework for provisioning the QoS for both real-time and non-real-time traffic in a multimedia wireless network. Unlike most conventional schemes, which gradually scale down the bandwidth of ongoing connections to accommodate new connection/hand-off requests, CAQoS introduces an early scaling-down of bandwidth for new connections based on a designated provisioning model. The performance of a CAQoS system is evaluated through simulations of a realistic wireless environment. Simulation results show that CAQoS meets our design goals and outperforms conventional schemes.     

Dynamic core provisioning for quantitative differentiated services

Efficient network provisioning mechanisms that support service differentiation are essential to the realization of the Differentiated Services (DiffServ) Internet. Building on our prior work on edge provisioning, we propose a set of efficient dynamic node and core provisioning algorithms for interior nodes and core networks, respectively. The node provisioning algorithm prevents transient violations of service level agreements (SLA) by predicting the onset of service level violations based on a multiclass virtual queue measurement technique, and by automatically adjusting the service weights of weighted fair queueing schedulers at core routers. Persistent service level violations are reported to the core provisioning algorithm, which dimensions traffic aggregates at the network ingress edge. The core provisioning algorithm is designed to address the difficult problem of provisioning DiffServ traffic aggregates (i.e., rate-control can only be exerted at the root of any traffic distribution tree) by taking into account fairness issues not only across different traffic aggregates but also within the same aggregate whose packets take different routes through a core IP network. We demonstrate through analysis and simulation that the proposed dynamic provisioning model is superior to static provisioning for DiffServ in providing quantitative delay bounds with differentiated loss across per-aggregate service classes under persistent congestion and device failure conditions when observed in core networks.     

On-line provisioning of network services

The authors review the telecommunications service-provisioning process. They describe a view of future service provisioning and discuss the issues and technical problems involved in online service provisioning. They explore how intelligent network elements in conjunction with distributed algorithms and distributed databases can be used to achieve this objective. The authors also present and describe a system architecture for the implementation of online service provisioning     

QoS provisioning for multiple non‐real‐time services in cellular wireless networks

Future mobile services are expected to include various non‐voice oriented services. One important category of non‐voice oriented mobile services is non‐real‐time services. When a mobile user establishes a connection to access non‐real‐time service, the mobile user usually cares about whether the total time to complete its data transfer is within its time tolerance. In addition, different mobile users may have different bandwidth requirements and different tolerances in the total completion time. It is essential for wireless systems to provide various mobile users with different total completion times. In this paper, two quality‐of‐service (QoS) metrics, called stretch ratio and eligibility percentage, are employed at a connection level to present the degree of the length of the total completion time. We devise a measurement based call admission control scheme that provides multiple QoSs for various mobile users which have different requirements of stretch ratios, eligibility percentages, and bandwidths. Extensive simulation results show that the measurement based call admission control scheme not only provides various satisfactory QoSs for mobile users but also produces high throughput. Copyright © 2010 John Wiley & Sons, Ltd.     

Implications of proxy caching for provisioning networks and servers

In this paper, we examine the potential benefits of Web proxy caches in improving the effective capacity of servers and networks. Since networks and servers are typically provisioned based on a high percentile of the load, we focus on the effects of proxy caching on the tail of the load distribution. We find that, unlike their substantial impact on the average load, proxies have a diminished impact on the tail of the load distribution. The exact reduction in the tail and the corresponding capacity savings depend on the nature of the workload and the percentile of the load distribution chosen for provisioning networks and servers-the higher the percentile, the smaller the savings. For workloads considered in this study, compared with over a 50% reduction in the average load, the savings in network and server capacity was only 20%-35% for the 99th percentile of the load distribution. We also find that while proxies can be somewhat useful in smoothing out some of the burstiness in Web workloads; the resulting workload continues, however, to exhibit substantial burstiness and a heavy-tailed nature. We identify one-time requests for large objects to be the limiting factor that diminishes the impact of proxies on the tail of load distribution. We conclude that, while proxies are immensely useful to users due to the reduction in the average response time, they are less effective in improving the capacities of networks and servers.     

FRR for latency reduction and QoS provisioning in OBS networks

We propose a forward resource reservation (FRR) scheme to reduce the data burst delay at edge nodes in optical burst switching (OBS) systems. We also explore algorithms to implement the various intrinsic features of the FRR scheme. Linear predictive filter (LPF)-based methods are investigated and demonstrated to be effective for dynamic burst-length prediction. An aggressive resource reservation algorithm is proposed to deliver a significant performance improvement with controllable bandwidth cost. By reserving resources in an aggressive manner, an FRR system can reduce both the signaling retransmission probability and the bandwidth wastage as compared with a system without the aggressive reservation. An FRR-based QoS strategy is also proposed to achieve burst delay differentiation for different classes of traffic. Theoretical analysis and simulation results verify the feasibility of the proposed algorithms and show that our FRR scheme yields a significant delay reduction for time-critical traffic without incurring a deleterious bandwidth overhead.     

Bandwidth provisioning in ADSL access networks

We consider an ADSL access network consisting of subscribers, digital subscriber line access multiplexers (DSLAMs), metro Ethernet switches and a broadband remote access aggregation server (BRAS). We obtain expressions for dimensioning the access network in the upstream direction. Specifically, we show that the bandwidth required at each DSLAM, metro Ethernet switch, and BRAS can be expressed as an exponential function of the subscribers in a log scale when the number of subscribers does not surpass a certain value. After this value, it grows linearly as a function of the subscribers. Copyright © 2009 John Wiley & Sons, Ltd.     

Interconnecting networks with optimized service provisioning

Telecommunication Systems - A recent trend of peering at geo-diversified Internet exchange points (IXPs) has empowered decades-old proposal of inter-networking and opened up new avenues of business...     

Resource provisioning for multigranular traffic flows in multilayer networks

The paper deals with the problem of multilayer resource provisioning in a network in which an operator needs to route traffic flows with different data rates. It is shown that the blocking probability of high data rate requests is usually higher than the blocking probability of other requests. A relevant strategy aimed at solving this problem is proposed, investigated, and compared with the known solution. Numerical results confirm that the proposed strategy is able to substantially reduce the blocking probability of high data rate requests at the cost of increased utilization of the packet switching layer and more complex routing of lightpaths.     

Forwarding state scalability for multicast provisioning in IP networks

Forwarding state scalability is one of the critical issues that delay the multicast deployment in IP networks. With traditional multicast routing protocols, a forwarding tree is built for each multicast session, and each router is required to maintain a forwarding entry for each multicast session whose distribution tree passes through the router. This poses the multicast forwarding state scalability issue when the number of concurrent multicast sessions is very large. We first present a survey of existing work addressing this scalability issue for providing scalable IP multicast. Then we extend an existing multicast routing protocol, Multicast Extension to OSPF (MOSPF), to scale well with respect to the number of concurrent multicast sessions by introducing tunnel support. This extension aims to reduce the protocol overhead associated with MOSPF. Simulation results show that the extension can significantly reduce multicast forwarding state and computational overhead at routers without affecting the per-destination shortest path characteristic of a resulting tree or introducing extra control overhead.     

Customer service provisioning in intelligent networks

The author points out that despite the convenience of using functional components (FCs) as service building blocks, service provisioning in the intelligent network (IN) still presents major problems. He focuses on issues of responsiveness to customers. When customers are allowed to provision new IN services, it is necessary to consider whether the set of FCs can be directly used by customers without cause problems and whether the set of FCs of flexible enough to construct the services customers are likely to want. The author investigates customer service provisioning problems and flexibility in using the IN FCs, and proposes methods to support integrity control for service development. For concreteness, he focuses on the set of FCs defined in Bellcores's IN/1+ network     

Service level agreement and provisioning in optical networks

This article proposes a service level agreement applied to the optical domain (O-SLA), which is expected to be the near- and long-term network technology of the great bandwidth capacity offered by optical devices. After an exposition of the rationale behind an optical SLA, parameters that could be included in this O-SLA, as well as their values for four classes of services are proposed. Different client (wavelength or subwavelength) and service types (from leased wavelength to bandwidth on demand) are distinguished when necessary. The last part of this article presents issues related to the provisioning of services emanating from this O-SLA.     

The QoS provisioning tri-mode energy saving mechanism for EPON networks

The Ethernet passive optical network provides broadband Internet access but also consumes a lot of energy. Energy saving mechanisms using the dual-mode—Active and Sleep modes—design for optical network unit (ONU) in EPON still suffer unnecessary energy consumption, especially in asymmetric data flow such as video streaming downloading service. The Doze mode is particularly suitable for handling the asymmetric data flow since it allows the ONU’s transmitter to turn off while turning on its receiver to receive data from optical line termination (OLT). However, adding Doze mode into original dual-mode design incur a greater challenge for OLT to identify the current status of the ONU since the ONU cannot transmit any upstream message to OLT at either Doze or Sleep mode. In this paper, we propose a new QoS provisioning tri-mode energy saving scheme, by integrating the Doze mode into original dual-mode design, to allow the ONU to switch to one of the energy saving modes whenever no upstream traffic exists. A high-priority upstream packet, arriving at ONU of energy saving modes, is able to trigger the ONU back to Active mode for QoS provisioning purpose. Performance evaluation via simulation has demonstrated the effectiveness of such mechanism in various asymmetric data flow. Furthermore, we propose two additional enhanced approaches to increase the energy saving effects by deferring the triggering action of the high-priority upstream packet as well as coalescing new arrival packets during waiting time into the same scheduling cycle.     

Availability-aware service provisioning in SD-EON-based inter-datacenter networks

The tremendous transmission capacity and flexible spectrum allocation scheme make elastic optical networks (EONs) one of the most promising infrastructures for constructing the interconnection in datacenter (DC) networks. Meanwhile, as DC traffics exhibit highly dynamic and heterogeneous features, differentiated service provisioning schemes are desired. In this paper, we take the advantage of centralized network control and management provided by the software-defined elastic optical networks (SD-EONs) and investigate availability-aware service provisioning in SD-EON-based inter-datacenter (inter-DC) networks. We first describe the problem of availability-aware service provisioning in SD-EON-based inter-DC networks and present the theoretical analysis for service availability. Then, we propose an availability-aware service provisioning algorithm (ASP) that leverages different path protection schemes to satisfy different service availability requirements. A service downgrading (SD) strategy is also designed as a supplement of ASP to further improve its performance. Simulation results show that the proposed ASP-SD algorithm can effectively improve the spectrum efficiency without sacrificing availability.     

Algorithms for provisioning virtual private networks in the hose model

Virtual private networks (VPNs) provide customers with predictable and secure network connections over a shared network. The recently proposed hose model for VPNs allows for greater flexibility since it permits traffic to and from a hose endpoint to be arbitrarily distributed to other endpoints. We develop novel algorithms for provisioning VPNs in the hose model. We connect VPN endpoints using a tree structure and our algorithms attempt to optimize the total bandwidth reserved on edges of the VPN tree. We show that even for the simple scenario in which network links are assumed to have infinite capacity, the general problem of computing the optimal VPN tree is NP-hard. Fortunately, for the special case when the ingress and egress bandwidths for each VPN endpoint are equal, we can devise an algorithm for computing the optimal tree whose time complexity is O(mn), where m and n are the number of links and nodes in the network, respectively. We present a novel integer programming formulation for the general VPN tree computation problem (that is, when ingress and egress bandwidths of VPN endpoints are arbitrary) and develop an algorithm that is based on the primal-dual method. Our experimental results with synthetic network graphs indicate that the VPN trees constructed by our proposed algorithms dramatically reduce bandwidth requirements (in many instances, by more than a factor of 2) compared to scenarios in which Steiner trees are employed to connect VPN endpoints.     

IDC服务内容和提供策略

本文基于对现在IDC市场IDC服务及其相关功能进行了详细描述，并提出了具体的需求和对策。     

A balancing scheme for QoS-aware service provisioning in OPS networks

The service level agreement (SLA) specifies the service availability that a service provider promises to a customer. Through introducing SLA into optical packet switching networks, a balancing scheme for quality of service-aware service provisioning is proposed, which uses packet loss ratio as availability index to obtain a larger number of SLA-satisfied services and achieve a better balance among different level services by means of discarding packets with better packet loss capacity in priority to fully make use of the capacity of services themselves. The theoretical analysis and simulation results show that the proposed scheme outperforms the other two typical service provisioning schemes in terms of the above two aspects.     

End-to-end QoS provisioning in mobile heterogeneous networks

The remarkable advances in information technologies bring a heterogeneous environment for mobile users and service providers. This heterogeneity exists in wireless access technologies, networks, user terminals, applications, service providers, and so on. The ability to provide seamless and adaptive quality of service in such a heterogeneous environment is key to the success of next-generation wireless communications systems. There has been a considerable amount of QoS research recently. However, the main part of this research has been in the context of individual architectural components, and much less progress has been made in addressing the issue of an overall QoS architecture for the mobile Internet. This article first summarizes the state-of-the-art QoS techniques and standardization activities, then examines in detail important challenges in building a ubiquitous QoS framework over the heterogeneous environment, and finally proposes a QoS framework integrating a three-plane network infrastructure and a unified terminal cross-layer adaptation platform to provide seamless support for future applications.