Edge-limited scalable QoS flow set-up

Although the Differentiated Services architecture supports scalable packet forwarding based on aggregate flows, the detailed procedure of Quality of Service (QoS) flow set-up within this architecture has not been well established. In this paper we explore the possibility of a scalable QoS flow set-up using a sink-tree paradigm. The paradigm initially constructs a sink tree at each egress edge router using network topology and bandwidth information provided by a QoS extended version of Open Shortest Path First (OSPF), which is a widely used link-state routing protocol. Our sink-tree paradigm dynamically reallocates network bandwidths online according to traffic demands. As a consequence, our paradigm easily supports QoS routing, resource allocation, and admission control at ingress edge routers without consulting core routers in a way that the QoS flow set-up time and overhead are minimized. Simulation results are very encouraging in that the proposed methodology requires significantly less communication overhead in setting up QoS flows compared to the traditional per-flow signaling-based methodology while still maintaining high resource utilization.     

区分服务网络的一种基于聚集状态的动态资源管理方案

提出了一种适用于区分服务网络的确保服务的动态资源提供方案:基于聚集状态的分布式动态资源管理。该方案在网络节点基于聚集状态进行接纳控制和动态资源预留,不需要核心节点保存单个流的状态,因此,该方案是可扩展的。方案包括一种轻量级动态资源预留协议和网络节点动态资源管理的相关算法,如状态建立、管理和接纳控制决策算法。该方案中,网络节点聚集状态的建立使用了基于实时流量测量的方法,在提供服务质量(QualityofService,QoS)保证的同时,得到了统计复用的增益,提高了资源的利用率,简化了信令协议,也使该方案具有很好的健壮性。     

基于MPLS的区分服务技术

传统的Internet只能提供尽力传送服务,但这种没有任何保证、不可预测的服务已不能满足许多应用的需要。IETF提出的Intserv/RSVP方案从技术角度可以提供灵活的服务质量,满足各种应用的需要,但该方案要求每个路由器保存每个连接的状态,复杂化了核心路由器的处理,因此实现 和配置是非常困难的。IETF提出的Diffserv将各种复杂的接入控制、每个连接的管理交给边界路由器处理,核心路由器只处理流量聚合,因而具有更好的伸缩性和鲁棒性。MPLS是一种可以在多种第二层媒质上进行标签交换的网络技术,将探讨如何结合Diffserv和MPLS两种技术,提供各种服务质量,特别是VPN。     

流感知网络的可扩展性研究

流感知网络作为新型综合业务区分服务质量保障体系,为关键业务提供了端到端的性能保障。它需要维护每个流的状态信息并基于单流实现策略控制,因此可扩展性便成为其应用基础。流感知网络可扩展性问题可分解为每流调度可扩展性和进程流规模控制策略两个方面,证明单流调度策略独立于链路速率且调度流数量有限是核心内容,流规模的控制策略说明流感知网络是基于有限流进行控制和管理,体系具有应用可扩展性。     

FIAC: a resource discovery-based two-level admission control for differentiated service networks

Differentiated Service (DiffServ) architecture has been proposed as a scalable QoS architecture for Internet. DiffServ, however, could not control its loads under heavy traffic conditions, and it could not provide strong QoS responses for individual traffic flow. This paper introduces at the edges of a DiffServ domain, a novel two-level admission control scheme, the Fair Intelligent Admission Control (FIAC). At per-class level, FIAC admits traffic according their fairshare and usage while preventing possible congestion within the DiffServ core. At per-flow level, FIAC estimates and allocates the fairshare for each flow within a class.

Simulation results demonstrate that FIAC adheres extremely well to the bandwidth requirements of DiffServ classes while preventing congestion within the DiffServ core. With FIAC's per-flow level enabled, FIAC is able to prevent uncontrolled UDP flows from depriving TCP-controlled flows of bandwidth share when they are aggregated to the DiffServ domain.     

Revisiting flow-based load balancing: Stateless path selection in data center networks

Hash-based load-balancing techniques are widely used to distribute the load over multiple forwarding paths and preserve the packet sequence of transport-level flows. Forcing a long-lived, i.e., elephant, flow to follow a specific path in the network is a desired mechanism in data center networks to avoid crossing hot spots. This limits the formation of bottlenecks and so improves the network use. Unfortunately, current per-flow load-balancing methods do not allow sources to deterministically force a specific path for a flow.In this paper, we propose a deterministic approach enabling end hosts to steer their flows over any desired load-balanced path without relying on any packet header extension. By using an invertible mechanism instead of solely relying on a hash function in routers, our method allows to easily select the packet’s header field values in order to force the selection of a given load-balanced path without storing any state in routers.We perform various simulations and experiments to evaluate the performance and prove the feasibility of our method using a Linux kernel implementation. Furthermore, we demonstrate with simulations and lab experiments how MultiPath TCP can benefit from the combination of our solution with a flow scheduling system that efficiently distributes elephant flows in large data center networks.     

Fairness of traffic controls for inelastic flows in the Internet

In best-effort networks, fairness has been used as a criterion to guide the design of traffic controls. The notion of fairness has evolved over time, from simple equality to a form of equality modulated by the user’s need (e.g. max–min and proportional fairness). However, fairness has always been defined on a per-user basis for a deterministic workload. In this paper, we argue that we must redefine the notion of fairness when we study traffic controls for the co-existence of elastic and inelastic traffics. Our results indicate that subjecting inelastic flows to fairness congestion control on a per-flow basis does not necessarily maximize the network’s utility. Instead, inelastic flows may follow their own form of traffic control, such as admission control (without congestion control). At the aggregate level, our results indicate that it still makes sense to maintain a balance between elastic and inelastic traffic. In order to support our arguments, we develop a methodology for comparing different traffic controls for given utility functions and different workloads, both deterministic and stochastic.     

A core stateless bandwidth broker architecture for scalable support of guaranteed services

We present a novel bandwidth broker architecture for scalable support of guaranteed services that decouples the QoS control plane from the packet forwarding plane. More specifically, under this architecture, core routers do not maintain any QoS reservation states, whether per-flow or aggregate. Instead, the QoS reservation states are stored at and managed by a bandwidth broker. There are several advantages of such a bandwidth broker architecture. Among others, it avoids the problem of inconsistent QoS states faced by the conventional hop-by-hop, distributed admission control approach. Furthermore, it allows us to design efficient admission control algorithms without incurring any overhead at core routers. The proposed bandwidth broker architecture is designed based on a core stateless virtual time reference system developed recently. This virtual time reference system provides a unifying framework to characterize, in terms of their abilities to support delay guarantees, both the per-hop behaviors of core routers and the end-to-end properties of their concatenation. We focus on the design of efficient admission control algorithms under the proposed bandwidth broker architecture. We consider both per-flow end-to-end guaranteed delay services and class-based guaranteed delay services with flow aggregation. Using our bandwidth broker architecture, we demonstrate how admission control can be done on a per domain basis instead of on a "hop-by-hop" basis. Such an approach may significantly reduce the complexity of the admission control algorithms. In designing class-based admission control algorithms, we investigate the problem of dynamic flow aggregation in providing guaranteed delay services and devise a new apparatus to effectively circumvent this problem. We conduct detailed analyses to provide theoretical underpinning for our schemes as well as to establish their correctness. Simulations are also performed to demonstrate the efficacy of our schemes.     

Integration of admission,congestion, and peak power control in QoS-aware clusters

Admission, congestion, and peak power control mechanisms are essential parts of a cluster network design for supporting integrated traffic. While an admission control algorithm helps in delivering the assured performance, a congestion control algorithm regulates traffic injection to avoid network saturation. Peak power control forces to meet pre-specified power constraints while maintaining the service quality by regulating the injection of packets. In this paper, we propose these control algorithms for clusters, which are increasingly being used in a diverse set of applications that require QoS guarantees. The uniqueness of our approach is that we develop these algorithms for wormhole-switched networks, which have been used in designing clusters. We use QoS-capable wormhole routers and QoS-capable network interface cards (NICs), referred to as Host Channel Adapters (HCAs) in InfiniBand™ Architecture (IBA), to evaluate the effectiveness of these algorithms. The admission control is applied at the HCAs and the routers, while the congestion control and the peak power control are deployed only at the HCAs. A mixed workload consisting of best-effort, real-time, and control traffic is used to investigate the effectiveness of the proposed schemes.     

区分服务网络中基于主动队列管理的病态流控制

在区分服务（Differentiated Services,DiffServ）网络中,为了消除病态流对确定性转发（Assured Forwarding,AF）服务的不良影响,提出了一种基于RIO（RED IN and OUT）的主动队列管理机制RIO SD。这种机制不需要在核心路由器维护每一个流的状态,而是通过观测虚拟队列的丢包历史记录来鉴别病态流,并通过虚拟队列的前置滤波器加大病态流的丢包率,实现对病态流的控制。仿真结果表明RIO SD可以有效抑制病态流对带宽的占用,提高其他正常流的性能。     

Conformance analysis in networks with service level agreements

To achieve some level of Quality of Service (QoS) assurance, a network usually has Service Level Agreements (SLAs) with its users and neighboring domains, which describe the QoS level that the service provider is committed to provide, and the specification of traffic that users or neighboring domains are allowed to send. An interesting and important question arises as to whether a flow is still conformant to its original traffic specification after crossing the network since it may interact with other flows within the network. In this paper, we study analytically the extent to which a flow and an aggregate of flows become non-conformant through an analysis of the stochastic burstiness increase of flows after crossing a per-flow scheduling network and an aggregate scheduling network . The stochastic behavior of a server in aggregate scheduling networks is also studied to determine the conformance deterioration of individual flows, which provides the theoretical conformance deterioration bound and provides useful results for conformance analysis in an aggregate scheduling network with general topology. Our theoretical results are verified by extensive simulations.     

A novel contention-on-demand design for WiFi hotspots

In widely deployed wireless “hot-spot” networks, nodes frequently join or leave, and inelastic/elastic and saturated/nonsaturated flows coexist. In such dynamic and diverse environments, it is challenging to maximize the channel utilization while providing satisfactory user experiences. In this paper, we propose a novel contention-on-demand (CoD) MAC scheme to address this problem. The CoD scheme consists of a fixed-CW algorithm, a dynamic-CW algorithm, and an admission control rule. The fixed-CW algorithm allows elastic flows to access limited system bandwidth; the dynamic-CW algorithm enables inelastic flows to contend for channel on demand and quickly adapt to network change; and the admission control rule rejects overloaded traffic for providing good user experiences. We then perform an asymptotic analysis to develop a simple and practical admission control rule for homogeneous and heterogeneous traffic; our rule can not only adapt to the change in offered loads and node number, but also maximize the system utilization. Finally, extensive simulations verify that our scheme is very effective and our theoretical result is very accurate.     

Mobility label based network: Hierarchical mobility management and packet forwarding architecture

Scalability of the network layer mobility management solution is one of the most important requirements for the mobility control plane. Mobility Label Based Network (MLBN) is a new approach to the network layer mobility management problem that relies solely on MPLS to provide both macro- and micro-mobility for IPv4 and IPv6 mobile hosts and routers. This new approach does not rely on the existing IP mobility management protocols such as Mobile IP and is based on the combination of Multi-Protocol BGP (MP-BGP) and MPLS. In the context of the MLBN the scalable control plane should be capable of efficient Mobility Label distribution while allowing the MPLS-based forwarding plane to deliver mobile traffic in an optimal manner. This paper presents a hierarchical mobility management system capable of both macro- and micro-mobility support without the use of Mobile IP and its derivatives and allows scalable Mobility Label distribution and MPLS label stack based packet forwarding in support of optimal traffic delivery between the communicating mobile users.     

新型路由器体系结构研究综述

计算机网络的管理与控制变得越来越复杂,人们对网络互连的关键设备--路由器提出了越来越高的要求。新型路由器控制体系结构必须考虑系统的开放性、可扩展性、可伸缩性、安全性、用户选择与应用感知等问题。本文分析了新型路由器控制体系结构的研究现状,重点介绍了主动网络、可编程网络、开放网络控制、软件可扩展路由器以及面向用户选择的网络控制等技术。     

Efficient link scheduling for online admission control of real-time traffic in wireless mesh networks

Link scheduling is used in wireless mesh networks (WMNs) to guarantee interference-free transmission on the shared wireless medium in a time division multiple access approach. Several papers in the literature address the problem of link scheduling guaranteeing a minimum throughput to the flows traversing the WMN. However, none of the existing works address the problem of computing a schedule that guarantees that pre-specified end-to-end delay constraints are met. In this paper, we make a first step forward in this direction by defining a link scheduling algorithm that works in sink-tree WMNs, i.e. those whose traffic is routed towards a common sink (i.e. the Internet gateway). Our iterative algorithm exploits a delay-based admission control procedure, devised through Network Calculus, which solves an optimization problem and tests the feasibility of a schedule from the point of view of delay guarantees. Thanks to a clever solution approach for the optimization problem, the iterative algorithm computes feasible solutions in affordable times for networks of several tens of nodes, and is thus amenable to online admission control of real-time traffic.     

利用直连网络实现可扩展路由器

Internet的迅速发展直接表现为用户流量的迅速增长,这就要求路由器必须提供更大的容量.传统的路由器由线卡和集中式交换网络构成.集中式交换网络只能支持有限的端口数目,而且随着端口数目的增加,调度算法也变得越来越复杂,所以交换网络正成为整个路由器的性能瓶颈.集中式交换网络还是路由器的单一失效点,无法提供令人满意的容错性能.直连网络具有良好的扩展性和容错性.其中,3-D Torus拓扑结构已被成功应用到可扩展路由器的设计当中.但是在实际应用中,3-D Torus结构受到等分带宽的约束,限制了扩展规模.介绍了一种新型的直连网络结构,称为蜂巢式结构.将对蜂巢结构作简单的改动,修改后的拓扑表现出很好的拓扑属性.基于该结构,提出了两类最短路径路由算法.其中,负载均衡的最短路径路由算法较好地利用了直连网络路径多样性的特点,针对均匀随机和Tornado两种类型的流量都表现出较低的分组延时和较高的吞吐量.另就队列长度和单节点调度算法等方面对路由算法的影响进行了讨论.蜂巢结构为可扩展路由器的设计提供了新的选择.     

Linux环境下路由器中的网络带宽管理

Linux是目前广泛用于路由设备中的操作系统,而流量管理是这种网络操作系统的一个重要功能.研究了Linux系统的流量管理机制,发现当前Linux系统所采用的在网络接口的出口实现的基于网络包调度的流量管理机制缺乏对于网络带宽在系统范围的全局性的管理,也缺乏对于输入流的网络处理的有效管理和调度,从而造成不必要的CPU资源的浪费.为解决这一问题,提出了一种新颖的网络带宽管理机制,它通过调度网络协议处理所占用的CPU时间来实现带宽管理.这种新的机制将网络带宽的管理和调度从网络接口的出口点转移到处理接收到的网络包的系统软件中断处理程序中,从而克服了原来的流量管理机制的缺点.通过系统实现和对比实验,发现该机制本身给系统带来的负载小于Linux原来的流量管理机制,同时可以提供更好的流量隔离,并且能够有效地节省用于网络处理的CPU资源.     

Availability of end-to-end ideal QoS in IP packet networks

Network processor technology has advanced to the point where high-precision time-based store-and-forward logic is readily incorporated into packet switches and routers. With appropriate scheduling, packets from multiple flows can be serviced without contending for link resources. Accordingly, packet flows traversing a network of switching elements can have both path and time determinacy attributes which support ideal end-to-end QoS (zero jitter, zero loss, acceptable end-to-end latency) for real-time UDP packet flows and guaranteed goodput for TCP flows. One approach to packing a network with a relatively large number of such deterministic flows, i.e. achieving high availability of the ideal QoS service in a network, uses precise buffering of packets at each switch, which introduces latency. This paper describes analysis methods for quantifying how much buffering may be necessary to achieve high (99.999%) availability. For typical network topologies the analysis shows that buffering latency requirements are very small compared to transport delays, even when the network is highly utilized with heterogeneous (e.g. voice, video, circuit emulation, and data) traffic. Actual physical implementations have empirically validated the analysis results as well as the scalability of the end-to-end, time-based forwarding approach and the end-to-end availability of ideal QoS services in IP packet networks.     

一种区分服务网络的设计与实现

给出了DiffServ网络的一种设计与实现，提出了一种有效的资源管理模型。在该模型中，存在一个带宽代理(Bandwidth Broker)同步多个边界节点的接纳控制，负责区分服务网络之间的SLA(Server Level Agreement)协商，进行路由器配置；边界节点使用RSVP协议为集聚流预留资源；RSVP使用预计算QoS路由寻找路径。整个模型不仅综合了已有资源管理方案的优点，而且保持了区分服务网络的可扩展性。     

Mean–variance relationship of the number of flows in traffic aggregation and its application to traffic management

We consider the mean–variance relationship of the number of flows in traffic aggregation, where flows are divided into several groups randomly, based on a predefined flow aggregation index, such as source IP address. We first derive a quadratic relationship between the mean and the variance of the number of flows belonging to a randomly chosen traffic aggregation group. Note here that the result is applicable to sampled flows obtained through packet sampling. We then show that our analytically derived mean–variance relationship fits well those in actual packet trace data sets. Next, we present two applications of the mean–variance relationship to traffic management. One is an application to detecting network anomalies through monitoring a time series of traffic. Using the mean–variance relationship, we determine the traffic aggregation level in traffic monitoring so that it meets two predefined requirements on false positive and false negative ratios simultaneously. The other is an application to load balancing among network equipments that require per-flow management. We utilize the mean–variance relationship for estimating the processing capability required in each network equipment.