Buffer management and scheduling schemes for TCP/IP over ATM‐GFR

Today ATM technology is facing challenges from integrated service IP, IP switching, gigabit IP router and gigabit ethernet. Although ATM is approved by ITU‐T as the standard technology in B‐ISDN, its survivability is still in question. Since ATM‐UBR (unspecified bit rate) provides no service guarantee and ATM‐ABR (available bit rate) is still unattainable for most users, many existing users have little or no incentives to migrate to ATM technology. The guaranteed frame rate (GFR) service is introduced to deal with this dilemma. The GFR can guarantee the minimum cell rate (MCR) with fair access to excess bandwidth. This paper studies various schemes to support the GFR. We have studied different discarding and scheduling schemes, and compared their throughput and fairness when TCP/IP traffic is carried. Through simulations, it is shown that only per‐VC queueing with weighted round robin (WRR) can guarantee minimum cell rate. Among all the schemes that have been explored, we recommend dynamic threshold–early packet discard (DT–EPD) integrated with MCR+ (a WRR variant) to support the GFR service. Copyright © 2001 John Wiley & Sons, Ltd.     

Traffic management for TCP/IP over satellite ATM networks

Several Ka-band satellite systems have been proposed that will use ATM technology to seamlessly transport Internet traffic. The ATM UBR, GFR, and ABR service categories have been designed for data. However, several studies have reported poor TCP performance over satellite ATM networks. We describe techniques to improve TCP performance over satellite ATM networks. We first discuss the various design options available for TCP end systems, IP-ATM edge devices, as well as ATM switches for long-latency connections. We discuss buffer management policies, guaranteed rate services, and the virtual source/virtual destination option in ATM. We present a comparison of ATM service categories for TCP transport over satellite links. The main goal of this article is to discuss design and performance issues for the transport of TCP over UBR, GFR, and ABR services for satellite ATM networks     

A detailed experimental performance evaluation on TCP over UBR

The ATM service category UBR is intended for non-real-time applications that do not require guaranteed QoS commitments. With additional, relatively inexpensive control functions such as packet discard schemes, UBR could become a cost-effective alternative for the transmission of data traffic, offering a straightforward and flexible solution as opposed to nrt-VBR and GFR that applies stricter traffic specifications as well as ABR with its sophisticated and complex rate-control protocol. This paper presents the results obtained from a comprehensive set of experiments with TCP over UBR, comprising measurements taken on different protocol layers. The goal is to experimentally investigate the performance of UBR to carry TCP traffic, to evaluate the performance gain achievable by packet discard schemes and TCP parameter tuning, to study the influence of the TCP implementation, and in a final step, to relate the measurements to simulation results. This revised version was published online in August 2006 with corrections to the Cover Date.     

ATM traffic management in an LMDS wireless access network

We investigate the capacity of LMDS to support ATM services in the local loop. In particular, we evaluate the performance of a MAC protocol for this system when transporting voice and IP traffic using the VBR and GFR service categories of ATM, respectively. Our results show that the MAC protocol is well suited for voice traffic but in general lacks efficient bandwidth management mechanisms to support the more dynamic bandwidth requirements of IP traffic     

Data communication in ATM networks

An increasing number of customers require LAN access with high bandwidth and low delay over long distances. To satisfy these needs, several high-speed network techniques have been developed. Asynchronous transfer mode (ATM) is superior compared to other networking technologies, as it offers high bandwidth and is scalable in the sense that the bandwidth capacity of an ATM system is not fundamentally limited to the technology itself. Initial ATM installations will operate as subnetworks of existing networks and MAC layer protocols. One of the main challenges in ATM is the transparent support of existing connectionless LAN services. Several activities have been launched within international standard bodies and forums to specify ways of providing data communication services over ATM. Most notable examples are Switched Multimegabit Data Service (SMDS) and the similar Connectionless Broadband Data Service (CBDS) supported mostly by public network service providers. But also other approaches such as IP over ATM, and LAN emulation that show more adherence to the existing local and campus area networking paradigms. The article presents the requirements and architecture of the LAN emulation service. It describes the alternative methods for carrying IP packets over ATM, a public broadband service architecture and CBDS. The traffic management aspects of the data communication services are also discussed     

Modelling and analysis of routing and resource allocation techniques in multi‐service networks

In this paper a novel call level model based on the extension of the classical Erlang multi‐rate model for broadband integrated services networks is proposed. We use the model to study routing strategies in multi‐service networks where service classes with/without QoS guarantees coexist. Examples for such networks include ATM and IP‐based integrated networks. In ATM, the CBR and VBR service classes provide QoS guarantees, while the ABR and UBR service classes are of the best effort type. In IP, traditional TCP/IP traffic is of the best effort type, while new protocols like the RSVP or the differentiated services with central resource handling attempt to support QoS parameters. The coexistence of guaranteed and best effort traffic gives rise to new challenging problems since for a given elastic (best effort) connection the bottleneck link determines the available bandwidth and thereby puts constraints on the bandwidth at the other links along the connection's path. Since the available bandwidth fluctuates in time following the load on the links, routing and link allocation in this environment together with blocking probability calculations and fairness issues need to be studied. By means of our proposed model we are able to conduct a survey of various routing and link allocation techniques as well as to develop a modified shortest path routing algorithm which, according to the numerical examples, performs well in this environment. Copyright © 1999 John Wiley & Sons, Ltd.     

An update on ATM traffic management

The Traffic Management (TM) group at the ATM Forum has been fairly focused on developing a new service called guaranteed frame rate (GFR) service. This service is expected to be positioned as an intermediate alternative between the available bit rate (ABR) service, which uses a closed-loop feedback control framework, and the unspecified bit rate (UBR) service, which provides no assurances whatsoever. The motivation for GFR service comes from the perceived need for frame-based minimum rate guarantees. The article gives an update on the status of the GFR service definition, as well as other topics discussed in the ATM Forum     

Performance and buffering requirements of Internet protocols overATM ABR and UBR services

ATM networks are quickly being adopted as backbones over various parts of the Internet. This article studies the dynamics and performance of the TCP/IP protocol over the ABR and UBR services of ATM networks. Specifically the buffering requirements in the ATM switches as well as the ATM edge devices. It is shown that with a good switch algorithm, ABR pushes congestion to the edges of the ATM network while UBR leaves it inside the ATM portion. As a result, the switch ABR buffer requirement for zero-packet-loss high-throughput TCP transmission is a sublinear function of the number of TCP connections     

ATM通信量管理技术及其在视频点播中的应用

ATM现有的5种通信量业务类型,即：恒定比特率（CBR）、可变比特率（VBR）、可用比特率（ABR）、不定比特率（UBR）和可保证帧速率（GFR）业务,在应用中各有所长。阐述了将ATM ABR业务运用于视频点播系统时,客户端和服务器端的缓存尺寸设置方法,为ABR业务在视频点播系统中的实际应用提供了有力的理论基础。     

Integrating Differentiated Services with ATM

IP in the edge and ATM in the core are commonplace in today's internetworks. The IETF has proposed a new Quality of Service (QoS) mechanism namely Differentiated Services (DiffServ) for IP networks. On the other hand, QoS is an inherent feature in ATM. It is imperative that IP and ATM QoS interoperate efficiently to provide an end-to-end service guarantee. DiffServ provides a class of service named Assured Forwarding (AF) that does not exactly correlate to any of the service categories offered by ATM. AF is targeted towards a range of applications, such as real-time (rt) that do not require a constant bit rate service provided by Expedited Forwarding, and other non-real-time (nrt) applications that expect a service better than best effort. In this paper we propose the mapping of AF to the Variable Bit Rate (VBR) service category in ATM. VBR is suitable because it is available in the form of rt-VBR and nrt-VBR and could be translated appropriately based on the applications. The mapping is implemented and verified using the LBNL Network Simulator. The results of the experiments show that VBR is a better match for AF than any other service category in ATM.     

Design and evaluation of multicast GFR for supporting TCP/IP over hybrid wired/wireless ATM

The major challenges of designing multicast traffic control protocols for a combined wired/wireless network are the varying transmission characteristics (bandwidth, error, and propagation delay) of the wireless and wired media, and the different, possibly conflicting frame rate requests from multiple sources. To address these issues, in this paper we design and evaluate new unicast and multicast guaranteed frame rate (GFR) schemes for supporting TCP/IP traffic over a combined wired/wireless ATM network. We first propose a new, flexible weighted buffer management, and a frame‐based virtual spacing (VS) mechanism implementing weighted fair queueing. The unicast GFR scheme is based on the integration of the new weighted buffer management, and either cell‐based or frame‐based VS. It is then extended to support multicast GFR flows. The multicast scheme presented in this paper is the first multicast GFR scheme appeared in the literature. These schemes are carefully evaluated over several network configuration, supporting heterogeneous TCP/IP traffic with various frame rates. Simulation results show that the new schemes guarantee the minimum rates requested, provide excellent fairness, and achieve reasonably high efficiency. The new schemes may be extended to provide differentiated service in both IP and mobile IP frame work. This revised version was published online in July 2006 with corrections to the Cover Date.     

基于ATM的无源光网络媒质接入控制协议的设计

针对基于ＡＴＭ的无源光网络（ＡＴＭ－ＰＯＮ）设计了一种媒质接入控制（ＭＡＣ）协议。该协议的设计遵照ＩＴＵ－Ｔ有关ＡＴＭ－ＰＯＮ的建议Ｇ．９８３．１（１９９８年１０月）中所定义的帧结构。该协议能够支持ＣＢＲ／ＶＢＲ、ＡＢＲ和ＵＢＲ等多种业务,并能够保证用户公平地接入和充分地利用网络的资源。其中ＣＢＲ／ＶＢＲ业务的优先级最高,同时也确保ＡＢＲ业务至少能以最小信元通信。系统的剩余带宽资源被分配给ＵＢＲ业     

Guaranteed frame rate: a better service for TCP/IP in ATM networks

Guaranteed frame rate, approved by the ATM Forum, is expected to become an important service category to efficiently support TCP/IP traffic in ATM networks. We first describe the GFR traffic contract in detail. We then present different types of switch implementations that have been proposed to support GFR. We analyze the performance of three of these switch implementations by simulations in two different network environments. These simulations show that the scheduler-based implementations provide a much better performance than the simple switch implementation. However, we also show that coupling an active packet discard mechanism to a scheduler-based switch implementation does not produce a performance gain when many TCP connections are multiplexed inside one ATM VC     

MSTP应用和发展展望

在概述MSTP演进路线的基础上,文章从运营角度探讨了今后一段时期内MSTP设备应具有的功能和新业务支持能力,如VLAN组网能力、业务分类和服务质量(QoS)控制能力、多业务传送能力等.文章认为多业务传送节点(MSTP)继承了同步数字系列(SDH)的技术优势,同时又能承载以太网、IP和ATM等多种数据业务,并在同一个平台上实现统一的网络控制和管理,将发展成为城域传送网的主流技术.由于具有丰富的接口、灵活的带宽管理和服务质量保证机制,MSTP可很好地将传输技术与数据业务的发展需求结合起来,较好地解决电路和数据业务的综合传送,为电信运营商提供更加广阔的运营空间.     

A Partially-Blocking Queueing System with CBR/VBR and ABR/UBR Arrival Streams

In this paper we consider an ATM transmission link, to which CBR or VBR and ABR or UBR calls arrive according to independent Poisson processes. CBR/VBR calls (characterized by their equivalent bandwidth) are blocked and leave the system if the available link capacity is less than required at the time of arrival. ABR/UBR calls, however, accept partial blocking, meaning that they may enter service even if the available capacity is less than the specified required peak bandwidth, but greater than the so called minimal accepted bandwidth. Partially blocked ABR/UBR calls instead experience longer service time, since smaller given bandwidth entails proportionally longer time spent in the system, as first suggested in [3] and analyzed in details herein. Throughout the life time of an ABR/UBR connection, its bandwidth consumption fluctuates in accordance with the current load on the link but always at the highest possible value up to their peak bandwidth (greedy sources). Additionally, if this minimal accepted bandwidth is unavailable at the time of arrival, ABR/UBR calls are allowed to wait in a finite queue. This system is modeled by a Continuous Time Markov Chain (CTMC) and the CBR/VBR and ABR/UBR blocking probabilities and the mean ABR/UBR waiting- and service times are derived.     

QoS support for integrated services over CATV

Cable TV has emerged as a promising access network infrastructure for the delivery of voice, video, and high-speed data traffic. A central issue in the design of protocols for CATV networks is to support different levels of QoS for diverse user applications. While CATV service providers and equipment have standardized, in the so-called MCNS protocol, the basic network architecture and interfaces, issues in the MAC layer for QoS support are likely to be left for differentiation in vendor products. This article first presents an overview of the basic CATV network architectural assumptions and the set of QoS requirements for supporting integrated services over CATV. It then discusses a MAC layer scheduling protocol that can efficiently multiplex constant bit rate traffic, such as voice over IP with guaranteed delay bound, and best-effort traffic, such as data services with minimum bit rate guarantee, while achieving fairness on any excess available bandwidth. The performance of this algorithm is illustrated by simulation results using Opnet. We also discuss a dynamic polling mechanism that enhances the link utilization while preserving delay bounds for latency-critical traffic     

TCP/IPoverATM的性能研究

 本文讨论将传统的TCP拥塞控制机制和CIPOA用于采用ER(FB)算法和EFCI(CCR)算法的ATM网络时可用比特率业务(ABR)和未确定比特率业务(UBR)的性能.仿真结果表明,无论在缓冲区的需求、带宽分配的公平性、吞吐量和链路利用率方面ABR业务的TCP性能均明显优于UBR业务.对于较简单的网络模型ER(FB)算法的TCP性能优于EFCI(CCR)算法,更优于传统的EFCI算法.     

Real-time voice over packet-switched networks

We discuss the architecture and technical viability of transporting real-time voice over packet-switched networks such as the Internet. The value of integrating voice and data networks onto a common platform is well known. The telephony industry has proposed the ATM standard as a means of upgrading the Internet to provide both real-time and data services. In contrast, voice services may be added to traditional IP networks that were originally designed for data transmission alone. We consider the feasibility and expected quality of service of audio applications over IP networks such as the Internet. In particular, we examine possible architectures for voice over IP and discuss measured Internet delay and loss characteristics     

ATM局域网性能测试与分析

异步传输模式（ＡＴＭ）常被认为是一种解决网络带宽瓶颈的技术，但是应用程序从ＡＴＭ网络获得的带宽受到许多因素的限制。本文通过在ＣＥＲＮＥＴ的ＡＴＭ测试床上的实验，对基于ＡＴＭＡＰＩ的数据交换和基于ＴＣＰ／ＩＰ／ＡＴＭ的数据交换进行了对比。结果表明，原始的ＡＴＭ信元交换其效率是非常好的，但在不同协议栈上的应用所能获得的带宽受系统软件、传统的网络协议和端系统等诸多方面的限制。通过调整这些因素，可以提高应用程序可获得的带宽。     

ATM in MPLS-based converged core data networks

Services supported by asynchronous transfer mode account for the majority of data and Internet service revenues generated by carrier networks today. This is based on ATM's ability to support high availability services with quality of service. However, the influences of the Internet and a highly dynamic telecommunications market have raised demands for increased flexibility while controlling costs. Therefore, future carrier networks are likely to continue to be based on established technologies, such as ATM, as well as IP. In many cases, this is achieved through maintaining separate ATM and IP core networks, with the IP network supporting Internet services, while the ATM network continues to support guaranteed services such as private lines, broadband access, and video. In some cases, however, it can be advantageous for a carrier to transport segments of their ATM network over their IP network core; for example, to transport ATM traffic currently carried on leased facilities onto an IP network where the service provider owns the facilities. Developments in IP and MPLS-based traffic engineering and QoS may increase the ability of IP-based networks to support ATM services using MPLS. This article provides an overview of approaches enabling a network based on MPLS that naturally supports IP services to also support ATM services. The drivers and requirements for convergence on an IP/MPLS core network are presented, followed by an overview of the different approaches and associated challenges currently being debated in the standards bodies.