High-performance protocol architecture

The development of high-speed networking applications requires improvements to data communication protocols in order to efficiently provide the required services to the applications. In this survey paper, we will first give a rapid presentation of protocol optimization techniques and of some high-speed transport protocols developed for specific application needs.We will then present a new protocol architecture based on the “Application Level Framing” (ALF) and “Integrated Layer Processing” (ILP) concepts. ALF states that applications send data as a sequence of autonomous “frames”, which will be at the same time the unit of transmission, the unit of control and the unit of processing. This allows for efficient design of application specific protocols. It also enables ILP, i.e., the integration of multiple transmission control layers in a single loop, maximizing the efficiency of modern processors.We will present both the architectural design aspects and the corresponding implementation problems. Practical experimentations with ALF and ILP will also be described and their benefits and limitations assessed. We will also present a new approach to network architecture namely the active networks approach, and discuss its pros and cons.     

通信协议的并行机制研究

通信技术的迅速发展，通信网络所提供的带宽达到几百Ｍｂｐｓ～Ｇｂｐｓ数量级。一些应用，象多媒体通信，就能利用网络的高吞吐量。但是，目前的通信协议并不能适应高速传输的环境。运输层和其它较高层协议的处理，已成为通信系统的性能瓶颈。本文首先回顾近十年来通信环境所发生的一些变化，并分析其对运输层协议的影响；第二部分分析标准协议并行实现时的一些关键技术；最后介绍了一个运输层系统模型，该模型采用并行机制，向上提供高性能的运输层服务。     

Network-based multicomputers: a practical supercomputerarchitecture

Multicomputers built around a general network are an attractive architecture for a wide class of applications. The architecture provides many benefits compared with special-purpose approaches, including heterogeneity, reuse of application and system code, and sharing of resources. The architecture also poses new challenges to both computer system implementers and users. First, traditional local-area networks do not have enough bandwidth and create a communication bottleneck, thus seriously limiting the set of applications that can be run effectively. Second, programmers have to deal with large bodies of code distributed over a variety of architectures, and work in an environment where both the network and nodes are shared with other users. Our experience in the Nectar project shows that it is possible to overcome these problems. We show how networks based on high-speed crossbar switches and efficient protocol implementations can support high bandwidth and low latency communication while still enjoying the flexibility of general networks, and we use three applications to demonstrate that network-based multicomputers are a practical architecture. We also show how the network traffic generated by this new class of applications poses severe requirements for networks     

一种基于ATM的支持并行处理的高速通信机制

随着高速网络（如ＡＴＭ）的发展以及工作站性能的不断提高，工作站网络（ＮＯＷ）作为一种新型的并行计算结构越来越受到人们的重视。传统的传输协议和报文传递系统不能充分作为高速网的传输能力。本文提出一种基于ＡＴＭ的支持并行处理的高速通信机制ＨＰＭＰＡ。在ＨＰＭＰＡ中，可靠的端－端传输协议ＨＳＴＰ为并行应用提供高速可靠的数据传输，而不可靠的端－端传输协议ＵＴＰ则提供不可靠的高速数据报服务，以混合树结构为基础     

A parallel hardware approach to implement network protocols

The dramatic increase in the operating bandwidth of current communication networks requires a very high processing speed for communications protocols to offer users real broadband services.Although there are different causes determining slowness in protocol processing, some implementation approaches for communication protocols seem to be the most limiting factor. This paper proposes a parallel architecture, based on VLSI components, to implement the OSI stack system in HSLANs and MANs. The proposed architecture is coupled with a model based on Petri nets which can characterize the implementation of each level according to the necessary requirements. We show that a hard protocol, as the data link layer, implemented according to the proposed architecture can effectively work at a data rate which is adequate for current networks.     

MAP协议集及其在CIMS中的实现方法

MAP协议集是CIMS制造自动化分系统理想的网络体系结构.本文论述第一层基于IEEE 802.3协议的MAP功能的实现以及IEEE 802.3、IEEE 802.4协议的特点、CSMA/CD总线网的效率、响应时间,给出了基于IEEE 802.3 MAP功能实现的限定条件.     

基于高速通信协议的COSMOS机群文件系统性能研究

作为曙光3000超级服务器的重要组成部分,COSMOS机群文件系统对机群文件系统协议,结构及性能优化等问题进行全面深入的探讨,首先描述了基于曙光3000机群高速通协议BCL－3的COSMOS文件系统的实现,然后引入并发带宽利用率,描述了通信与I／O对机群文件系统性能影响程序,最后介绍了有关性能实验并对实验结果作出解释。     

运输层协议差错恢复的数学模型与性能分析

差错恢复是运输层协议差错控制及其它相关协议机制和控制策略的重要内容,对运输协议的功能和性能有着显著的影响,因此运输层协议差错恢复的模型建立和分析工作就相当关键。本文先简单介绍传统和新型运输层协议在差错恢复方面的功能差异,定义性能分析和比较度量,再着重建立后退重发和选择重发差错恢复控制策略的数学模型,分析上述控制策略的性能差异,及对高速网络和新型应用的适应程度。     

High speed architectures for Leviathan: a binary tree based stream cipher

Connectionless routed networks, built atop high-speed communication medium, require cryptographic algorithms capable of out-of-order keystream generation and high throughput. Binary tree based stream ciphers, of which Leviathan is an example, are capable of meeting both of these requirements. We investigate high-speed architectures for the binary tree traversal and show that the traversal approaches discussed can be extended to m-ary tree of height h. Of the two architectures presented, the pipeline architecture computes keystream at uniform rate and the parallel architecture bounds the worst-case variance in the time period between computations of consecutive output key words, which form the keystream. The design and implementation of Leviathan keystream generator based on the pipeline architecture for binary tree traversal are presented. We show that it is possible to achieve keystream generation rates approaching 1 Gbps with the pipeline architecture. The design was implemented in two parts, the keysetup and the keystream pipeline, targeting commercially available Xilinx XC2V4000 and XC2V3000 FPGAs. The keystream pipeline implementation operated at frequency of 50 MHz and occupied 6864 slices. The results were verified performing the timing simulation.     

Unreliable transport protocol using congestion control for high-speed networks

Currently there is no control for the real-time traffic of multimedia applications using UDP (User Datagram Protocol) in high-speed networks. Therefore, although a number of high-speed TCP (Transmission Control Protocol) protocols have been developed for gigabit-speed (or faster) links, the real-time traffic could also congest the network and result in unfairness and throughput degradation of TCP traffic. In this paper, a new unreliable transport protocol, FAST DCCP, is presented for the real-time traffic in high-speed networks. FAST DCCP is based on the DCCP protocol and adopts the FAST scheme to realize congestion control. Some modifications have been made to the mechanisms inherited from DCCP so as to let the proposed protocol can efficiently operate under a large size window. In addition, an enhanced protocol, EEFAST DCCP, using the measurements of one-way delay to dynamically adjust the window size is also proposed to improve the throughput of FAST DCCP with the effect of reverse traffic. Simulation results show that FAST DCCP not only can satisfy the requirements of real-time data delivery, but also perform well in bandwidth utilization and fairness in high-speed networks. Meanwhile, EEFAST DCCP is able to effectively conquer the throughput degradation caused by the reverse traffic.     

A conceptual model of service customization and its implementation

With the development of Internet and next generation networks in telecommunications, more and more new services are required to be introduced into networks. Introducing new services into traditional network is always associated with standardizing new protocols. The progress of protocol standardization usually takes several years, which cannot meet the increasing demands of the applications in Internet and next generation networks. Service customization in network systems may be one possible solution to cope with this problem. Based on the principle that network service is provided by interactions among protocol entities, this paper proposes a conceptual model of service customization (SECUM) by separating the service logic from protocol interactive logic within existing network architecture. The theory of Communicating Sequential Processes (CSP) is used to formalize the SECUM in order to locate exactly the service logic and to define precisely the SECUM. For validating the SECUM‘s usability in practical network systems, this paper also proposes an implementation model for SECUM: a component-based protocol implementation model (CPIM). CPIM discomposes protocol entity into application component, service component, message component and communication component. Service component associates application component with message component. Users or network managers can customize network services by configuring service component. The paper shows respectively the applications of SECUM and CPIM by proposing a customizable IP service model based on SECUM and describing an implementation of Session Initiation Protocol (SIP) based on CPIM. Compared with the existing service-customization techniques, SECUM is a service customization model internal to network system and may provide more powerful capabilities of service customization.     

基于SEP5010嵌入式系统中USB功能的实现

随着数字信号处理速度的提高,串口、并口已不能满足主机与嵌入式设备之阍的高速数据传输要求．设计了一种采用具有东南大学自主知识产权ARM9的SEP5010芯片硬件平台的USB系统功能,简单分析USB的驱动架构并在应用层中提出了新的传输协议,提出了一种对平台中的USB模块进行高性能读写的设计方案,使得USB模块达到了702KB／s的读写速度。同时此测试方案的不依赖于底层硬件,可以在适用于各种类型的硬件平台。     

Uninet SS:支持网络融合和网络开放的软交换系统

讨论了支持网络融合和网络开放的软交换系统的特点，提出了一个层次化的软交换体系结构，包括协议适配子系统、呼叫控制子系统、业务适配子系统和OAM子系统。分析了层次化的呼叫控制模型、基于统一消息的协议适配方法和可扩展的Parlay API实现方案。测试结果表明了该系统能够满足国家标准规定的性能要求。     

一种基于RTLinux的实时网络子系统

提出了一种基于RTLinux的实时网络子系统。该系统能为不同应用的数据包进行资源的预留,再分别建立独自的协议处理线程以及与之绑定的实时任务进程,通过维护3条不同优先级机制的调度队列高效地实现系统的实时性能。     

Modeling and distributed simulation of a broadband-ISDN network

A distributed approach to communication network simulation using a network of workstations configured as a loosely coupled parallel processor to model and simulate the broadband integrated services digital network (B-ISDN) is proposed. In a loosely coupled parallel processor system, a number of concurrently executable processors communicate asynchronously using explicit messages over high-speed links. Since this architecture is similar to that of B-ISDN networks, it constitutes a realistic testbed for their modeling and simulation. The authors describe an implementation of this approach on 50 Sun workstations at Brown University. Performance results, based on representative B-ISDN networks and realistic traffic models, indicate that the distributed approach is efficient and accurate     

Link layer for cooperating processes on a LAN with enhanced communication services

Most local area networks support multicast and broadcast communication facilities among the network nodes. However, their use among processes has been limited by the lack of support in the distributed operating system kernel and by the limited requirements of traditional computer network applications. Some applications (e.g. operating in a multicast environment) and distributed algorithms require advanced associations among processes, whose support by the operating system kernel can take advantage of properly designed functions provided by the communication subsystem. This problem is examined in the context of an architecture for LANs, and enhanced protocols are described that can be provided by a communication subsystem based on a LAN. Virtual network protocols provide classes of service suited to a multicast environment. Their availability at communication subsystem layers offers higher layers a common framework for the implementation of a computing system with distributed control.     

嵌入式智能家居终端通信模块的设计与实现

针对嵌入式系统的存储和处理器等硬件资源受限,设计出一个简单高效的串口通信协议和网络通信协议是开发的难点.讨论了基于ARM嵌入式Linux平台实现智能家居终端通信子系统的技术难点和关键实现技术,给出了其中串口通信协议以及网络通信协议的设计与实现,很好地解决了不同带宽和速率的通信匹配问题.该通信子系统已在Linux下调试通过,可与相应串口设备以及网络中心进行正常通信,目前已投入实际使用.     

高速网络PPP协议的实现及其安全机制的研究

研究了PPP协议在高速SDH／SONET光纤网络上的实现技术,并将其应用到高速路由器的工程实践中,实现的PPP协议软件向下和底层硬件驱动程序接口处理收到的PPP报文;向上和网络层协议IP,IPX接口将IP／IPX报文送交IP／IPX协议处理,形成发送网络层报文的控制机制。还研究了PPP协议的安全问题,包括协议的加密和认证机制,确保报文转发过程中的高速和安全性。     

一种3G网络下的LBS架构研究及实现

随着移动3G网络的普及,新型的增值业务不断涌现市场。LBS(Location Based Service)就是一类,基于移动网络高速的带宽提供人们与位置相关的服务。本文介绍了LBS的概念,提出了一种基于J2EE平台的瘦客户端/服务器的LBS体系架构,分析了业务的工作流程,并给出相关的实现技术。在实践中证明适应LBS的应用需求。     

Enabling PVM to exploit the SCTP protocol

Recently, a new general purpose transport protocol, called SCTP, has been standardized by the IETF to be used in IP based applications. The features of SCTP, compared with TCP, better support the communication requirements of parallel applications and have motivated the development of SCTP-PVM, a PVM extension, that uses SCTP for direct communications among tasks. Using the LK-SCTP open source kernel module implementation of SCTP, we have compared the performances of SCTP-PVM with the standard version of PVM that, through the PvmRouteDirect directive, uses TCP. Due to the vast difference in the maturity level of both protocols where TCP, unlike SCTP, is able to offload checksum calculations and transport segmentations in hardware, in the tests performed, we have disabled the TCP hardware transport segmentations and we have analyzed the behaviour of SCTP-PVM both with the checksum enabled and disabled. SCTP-PVM, with the checksum disabled, resulted better as the messages exchanged among tasks increase. In addition, thanks to the SCTP multi-streaming feature, SCTP-PVM achieved higher throughput in error prone networks. On the other hand, due to the high-protocol processing cost of the checksum enabled, SCTP-PVM resulted slower. Finally, the SCTP-PVM extension has been designed to easily enable existing PVM applications to use the SCTP protocol features.