imtoken官网正版下载|ethernet interface

作者： imtoken官网正版下载

2024-03-14 19:50:18

以太网（Ethernet） - 知乎

以太网（Ethernet） - 知乎首页知乎知学堂发现等你来答切换模式登录/注册以太网（Ethernet）以太网的标准拓扑结构为总线型拓扑，但目前的快速以太网（100BASE-T、1000BASE-T标准）为了减少冲突，将能提高的网络速度和使用效率最大化，使用交换机（Switch hub）来进行网络连…查看全部内容关注话题管理分享百科讨论精华视频等待回答详细内容以太网（英语：Ethernet）是一种计算机局域网技术。IEEE组织的IEEE 802.3标准制定了以太网的技术标准，它规定了包括物理层的连线、电子信号和介质访问控制的内容。以太网是目前应用最普遍的局域网技术，取代了其他局域网标准如令牌环、FDDI和ARCNET。以太网的标准拓扑结构为总线型拓扑，但目前的快速以太网（100BASE-T、1000BASE-T标准）为了减少冲突，将能提高的网络速度和使用效率最大化，使用交换机（Switch hub）来进行网络连接和组织。如此一来，以太网的拓扑结构就成了星型；但在逻辑上，以太网仍然使用总线型拓扑和CSMA/CD（Carrier Sense Multiple Access/Collision Detection，即载波多重访问/碰撞侦测）的总线技术。概述：1990年代的以太网网卡或叫NIC（Network Interface Card，以太网适配器）。这张卡可以支持基于同轴电缆的10BASE2 (BNC连接器，左)和基于双绞线的10BASE-T（RJ-45，右）。以太网实现了网络上无线电系统多个节点发送信息的想法，每个节点必须获取电缆或者信道才能传送信息，有时也叫作以太（Ether）。这个名字来源于19世纪的物理学家假设的电磁辐射媒体——光以太。每一个节点有全球唯一的48位地址也就是制造商分配给网卡的MAC地址，以保证以太网上所有节点能互相鉴别。由于以太网十分普遍，许多制造商把以太网卡直接集成进计算机主板。以太网通讯具有自相关性的特点，这对于电信通讯工程十分重要。CSMA/CD共享介质以太网：带冲突检测的载波侦听多路访问（CSMA/CD）技术规定了多台电脑共享一个通道的方法。这项技术最早出现在1960年代由夏威夷大学开发的ALOHAnet，它使用无线电波为载体。这个方法要比令牌环网或者主控制网简单。当某台电脑要发送信息时，在以下行动与状态之间进行转换：开始 - 如果线路空闲，则启动传输，否则跳转到第4步。发送 - 如果检测到冲突，继续发送数据直到达到最小回报时间（min echo receive interval）以确保所有其他转发器和终端检测到冲突，而后跳转到第4步。成功传输 - 向更高层的网络协议报告发送成功，退出传输模式。线路繁忙 - 持续等待直到线路空闲。线路空闲 - 在尚未达到最大尝试次数之前，每隔一段随机时间转到第1步重新尝试。超过最大尝试传输次数 - 向更高层的网络协议报告发送失败，退出传输模式。就像在没有主持人的座谈会中，所有的参加者都通过一个共同的介质（空气）来相互交谈。每个参加者在讲话前，都礼貌地等待别人把话讲完。如果两个客人同时开始讲话，那么他们都停下来，分别随机等待一段时间再开始讲话。这时，如果两个参加者等待的时间不同，冲突就不会出现。如果传输失败超过一次，将延迟指数增长时间后再次尝试。延迟的时间通过截断二进制指数后移（英语：Exponential_backoff）（truncated binary exponential backoff）算法来实现。最初的以太网是采用同轴电缆来连接各个设备的。电脑通过一个叫做附加单元接口（Attachment Unit Interface，AUI）的收发器连接到电缆上。一条简单网路线对于一个小型网络来说很可靠，而对于大型网络来说，某处线路的故障或某个连接器的故障，都会造成以太网某个或多个网段的不稳定。因为所有的通信信号都在共享线路上传输，即使信息只是想发给其中的一个终端（destination），却会使用广播的形式，发送给线路上的所有电脑。在正常情况下，网络接口卡会滤掉不是发送给自己的信息，接收到目标地址是自己的信息时才会向CPU发出中断请求，除非网卡处于混杂模式（Promiscuous mode）。这种“一个说，大家听”的特质是共享介质以太网在安全上的弱点，因为以太网上的一个节点可以选择是否监听线路上传输的所有信息。共享电缆也意味着共享带宽，所以在某些情况下以太网的速度可能会非常慢，比如电源故障之后，当所有的网络终端都重新启动时。以太网中继器和集线器：在以太网技术的发展中，以太网集线器（Ethernet Hub）的出现使得网络更加可靠，接线更加方便。因为信号的衰减和延时，根据不同的介质以太网段有距离限制。例如，10BASE5同轴电缆最长距离500米（1,640英尺）。最大距离可以通过以太网中继器实现，中继器可以把电缆中的信号放大再传送到下一段。中继器最多连接5个网段，但是只能有4个设备（即一个网段最多可以接4个中继器）。这可以减轻因为电缆断裂造成的问题：当一段同轴电缆断开，所有这个段上的设备就无法通讯，中继器可以保证其他网段正常工作。类似于其他的高速总线，以太网网段必须在两头以电阻器作为终端。对于同轴电缆，电缆两头的终端必须接上被称作“终端器”的50欧姆的电阻和散热器，如果不这么做，就会发生类似电缆断掉的情况：总线上的AC信号当到达终端时将被反射，而不能消散。被反射的信号将被认为是冲突，从而使通信无法继续。中继器可以将连在其上的两个网段进行电气隔离，增强和同步信号。大多数中继器都有被称作“自动隔离”的功能，可以把有太多冲突或是冲突持续时间太长的网段隔离开来，这样其他的网段不会受到损坏部分的影响。中继器在检测到冲突消失后可以恢复网段的连接。随着应用的拓展，人们逐渐发现星型的网络拓扑结构最为有效，于是设备厂商们开始研制有多个端口的中继器。多端口中继器就是众所周知的集线器（Hub）。集线器可以连接到其他的集线器或者同轴网络。第一个集线器被认为是“多端口收发器”或者叫做“fanouts”。最著名的例子是DEC的DELNI，它可以使许多台具有AUI连接器的主机共享一个收发器。集线器也导致了不使用同轴电缆的小型独立以太网网段的出现。像DEC和SynOptics这样的网络设备制造商曾经出售过用于连接许多10BASE-2细同轴线网段的集线器。非屏蔽双绞线（unshielded twisted-pair cables , UTP）最先应用在星型局域网中，之后也在10BASE-T中应用，最后取代了同轴电缆成为以太网的标准。这项改进之后，RJ45电话接口代替了AUI成为电脑和集线器的标准线路，非屏蔽3类双绞线/5类双绞线成为标准载体。集线器的应用使某条电缆或某个设备的故障不会影响到整个网络，提高了以太网的可靠性。双绞线以太网把每一个网段点对点地连起来，这样终端就可以做成一个标准的硬件，解决了以太网的终端问题。采用集线器组网的以太网尽管在物理上是星型结构，但在逻辑上仍然是总线型的，半双工的通信方式采用CSMA/CD的冲突检测方法，集线器对于减少数据包冲突的作用很小。每一个数据包都被发送到集线器的每一个端口，所以带宽和安全问题仍没有解决。集线器的总传输量受到单个连接速度的限制（10或100 Mbit/s），这还是考虑在前同步码、传输间隔、标头、档尾和封装上都是最小花费的情况。当网络负载过重时，冲突也常常会降低传输量。最坏的情况是，当许多用长电缆组成的主机传送很多非常短的帧(frame)时，可能因冲突过多导致网络的负载在仅50%左右程度就满载。为了在冲突严重降低传输量之前尽量提高网络的负载，通常会先做一些设定以避免类似情况发生。桥接和交换：尽管中继器在某些方面分隔了以太网网段，使得电缆断线的故障不会影响到整个网络，但它向所有的以太网设备转发所有的数据。这严重限制了同一个以太网网络上可以相互通信的机器数量。为了减轻这个问题，桥接方法被采用，在工作在物理层的中继器之基础上，桥接工作在数据链路层。通过网桥时，只有格式完整的数据包才能从一个网段进入另一个网段；冲突和数据包错误则都被隔离。通过记录分析网络上设备的MAC地址，网桥可以判断它们都在什么位置，这样它就不会向非目标设备所在的网段传递数据包。像生成树协议这样的控制机制可以协调多个交换机共同工作。早期的网桥要检测每一个数据包，因此当同时处理多个端口的时候，数据转发比Hub（中继器）来得慢。1989年网络公司Kalpana发明了EtherSwitch，第一台以太网交换机。以太网交换机把桥接功能用硬件实现，这样就能保证转发数据速率达到线速。大多数现代以太网用以太网交换机代替Hub。尽管布线方式和Hub以太网相同，但交换式以太网比共享介质以太网有很多明显的优势，例如更大的带宽和更好的异常结果隔离设备。交换网络典型的使用星型拓扑，虽然设备在半双工模式下运作时仍是共享介质的多节点网，但10BASE-T和以后的标准皆为全双工以太网，不再是共享介质系统。交换机启动后，一开始也和Hub一样，转发所有数据到所有端口。接下来，当它记录了每个端口的地址以后，他就只把非广播数据发送给特定的目的端口。因此线速以太网交换可以在任何端口对之间实现，所有端口对之间的通讯互不干扰。因为数据包一般只是发送到他的目的端口，所以交换式以太网上的流量要略微小于共享介质式以太网。然而，交换式以太网仍然是不安全的网络技术，因为它很容易因为ARP欺骗或者MAC满溢而瘫痪，同时网络管理员也可以利用监控功能抓取网络数据包。当只有简单设备（除Hub之外的设备）连接交换机端口时，整个网络可能处于全双工模式。如果一个网段只有2个设备，那么冲突探测也不需要了，两个设备可以随时收发数据。这时总带宽是链路的2倍，虽然双方的带宽相同，但没有发生冲突就意味着几乎能利用到100%的带宽。交换机端口和所连接的设备必须使用相同的双工设置。多数100BASE-TX和1000BASE-T设备支持自动协商特性，即这些设备通过信号来协调要使用的速率和双工设置。然而，如果自动协商功能被关闭或者设备不支持，则双工设置必须通过自动检测进行设置或在交换机端口和设备上都进行手工设置以避免双工错配——这是以太网问题的一种常见原因（设备被设置为半双工会报告迟发冲突，而设备被设为全双工则会报告runt）。许多较低层级的交换机没有手工进行速率和双工设置的能力，因此端口总是会尝试进行自动协商。当启用了自动协商但不成功时（例如其他设备不支持），自动协商会将端口设置为半双工。速率是可以自动感测的，因此将一个10BASE-T设备连接到一个启用了自动协商的10/100交换端口上时将可以成功地创建一个半双工的10BASE-T连接。但是将一个配置为全双工100Mb工作的设备连接到一个配置为自动协商的交换端口时（反之亦然）则会导致双工错配。即使电缆两端都设置成自动速率和双工模式协商，错误猜测还是经常发生而退到10Mbps模式。因此，如果性能差于预期，应该查看一下是否有计算机设置成10Mbps模式了，如果已知另一端配置为100Mbit，则可以手动强制设置成正确模式。.当两个节点试图用超过电缆最高支持数据速率（例如在3类线上使用100Mbps或者3类/5类线使用1000Mbps）通信时就会发生问题。不像ADSL或者传统的拨号Modem通过详细的方法检测链路的最高支持数据速率，以太网节点只是简单的选择两端支持的最高速率而不管中间线路，因此如果速率过高就会导致链路失效。解决方案为强制通讯端降低到电缆支持的速率。以太网类型：除了以上提到的不同帧类型以外，各类以太网的差别仅在速率和配线。因此，同样的网络协议栈软件可以在大多数以太网上执行。以下的章节简要综述了不同的正式以太网类型。除了这些正式的标准以外，许多厂商因为一些特殊的原因，例如为了支持更长距离的光纤传输，而制定了一些专用的标准。很多以太网卡和交换设备都支持多速率，设备之间通过自动协商设置最佳的连接速度和双工方式。如果协商失败，多速率设备就会探测另一方使用的速率但是默认为半双工方式。10/100以太网端口支持10BASE-T和100BASE-TX。10/100/1000支持10BASE-T、100BASE-TX和1000BASE-T。部分以太网类型局域网（英语：Local Area Network，简称LAN）是连接住宅、学校、实验室、大学校园或办公大楼等有限区域内计算机的计算机网络。相比之下，广域网（WAN）不仅覆盖较大的地理距离，而且还通常涉及固接专线和对于互联网的链接。相比来说互联网则更为广阔，是连接全球商业和个人电脑的系统。在历经使用了链式局域网（英语：ARCNET）、令牌环与AppleTalk技术后，以太网和Wi-Fi（无线网络连接）是现今局域网最常用的两项技术。机理：局域网（Local Area Network, LAN），又称内网。指覆盖局部区域（如办公室或楼层）的计算机网络。按照网络覆盖的区域（距离）不同，其他的网络类型还包括个人网、城域网、广域网等。早期的局域网网络技术都是各不同厂家所专有，互不兼容。后来，电机电子工程师学会推动了局域网技术的标准化，由此产生了IEEE 802系列标准。这使得在建设局域网时可以选用不同厂家的设备，并能保证其兼容性。这一系列标准覆盖了双绞线、同轴电缆、光纤和无线等多种传输介质和组网方式，并包括网络测试和管理的内容。随着新技术的不断出现，这一系列标准仍在不断的更新变化之中。以太网（IEEE 802.3标准）是最常用的局域网组网方式。以太网使用双绞线作为传输介质。在没有中继的情况下，最远可以覆盖200米的范围。最普及的以太网类型数据传输速率为100Mb/s，更新的标准则支持1000Mb/s和10Gb/s的速率。其他主要的局域网类型有令牌环和FDDI（光纤分布数字接口，IEEE 802.8）。令牌环网络采用同轴电缆作为传输介质，具有更好的抗干扰性；但是网络结构不能很容易的改变。FDDI采用光纤传输，网络带宽大，适于用作连接多个局域网的骨干网。近两年来，随着802.11标准的制定，无线局域网的应用大为普及。这一标准采用2.4GHz 和5.8GHz 的频段，数据传输速度最高可以达到300Mbps和866Mbps。局域网标准定义了传输介质、编码和介质访问等底层（一二层）功能。要使数据通过复杂的网络结构传输到达目的地，还需要具有寻址、路由和流量控制等功能的网络协议的支持。TCP/IP（传输控制协议/互联网络协议）是最普遍使用的局域网网络协议。它也是互联网所使用的网络协议。其他常用的局域网协议包括，IPX、AppleTalk等。在无线 LAN 中，用户可以在覆盖区域内不受限制地移动。无线网络因其易于安装而在住宅和小型企业中流行起来。大多数无线局域网都使用 Wi-Fi，因为它内置于智能手机、平板电脑和笔记本电脑中。客人通常可以通过热点服务上网。网络拨接互联网（英语：Internet）是指20世纪末期兴起电脑网络与电脑网络之间所串连成的庞大网络系统。这些网络以一些标准的网络协议相连。它是由从地方到全球范围内几百万个私人、学术界、企业和政府的网络所构成，通过电子、无线和光纤网络技术等等一系列广泛的技术联系在一起。互联网承载范围广泛的信息资源和服务，比方说相互关系的超文本文件，还有万维网（WWW）的应用、电子邮件、通话，以及文件共享服务。互联网的起源可以追溯到1960年代美国联邦政府委托进行的一项研究，目的是创建容错与电脑网络的通信。互联网的前身ARPANET最初在1980年代作为区域学术和军事网络连接的骨干。1980年代，NSFNET（英语：NSFNET）成为新的骨干而得到资助，以及其他商业化扩展得到了私人资助，这导致了全世界网络技术的快速发展，以及许多不同网络的合并结成更大的网络。到1990年代初，商业网络和企业之间的连接标志着向现代互联网的过渡。尽管互联网在1980年代只被学术界广泛使用，但商业化的服务和技术，令其极快的融入了现代每个人的生活。互联网并不等同万维网，互联网是指凡是能彼此通信的设备组成的网络就叫互联网，指利用TCP/IP通讯协定所创建的各种网络，是国际上最大的互联网，也称“国际互联网”。万维网是一个由许多互相链接的超文本组成的系统，通过互联网访问。在此定义下，万维网是互联网的一项服务。不过多数民众并不区分两者，常常混用。连接技术：任何需要使用互联网的计算机必须通过某种方式与互联网进行连接。互联网接入技术的发展非常迅速，带宽由最初的14.4Kbps发展到目前的100Mbps甚至1Gbps带宽，接入方式也由过去单一的电话拨号方式，发展成现在多样的有线和无线接入方式，接入终端也开始朝向移动设备发展。并且更新更快的接入方式仍在继续地被研究和开发。架构：最顶层的是一些应用层协议，这些协议定义了一些用于通用应用的数据报结构，包括FTP及HTTP等。中间层是UDP协议和TCP协议，它们用于控制数据流的传输。UDP是一种不可靠的数据流传输协议，仅为网络层和应用层之间提供简单的接口。而TCP协议则具有高的可靠性，通过为数据报加入额外信息，并提供重发机制，它能够保证数据不丢包、没有冗余包以及保证数据包的顺序。对于一些需要高可靠性的应用，可以选择TCP协议；而相反，对于性能优先考虑的应用如流媒体等，则可以选择UDP协议。最底层的是互联网协议，是用于报文交换网络的一种面向数据的协议，这一协议定义了数据包在网际传送时的格式。目前使用最多的是IPv4版本，这一版本中用32位定义IP地址，尽管地址总数达到43亿，但是仍然不能满足现今全球网络飞速发展的需求，因此IPv6版本应运而生。在IPv6版本中，IP地址共有128位，“几乎可以为地球上每一粒沙子分配一个IPv6地址”。IPv6目前并没有普及，许多互联网服务提供商并不支持IPv6协议的连接。但是，可以预见，将来在IPv6的帮助下，任何家用电器都有可能连入互联网。互联网承载着众多应用程序和服务，包括万维网、社交媒体、电子邮件、移动应用程序、多人电子游戏、互联网通话、文件分享和流媒体服务等。提供这些服务的大多数服务器托管于数据中心，并且通过高性能的内容分发网络访问。万维网（英语：World Wide Web）亦作WWW、Web、全球广域网，是一个透过互联网访问的，由许多互相链接的超文本组成的信息系统。英国科学家蒂姆·伯纳斯-李于1989年发明了万维网。1990年他在瑞士CERN的工作期间编写了第一个网页浏览器。网页浏览器于1991年1月向其他研究机构发行，并于同年8月向公众开放。罗伯特·卡里奥设计的Web图标万维网是信息时代发展的核心，也是数十亿人在互联网上进行交互的主要工具。网页主要是文本文件格式化和超文本置标语言（HTML）。除了格式化文字之外，网页还可能包含图片、视频、声音和软件组件，这些组件会在用户的网页浏览器中呈现为多媒体内容的连贯页面。万维网并不等同互联网，万维网只是互联网所能提供的服务其中之一，是靠着互联网运行的一项服务。参考文献： Wendell Odom. CCENT/CCNA ICND1 100-105 Official Cert Guide. Cisco Press. 2016: 43页. ISBN 978-1-58720-580-4.Internet协议观念与实现ISBN 9577177069Internet协议观念与实现ISBN 9577177069IEEE 802.3-2008 Section 3 Table 38-2 p.109IEEE 802.3-2008 Section 3 Table 38-6 p.111网络化生存，乔岗，中国城市出版社，1997年，ISBN 978-7-5074-0930-7Richard J. Smith, Mark Gibbs, Paul McFedries 著，毛伟、张文涛译，Internet漫游指南，人民邮电出版社，1998年. ISBN 978-7-115-06663-3世界是平的，汤马斯·佛里曼著，2005年出版. ISBN 978-986-80180-9-9内容采用CC BY-SA 3.0授权。浏览量2690 万讨论量9728 帮助中心知乎隐私保护指引申请开通机构号联系我们举报中心涉未成年举报网络谣言举报涉企侵权举报更多关于知乎下载知乎知乎招聘知乎指南知乎协议更多京 ICP 证 110745 号 · 京 ICP 备 13052560 号 - 1 · 京公网安备 11010802020088 号 · 京网文[2022]2674-081 号 · 药品医疗器械网络信息服务备案（京）网药械信息备字（2022）第00334号 · 广播电视节目制作经营许可证:（京）字第06591号 · 服务热线：400-919-0001 · Investor Relations · © 2024 知乎北京智者天下科技有限公司版权所有 · 违法和不良信息举报：010-82716601 · 举报邮箱：jubao@zhihu.

Ethernet接口和Interface：深入了解接口的结构和功能 - 技象科技

跳至内容

技象科技有限公司官方网站

技象天猫旗舰店技象天猫旗舰店

技象京东旗舰店技象京东旗舰店

IoT演示平台IoT演示平台

首页

解决方案

菜单切换

TPUNB®

城域物联网

菜单切换

智慧农业

环境监测

智慧停车

局域物联网

菜单切换

智慧楼宇

智慧养老

智慧照明

智慧能源

工业智控物联网

菜单切换

动环监控

终端自组网

菜单切换

自组网

产品终端

菜单切换

芯片产品

模组产品

网关产品

应用产品

云平台

关于我们

菜单切换

荣誉资质

合作伙伴

公司一览

招聘岗位

新闻资讯

菜单切换

公司新闻

下载专区

联系我们

首页

TPUNB®

物联网解决方案

菜单切换

城域物联网

局域物联网

工业智控物联网

终端自组网

物联网产品终端

菜单切换

芯片产品

模组产品

网关产品

物联网云平台

关于我们

菜单切换

荣誉资质

合作伙伴

公司新闻

联系我们

Main Menu

首页

TPUNB®

物联网解决方案

菜单切换

城域物联网

局域物联网

工业智控物联网

终端自组网

物联网产品终端

菜单切换

芯片产品

模组产品

网关产品

物联网云平台

关于我们

菜单切换

荣誉资质

合作伙伴

公司新闻

联系我们

技象科技首页 / 行业百科 / Ethernet接口和Interface：深入了解接口的结构和功能

Ethernet接口和Interface：深入了解接口的结构和功能作者：

技象物联网

/ 行业百科 / 电子技术 / 2023年10月21日 01:28:02 2023年10月21日 01:28:02

从最基本的概念开始，以接口和接口（interface）作为核心关键词，本文将深入介绍如何使用以太网接口（Ethernet interface），以及它们的结构和功能。本文将帮助您深入了解以太网接口，以及它们如何与网络系统相互作用。

什么是以太网接口？

以太网接口是一种网络接口，它可以连接计算机和其他网络设备，以便进行数据传输。它使用网络协议，如TCP/IP，以及物理媒介，如有线电缆，来连接设备。以太网接口可以是内置的，也可以是外接的。

以太网接口的结构

以太网接口通常由两个主要部件组成：以太网接口卡（Ethernet interface card）和连接器（connector）。以太网接口卡是一种硬件，它可以安装在计算机的硬件插槽中，用于连接设备和网络。连接器是一种外部设备，它可以将以太网接口卡连接到网络设备，如网络路由器，交换机，或以太网线。

以太网接口的功能

以太网接口的主要功能是提供网络连接，以便设备之间可以进行数据传输。它可以实现网络的物理层，以及网络的数据链路层功能。它还可以实现网络层，传输层和应用层功能，以及其他网络协议，如TCP/IP协议。

接口的优缺点

以太网接口有很多优点，它可以提供高速数据传输，支持多种网络协议，并可以通过网络设备进行连接。另外，它还可以支持多种类型的网络连接，如有线连接和无线连接。

然而，以太网接口也有一些缺点。它的数据传输速度受限于物理媒介的最大传输速率，并且它可能会受到干扰，从而降低数据传输的效率。此外，它也可能会受到病毒和恶意软件的攻击，从而破坏网络的安全性。

总结

本文介绍了以太网接口的结构和功能，以及它的优缺点。以太网接口可以提供高速数据传输，支持多种网络协议，并可以通过网络设备进行连接。然而，它也有一些缺点，如受限于物理媒介的最大传输速率，以及可能受到病毒和恶意软件的攻击。

以太网知识01 Media Independent Interface （MII) 媒体独立接口 - 知乎

以太网知识01 Media Independent Interface （MII) 媒体独立接口 - 知乎首发于塞米啃以太网Ethernet切换模式写文章登录/注册以太网知识01 Media Independent Interface （MII) 媒体独立接口塞米本文非原创，摘自：Media Independent InterfaceMedia Independent Interface ( MII )，媒体独立接口，起初是定义100M以太网（Fast Ethernet）的 MAC 层与 PHY 芯片之间的传输标准（802.3u）。介质独立的意思是指，MAC与PHY之间的通信不受具体传输介质（双绞线或光纤等）的影响，任何MAC和PHY都可以通过MII接口互连。MAC与PHY之间的MII连接可以是可插拔的连接器，或者是同一块PCB上MAC与PHY之间的走线。MDIO 是MII接口的一部分，用于在MAC和PHY之间传递配置信息。在系统上电瞬间，PHY芯片通过管脚的电平状态确定原始设置，进而通过MDIO更改配置。最初MII定义数据 4 bit 发送+ 4 bit 接收，每位数据速率 25MHz ，总数据速率 4*25Mbps=100Mbps 。其它 MII 标准的变种，包括 RMII，GMII，RGMII，XGMII，SGMII，基本上都是定位于更高速率或者更少的信号数的目标，图1表示在以太网通信层次模型中MII接口的位置。图1. IEEE 802.3 标准（100Mbps +）MII: Media Independent InterfaceMII接口信号包括三类，分别为：发送端信号：TXCLK, TXD[0-3], TXEN, TXER接收端信号：RXCLK, RXD[0-3], RXDV, RXER, CRS, COL配置信号：MDIO, MDC信号方向如下图所示，其中 TXER 为选配。MII 共计 18 根信号线，只有 MDIO/MDC 信号可以在不同PHY间级联。假定系统中有 8 个PHY，则MII信号总数为 8*16 + 2 = 130 根！为减少信号数，RMII接口应运而生。图2. MII InterfaceRMII: Reduced Media Independent Interface相比于MII接口，RMII有以下四处变化：TXCLK 和 RXCLK 两个时钟信号，合并为一个时钟 REFCLK时钟速率由 25MHz 上升到 50MHz，单向数据由 4 bits 变为 2 bitsCRS 和 RXDV 合并为一个信号 CRSDV取消了 COL 信号RMII信号如下图所示。RMII只要 9 根信号线，相比于MII的 18 根信号可谓有不少的删减，在同一个系统中的多个设备可以共享 MDIO, MDC 和 REFCLK 信号线。图3. RMII InterfaceGMII: Gigabit Media Independent InterfaceGMII接口的数据速率可达 1000Mbps，其时钟频率为 125MHz ，单向数据位宽 8 bits。GMII向下兼容MII，可以像MII一样工作在 100Mbps 和 10Mbps 的数据速率。GMII接口信号包括三类，分别为：发送端信号：GTXCLK, TXCLK, TXD[0-7], TXEN, TXER接收端信号：RXCLK, RXD[0-7], RXDV, RXER, CRS, COL配置信号：MDIO, MDC发送端包括两个时钟信号 GTXCLK 和 TXCLK，当设备工作于 1000Mbps 模式时，TXD, TXEN, TXER 是与 GTXCLK （125MHz）同步的，而在 10/100Mbps 工作模式时，以上数据信号是同步于由PHY提供的TXCLK 的，其中 100Mbps 时是 25MHz，10Mbps 时是 2.5MHz。接收端时钟只有一个时钟信号 RXCLK，它是从接收数据中恢复的时钟。图4. GMII InterfaceRGMII: Reduced Gigabit Media Independent InterfaceRGMII相比于GMII减小将近一半的管脚数（24 → 12），通过以下两种方式：1000Mbps模式下，在时钟的上/下边沿均采样数据取消不重要的如 CRS, COL 等信号在RGMII接口中 MAC 在 TXC 上一直提供时钟信号，而不像在GMII接口中那样，10/100Mbps 模式下时钟是由 PHY 提供（TXCLK），而 1000Mbps 模式下时钟是由 MAC 提供（GTXCLK）。在RGMII中应用到源同步时钟，即数据与时钟信号是同步的。这要求在PCB设计中，要对时钟信号额外增加 1.5~2 ns 的延迟以保证接收端的建立/保持时间满足要求。在 RGMII v2.0 规范中有定义MAC/PHY内部延迟（RGMII-ID），由此避免PCB设计中再要增加这个延迟。在RGMII接口中：1000Mbps 模式，数据在时钟的上/下边沿均采样10/100Mbps 模式，数据仅在时钟上升沿采样RXCTL 和 TXCLT 为复用的传输控制信号。RXCTL 在时钟的上升沿代表 RXDV，在时钟的下降沿代表（RXDV xor RXER）；TXCTL 在时钟的上升沿代表 TXEN，在时钟的下降沿代表（TXEN xor TXER）。RGMII v1.3 采用 2.5V CMOS 电平，RGMII v2 采用 1.5V HSTL 电平。图5. RGMII Interface以下为解释为什么需要对时钟添加delay (参考于https://ethernetfmc.com/docs/user-guide/rgmii-timing/)The RGMII standard specifies clock and data signals to be output with no skew, ie. the clock edges are aligned with the data edges. This is not ideal for the receiver’s sampling circuit, but it greatly simplifies the transmitter circuit. RGMII Interface without clock skew RGMII clock and data signals as they must be presented to the receiving circuit for optimal sampling. The clock skew has been added by the PCB trace or the receiving device. RGMII Interface with clock skew Adding the clock skew ：In an FPGA based system, there are three stages where the required skew (ie. delay) can be added to the TX and RX clock signals. The first stage is in the FPGA, the second stage is on the PCB traces (ie. with longer clock traces than the data traces) and the third stage is in the PHY. In an optimal RGMII interface, the skew is added at only one stage in the TX and RX path, and the other two delay stages are disabled or not implemented.Clock skew stages in the RGMII interface The skew of the TX and RX clocks can be managed independently, it does not have to be implemented at the same stage on each path, but it must be implemented somewhere on each path. So it is critical to understand each of the delay stages in your target system in order to ensure that the clocks in your RGMII interface are properly skewed. We will now discuss each skew stage and the various ways to enable or disable them.SGMII: Serial Gigabit Media Independent InterfaceSGMII发送和接收时钟频率均为 625MHz，采用 DDR 模式，因此数据速率为1.25Gbps。SGMII相比于GMII，功耗更低，采用 SerDes 接口后管脚数更少。SGMII发送和接受数据各 1 对差分信号（LVDS），另外还有 1 对差分时钟，共 6 根线。对于 MAC/PHY 中包括时钟恢复电路（CDR, Clock and Data Recovery circuitry ）的系统，TXCLK 可以省略，SGMII接口只需要 4 根线，相比于GMII（ 24 根）和RGMII（ 12 根）信号线大大减少！TX/RX在数据发送端必须同时产生时钟，而接收端的时钟是可选的，因为可以通过 CDR 恢复时钟。在 10/100Mbps 工作模式下，数据分别重复发送 100/10 次，因此时钟always是 625MHz。图6. SGMII Interface图7. 4-Wire/6-Wire SGMIIXGMII: 10 Gigabit Media Independent InterfaceXGMII 是用于10G以太网的MAC与PHY设备间通信的接口标准，它包括 32 bits 的数据通道（RXD & TXD），两组 4 bits 的控制通道（RXC & TXC）和两组时钟（收/发），时钟频率 156.25 MHz ，工作在 DDR 模式。图8表示XGMII接口的连接示意图，注意 RXD/TXD 信号上的 36 表示 32 bits 数据 + 4 bits 控制信号，其中每 8 bits 数据称为 1 个Lane，共用 1 路控制信号。10 Gbps = 156.25 MHz * 32 bits * 2XGMII信号数目（74 根）较多，通常用于芯片内的连接，不适合作为芯片间通信的接口，因此协议定义XGXS（XGMII eXtender Sublayer）子层以缩减信号数目，简化硬件设计。XGXS 子层主要完成 8b/10b 编码和不同Lane之间的去偏斜等功能。如图8所示，在信号链的两端，MAC和PHY 都包括XGXS子层，XAUI 是 XGXS 之间通信的接口。XAUI 接口包括4组发送差分对和4 组接收差分对，共 16 根信号。每组差分对（Lane）的数据速率为 3.125 Gbps，因此总的数据速率为 4 * 3.125 Gbps = 12.5 Gbps，考虑到8b/10b的效率为80%，因此实际数据速率为 12.5Gbps * 80% = 10 Gbps。图8. XGMII InterfaceAppendixXFI/XFPXFI 是10G以太网 PMA（Physical Medium Attachment）和 PMD（Physical Medium Dependent）之间的接口标准，它只有两对差分线（收/发），共 4 根线，如图9所示。XFI 接口速度达到 10.3125 Gbps，采用 64B/66B 编码，在XAUI与XFI之间使用到 SerDes 以减小信号数。图9. 10GbE Layer &amp;amp;amp;amp; InterfaceXFP（10 Gigabit Small Form Factor Pluggable）是指应用XFI接口的光模块，应用于10G以太网的光传输。XFP光模块的尺寸略大于 SFP 和 SFP+ 光模块，三种光模块的详细对比见链接文章，在此不再赘述。MDIO如上文所述，MDIO用于上层（MAC）配置底层（PHY）的参数，它包括时钟信号 MDC 和数据信号 MDIO 。如果系统中不止一个PHY，在使用同一组MDIO信号以级联方式配置PHY时，需要通过PHY芯片管脚的 Strap 来寻址不同芯片。PHY芯片的物理地址Strap管脚一般与 RXD 管脚复用。在MDIO规范中定义PHY地址为 5 bit，即同一组MDIO最多可配置 2^5 = 32 个PHY。图10表示MDIO配置的时序图。注意，这里提到的PHY芯片Strap的物理地址仅与MDIO的配置过程寻址有关，和通常意义上的 MAC 地址没有任何关系。图10. MIIM Timing接口速率计算摘自知乎博主碎碎思的媒体独立接口（MII，Meida Independent Interface）参考资料Media-independent interface -WikipediaIEEE 802.3 Ethernet PresentationEthernet Media Access Controller (EMAC)/ Management Data Input/Output (MDIO) Module -TIDP83867E/IS/CS Robust, High Immunity, Small Form Factor 10/100/1000 Ethernet Physical Layer Transceiver -TIXGMII Update -IEEE 802XAUI/XGXS Proposal -IEEE 802Overview of 10G Ethernet Family -IEEE 802编辑于 2022-10-27 14:53以太网（Ethernet）芯片（集成电路）赞同 232 条评论分享喜欢收藏申请转载文章被以下专栏收录塞米啃以太网Ethernet学习Ethernet他说《SoC设计从入门到

高性能网络 — SmartNIC、DPU 设备演进与运行原理 - 知乎

高性能网络 — SmartNIC、DPU 设备演进与运行原理 - 知乎首发于OpenStack IaaS切换模式写文章登录/注册高性能网络 — SmartNIC、DPU 设备演进与运行原理云物互联云计算、云原生、边缘计算、5G 网络、云网融合。Basic Ethernet NIC 设备组成Physical Interface（物理链路连接器）Physical Interface（物理链路连接器）负责将双绞线网口（电口）或光模块（光口）或连接到网卡上。一个 Physical Interface 通常具有多个 Ethernet Ports。电口：一般为 SFP、QSFP 等，例如 RJ45（Registered Jack，注册的插座）实现了网卡和网线的连接。光口：一般为光纤连接器。数据中心常见的网卡速率和接口类型：1GbE 千兆网卡：通常使用基于 RJ45 接口的电缆，如 Cat5e、Cat6 等。10GbE 万兆网卡：通常使用基于 SFP+ 接口（10GbE）的光缆。25GbE 网卡：通常使用基于 SFP28 接口（25GbE/10GbE）的光缆。100GbE 网卡：通常使用基于 QSFP28 接口（100GbE/40GbE）的光缆。PHY（物理层调制解调器）PHY（Physical Layer Modem，物理层调制解调器），TCP/IP 物理层实现，负责将计算机产生的数字信号转换成可以在物理介质上传输的模拟信号。例如：CSMA/CD、模数转换、编解码、串并转换等。硬件层面表现为 PHY 芯片，定义了数据发送和接收所需要的电与光信号、线路状态、时钟基准、数据编码、连接速度，双工能力等。并向数据链路层提供了 IEEE MII/GigaMII（Media Independed Interfade，介质独立接口）标准接口，用于连接 MAC 和 PHY 传输控制面和数据面的数据。此外还具有 PCS（Physical Coding Sublayer，物理编码）、PMA（Physical Medium Attachment，物理介质附加）、PMD（物理介质相关）、MDIO（Management Data Input/Output）等子层。软件层面执行 CSMA/CD（Carrier Sense Multiple Access/Collision Detection，载波监听多路访问/冲突检测）协议。CSMA/CD 协议具有 “冲突检测“ 和 “载波监听“ 功能，能够检测到网络上是否有数据在传送，如果有数据在传送中就等待，一旦检测到网络空闲，再等待一个随机时间后将送数据出去。PHY 和 Physical Interface 之间还具有一个 Transformer（变压器），具有提高传输距离、波形修复、电气隔离、抗干扰、防雷等作用。变压器使网卡的芯片组与外部隔离，增强了抗干扰能力，也提供了重要的保护作用。MAC（介质访问控制器）MAC（Media Access Control，介质访问控制器），TCP/IP 数据链路层实现，负责控制与物理层进行连接的的物理介质。硬件层面表现为一块 MAC 芯片，每个 Ethernet Ports 都具有一个全球唯一的 MAC 地址，就是 MAC 芯片的地址。另外， MAC 还具有对应的 Rx/Tx Queues 用于缓存接收/发送的 Frames。软件层面遵守 IEEE 802.3 Ethernet 协议，完成物理介质 Bit stream 和操作系统 Ethernet Frames 之间的转换，以及完成 Frames CRC 校验。还会控制 PHY 具体执行 CSMA/CD 协议。先进的 MAC 芯片还会提供数据链路层面的 Packet Filtering 功能。例如：L2 Filtering、VLAN Filtering、Host Filtering 等。Ethernet Controller（以太网控制器）Ethernet Controller（以太网控制器）是网卡的核心部件，相当于计算机的主机（CPU + Memory），提供了主要的控制面功能，并通过 Driver（驱动程序）与 Linux 操作系统进行交互。具有以下功能：设备初始化、启动、停止、重启等操作接口。DMA Control；Flow Control；定时与控制电路（Programmable Logic Array）；中断处理；等。Bus Interface（总线接口）Bus Interface（总线接口）是网卡和计算机主板之间的连接器，实现 CPU 和 NIC 之间的交互。主要包括 DMA Interface 和 PCIe Interface 这两大类型。DMA InterfaceDMA Controller 的功能：向 CPU 发出 HOLD（保持）信号，提出 Bus（总线）接管请求。当 CPU 发出允许接管信号后，负责对 Bus 的控制，进入 DMA I/O 模式。通过对 Main Memory 进行寻址以及修改地址指针，实现对 Memory 的读写操作。向 CPU 发出 DMA 结束信号，CPU 恢复正常工作模式。DMA Interface 的信号类型：DREQ（外设请求信号）：I/O 外设向 DMA Controller 发起请求。DACK（DMA 响应信号）：DMA Controller 向 I/O 外设的响应信号HRQ/HOLD（DMA 请求信号）：DMA Controller 向 CPU 发出，要求接管 Bus。HLDA（CPU 响应信号）：CPU 响应允许 DMA Controller 接管 Bus。PCIe InterfacePCI-E 外设接口标准历经了 PCI，PCI-X 和 PCI-E 三个阶段，发展了近 30 年的时光。PCI-E 的最终目的是为了替代并统一 PCI、磁盘、网卡、显卡等外设的接口，解决计算机内部数据传输的瓶颈问题。最新的 PCI-E 标准是 Intel 提出的新一代的外设总线接口，采用了目前业内流行的点对点串行连接。比起 PCI 以及更早期的计算机总线的共享并行架构而言，每个 PCI-E 设备都有自己的专用连接，不需要向整个总线请求带宽，而且可以把数据传输率提高到一个很高的频率，达到 PCI 所不能提供的高带宽。PCI-E（Peripheral Component Interconnect Express，PCI Express，PCI-E，快速外设组件互连）设备PCI-E 设备的接口类型PCI-E 设备的体积类型LP（半高）FH（全高）HL（半长）FL（全长）SW（单宽）DW（双宽）PCI / PCI-E 插槽：PCI-E 支持对 PCI / PCI-X 的软件兼容，但主板上的接口插槽却不兼容，因为 PCI-E 是串行接口，针数会更少，插槽会更短，PCI-E 插槽的长度跟信道的数目有关。SmartNIC 设备组成Basic NIC 是一个 PCIe 设备，它仅实现了与以太网的连接，即：实现了 L1-L2 层的逻辑，负责 L2 层数据帧的封装/解封装，以及 L1 层电气信号的相应处理；而由 Host CPU 则负责处理网络协议栈中更高层的逻辑。即：CPU 按照 L3-L7 层的逻辑，负责数据包的封装/解封装等工作；随着网络速率从 1Gbps、10Gbps、25Gbps、100Gbps 的发展，网络速率需求和 CPU 计算能力的差距持续扩大，激发网络侧专用计算的需求。也伴随着 NIC 片上芯片计算能力的发展，工业内陆续提出了各种各样的 Hardware Offload 方案，将各种 CPU 的 workload 卸载（Offload）到外设扩展卡上进行处理。最初的方式就是增加 NIC 的 workload 处理能力，例如：现代 NIC 普遍实现了部分 L3-L4 层逻辑的 offload，例如：校验和计算、传输层分片重组等，以此来减轻 Host CPU 的处理负担。甚至有些专用 NIC，例如：RDMA 网卡，还会将整个 L4 层的处理都 offload 到了硬件上。TSO（TCP Segmentation Offload）GSO（Generic Segmentation Offload）LRO（Large Receive Offload）GRO （Generic Receive Offload）随着 NIC 具备了一定计算能力，也称之为智能，所以这类 NIC 被分类为 SmartNIC。SmartNIC 通过在 NIC 上引入 ASIC、FPGA 或 SoC 芯片来加速（处理）某些特定的流量，从而加强网络的可靠性，降低网络延迟，提升网络性能。简而言之，SmartNIC 就是通过从 Host CPU 上 Offload（卸载）工作负载到网卡硬件，以此提高 Host CPU 的处理性能。其中的 “工作负载” 不仅仅是 Networking，还可以是 Storage、Security 等等。以 FPGA 来实现 Smart NIC 举例，了解到底有什么 workload 是可以 Offload 到 Smart NIC 上进行处理的。并且，使用 FPGA 可以根据需要轻松添加、或删除这些功能。示例 1 到 13 说明了可以添加到 Basic NIC 的处理元素，以创建功能更加强大的 Smart NIC。1. Basic NIC采用多个 Ethernet MAC 芯片和一个用于与 Host CPU 连接的 PCIe Interface。Host CPU 必须主动处理所有的 Ethernet Packets。2. 添加 DMA Engine 功能添加 DMA Controller 和 DMA Interface，将 NIC Memory 直接映射到 Main Memory ZONE_DMA。Host CPU 可以直接从 Main Memory 读取 Packets，而不再需要从 NIC Memory 中进行 Copy，从而减少了 Host CPU 的工作负载。3. 添加 Filter Engine 功能Packets Filter 模块提供 L2 Filtering、VLAN Filtering、Host Filtering 等功能，可以进一步减少了 Host CPU 的工作负载。4. 添加外部 DRAM 到 Filter Engine为 Packets Filter 添加用于存储 Filter Rules 的 DRMA 存储器，进一步增强 Packets Filter 的功能和灵活性。5. 添加 L2/L3 Offload Engine 功能添加 L2 Switching 和 L3 Routing 功能，卸载数据面转发功能，进一步减少 Host CPU 的工作负载。6. 添加 Tunnel Offload Engine 功能将 VxLAN、GRE、MPLSoUDP/GRE 等 Host Tunnel 数据面功能卸载到 SmartNIC，进一步减轻 Host CPU 进行隧道封装/解封装的工作负载。7. 添加 Deep Buffering 外部存储添加 Deep Buffer（深度缓冲）专用存储器，用于构建支撑 L2/L3/Tunnel Offloading 的差异化 Buffer Rings。8. 添加 Flows Engine 功能针对 vSwitch / vRouter 虚拟网元的 Fast Path 提供 Traffic Flows 模块，另外配置一个 DRAM 存储器，可以处理数百万个 Flow Table Entries。9. 添加 TCP Offload Engine 功能卸载全部或部分 TCP 协议功能，减轻 TCP 服务器的 Host CPU 工作负载。10. 添加 Security Offload Engine 功能卸载 TLS 此类加密/解密功能，针对相应的 Traffic Flow 可以选择开启/关闭 TLS 加速。11. 添加 QoS Engine 功能添加 QoS Engine 功能，卸载 TC 等流量控制模块，可以实现 Multi-Queues 和 QoS 调度功能。12. 添加一个 Programmable Engine 功能添加 P4 RMT（Reconfigurable Match Tables，可重配置 Match-Action 表）此类 Programmable Engine，提供一定的可编程 Pipeline 能力。13. 添加一个或多个 ASIC 板载处理器添加用于管理面和控制面的 CPU 处理器，提供完整的软件可编程性。DPU 设备组成失效的摩尔定律从前半导体技术高速发展，处理器芯片的性能每 18 个月就能翻倍，算力提升和软件需求处在一个供需平衡的状态。但近几年半导体技术的发展已经逼近物理极限，集成电路越来越复杂，单核芯片的工艺提升目前止步于 3nm。一个可行的方式是通过多核堆叠来提升算力，但是随着核数的增加，单位算力能耗比也会显著增加，而且堆叠无法实现算力的线性增长。例如：将 128 核堆叠至 256 核，但总算力水平也无法提升到 1.2 倍。计算单元的工艺演进已经逼近基线，每 18 个月翻一番的摩尔定律已经宣告失效。2016 年 3 月 24 日，英特尔宣布放弃 “Tick-Tock” 研发模式，未来研发周期将从两年周期向三年期转变。一边是失效的摩尔定律，但另一边却正在发生着 “数据摩尔定律” —— IDC 数据显示，全球数据量在过去 10 年的年均复合增长率接近 50%，并进一步预测每四个月对于算力的需求就会翻一倍。可见，算力的供需关系已然失衡。单从网络的角度出发，可以使用 RBP（Ratio of Bandwidth and Performance growth rate，带宽性能增速比）描述这一供需关系。RBP=BW GR/Perf. GR（网络带宽增速 / CPU 性能增速）。2010 年前，网络的带宽年化增长大约是 30%，到 2015 年增长到 35%，然后在近年达到 45%。相对应的，CPU 的性能增长从 10 年前的 23%，下降到 12%，并在近年直接降低到 3.5%。在这三个时间段内，RBP 指标从 RBP～1 附近（I/O 压力尚未显现出来），上升到 RBP～3，并在近年超过了 RBP～10。CPU 算力增速几乎已经无法应对网络带宽的增速。沉重的数据中心税在 10GbE 和 25GbE 场景中，传统 NIC 的表现让人可以接受，DPDK 等高性能数据面转发技术只是会占用部分 CPU 资源。但在逐渐普及的 40GbE 和 100GbE 场景中，CPU 就会出现阻塞。根据 Fungible 和 AWS 的统计，在大型数据中心中，网络流量的处理占到了计算的 30% 左右，即：CPU 30% 的 workload 都是在做流量处理，这个开销被形象的称作数据中心税（Datacenter Tax）。即还未运行业务程序，先接入网络数据就要占去的计算资源。冯诺依曼内存墙冯·诺依曼体系结构作为一种程序存储计算机，CPU 从 Main Memory 中读取数据，在完成计算后再将数据回写，该模型存在的潜规则是 CPU 计算速度和 Main Memory 传输速度相当。相对的，一旦失衡，那么慢的一方就会成为瓶颈。而 “内存墙“ 的现实就是 DDR（Main Memory 的容量和 Bus 传输带宽）已经成为了那块短板。这意味着传统计算机体系结构已经无法满足新兴的内存密集型应用程序的需求。例如 AI DL/ML 训练场景，具有高并发、高耦合的特点，不仅有大量的数据参与到整个算法运行的过程中，这些数据之间的耦合性也非常强，因此对 Main Memory 提出了非常高的要求。为了解决内存墙问题，业界目前有下列几种 “补丁式“ 的解决方法：加大存储带宽：采用高带宽的外部存储，如 HDM2，降低对 DDR 的访问。这种方法虽然看似最简单直接，但问题在于缓存的调度对深度学习的有效性就是一个难点；片上存储：在处理器芯片里集成大存储，抛弃 DDR，比如集成几十兆字节到上百兆的 SRAM。这种方法看上去也比较简单直接，但成本高昂也是显著的劣势。存算一体（In-Memory Computing)：在存储器上集成计算单元，现在也是一个比较受关注的方向。数据 I/O 路径冗长基于以上等等背景，DSA/DSL（专用的协处理器）和异构计算等领域成为了热门研究方向。针对不同的应用场景，在冯·诺依曼体系中部署专用的协处理器（e.g. GPU、ASIC、FPGA、DSA）来进行加速处理。但也存在一个直观的问题，CPU 和 Main Memory 以及 Device Memory 之间的数据 I/O 路径冗长，同样会成为计算性能的瓶颈。以 CPU+GPU 异构计算为例，GPU 具有强大的计算能力，能够同时并行工作数百个的内核，但 “CPU+GPU 分离” 架构中存在海量数据无法轻松存储到 GPU Device Memory 中，需要等待显存数据刷新。同时，海量数据在 CPU 和 GPU 等加速器之间来回移动，也加剧了额外的速率消耗。可见，以 CPU 为中心的体系架构在异构计算场景中，由于内存 IO 路径太长也会成为一种性能瓶颈。以 DPU 为中心的新型架构以 DPU 为中心的数据中心，是一种全新的计算机体系结构。在数据中心，将更多 CPU 和 GPU 的 workload offload 到 DPU（Data Processing Unit，数据处理单元）中，使得计算、存储和网络变得更加紧耦合。在以往，计算和网络是相互独立的，各自关心自己的事情。正如上文中提到的，随着 CPU 性能瓶颈摩尔定律失效，就需要通过各种方式进行计算加速，但这些手段都使得问题变得非常的复杂。再遵循旧思维以各自的方式去解决各自的问题变得难度都很大。而把计算和网络两者融合起来，用网络的方式解决计算的问题，用计算的方式解决网络的挑战，却是非常高效的新思路。未来的一个重要趋势是，计算、存储和网络都在不断融合。而 DPU 的核心就是让计算发生在靠近数据产生的地方。CPU 负责通用计算。GPU 负责加速计算DPU 负责数据中心内部的数据传输和处理。DPU 的抽象架构控制平面由通用处理器（x86 / ARM / MIPS）和片上内存实现，可运行 NIC OS（Linux），主要负责以下工作：DPU 设备运行管理安全管理：信任根、安全启动、安全固件升级、基于身份验证的容器和应用生命周期管理等。实时监控：对 DPU 的各个子系统进行监控，包括：数据平面处理单元等。实时观察设备是否可用、设备中流量是否正常，周期性生成报表，记录设备访问日志核配置修改日志。DPU 计算任务和资源配置网络功能控制面计算任务。存储功能控制面计算任务。等。数据平面由专用处理器（NP / ASIC / FPGA）和 NIC 实现，主要负责以下工作：可编程的数据报文处理功能。协议加速功能。I/O 子系统System I/O：由 PCIe 实现，负责 DPU 和其他系统的集成。支持 Endpoint 和 Root Complex 两种实现类型。Endpoint System I/O：将 DPU 作为 “从设备” 接入到 Host CPU 处理平台，将数据上传至 CPU 进行处理。Root Complex System I/O：将 DPU 作为 “主设备” 接入其他加速处理平台（e.g. FPGA、GPU）或高速外部设备（e.g. SSD），将数据分流至加速平台或外设进行处理。Network I/O：由 NIC（网络协议处理器单元）实现，与 IP/FC Fabric 互联。Main Memory I/O：由 DDR 和 HBM 接口实现，与片外内存互联，可作为 Cache 和 Shared Memory。DDR 可以提供比较大的存储容量（512GB 以上）。HBM 可以提供比较大的存储带宽（500GB/s 以上）。- END -关于 “云物互联” 微信公众号：欢迎关注 “云物互联” 微信公众号，我们专注于云计算、云原生、SDN/NFV、边缘计算及 5G 网络技术的发展及应用。热爱开源，拥抱开源！技术即沟通化云为雨，落地成林发布于 2023-04-29 11:30・IP 属地北京设备dpu智能网卡赞同 101 条评论分享喜欢收藏申请转载文章被以下专栏收录OpenStack IaaSOpenStack 开源云计算项目技术分享。Linux 操作系统和网络Linux 操作系统和网络技

Attention Required! | Cloudflare

Please enable cookies.

Sorry, you have been blocked

You are unable to access techopedia.com

Why have I been blocked?

This website is using a security service to protect itself from online attacks. The action you just performed triggered the security solution. There are several actions that could trigger this block including submitting a certain word or phrase, a SQL command or malformed data.

What can I do to resolve this?

You can email the site owner to let them know you were blocked. Please include what you were doing when this page came up and the Cloudflare Ray ID found at the bottom of this page.

Cloudflare Ray ID: 86440b7bfa1b09a0

•

Your IP:

Click to reveal

49.157.13.121

•

Performance & security by Cloudflare

Network 之二 Ethernet（以太网）中的 MAC、MII、PHY 详解_ethernet phy-CSDN博客

Network 之二 Ethernet（以太网）中的 MAC、MII、PHY 详解

ZC·Shou

已于 2023-02-28 08:43:20 修改

阅读量10w+

1.8k

点赞数

328

分类专栏：

Network

文章标签：

MAC

PHY

以太网

Ethernet

MII

于 2018-04-26 14:06:47 首次发布

本文链接：https://blog.csdn.net/ZCShouCSDN/article/details/80090802

版权

Network

专栏收录该内容

13 篇文章

12 订阅

¥29.90

¥99.00

订阅专栏

超级会员免费看

结构

从硬件的角度看，以太网接口电路主要由 MAC（Media Access Control，MAC）控制器和物理层接口 PHY（Physical Layer，PHY）两大部分构成。如下图所示：但是，在实际的设计中，以上三部分并不一定是独立分开的。由于，PHY 整合了大量模拟硬件，而 MAC 则是典型的全数字器件。考虑到芯片面积及模拟/数字混合架构的原因，通常，将 MAC 集成进微控制器而将 PHY 留在片外（现在，更灵活、密度更高的芯片技术已经可以实现 MAC 和 PHY 的单芯片整合）。

CPU 集成 MAC 与 PHY，目前来说并不多见。

CPU 集成 MAC，PHY 采用独立芯片，这种比较常见。

了解本专栏

订阅专栏解锁全文

超级会员免费看

优惠劵

ZC·Shou

关注

328

踩

1855

觉得还不错?

一键收藏

打赏

知道了

Network 之二 Ethernet（以太网）中的 MAC、MII、PHY 详解

结构从硬件的角度看，以太网接口电路主要由MAC（Media Access Control）控制器和物理层接口PHY（Physical Layer，PHY）两大部分构成。如下图所示但是，在实际的设计中，以上三部分并不一定独立分开的。由于，PHY整合了大量模拟硬件，而MAC是典型的全数字器件。考虑到芯片面积及模拟/数字混合架构的原因，通常，将MAC集成进微控制器而将PHY留在片外...

复制链接

扫一扫

专栏目录

订阅专栏

以太网（一）MAC、MII、PHY 介绍

ID2442512720的博客

06-20

2297

即媒体访问控制层协议。MAC由硬件控制器和MAC通信协议构成。该协议位于OSI七层协议中数据链路层的下半部分，主要负责控制与连接物理层的物理介质。MAC硬件框图如下图所示：在发送数据的时候，MAC 协议可以事先判断是否可以发送数据，如果可以发送将给数据加上一些控制信息，最终将数据以及控制信息以规定的格式发送到物理层；在接收数据的时候，MAC 协议首先判断输入的信息并是否发生传输错误，如果没有错误，则去掉控制信息发送至 LLC（逻辑链路控制）层。

MAC,PHY,MII的关系

09-25

本文主要介绍以太网的MAC（Media Access Control，即媒体访问控制子层协议）和PHY（物理层）之间的MII（Media Independent Interface ，媒体独立接口），以及MII的各种衍生版本——GMII、SGMII、RMII、RGMII等。

28 条评论

您还未登录，请先

后发表或查看评论

ETHERNET中MAC通过MII总线控制PHY的过程

瑞风轻拂

12-20

6611

一些芯片资料常常遇到MAC MODE 和PHY MODE 不知什么意思？

一般以太网芯片中涉及，一般交换芯片（比如ks8993吗，8305sb等）可以选择两种模式，mac和phy。

应该就是针对不同的外部接口采用的不同模式吧，主要是区别在于针对OSI七层协议中数据链路层中处理信息所处的层不一样，也就导致处理的对象不一样。

以下资料来自网络是针对有关MAC、PHY和MII 的详细解

STM32的以太网外设+PHY（LAN8720）使用详解（3）：PHY寄存器详解

kevin1499的博客

12-22

1847

STM32的以太网外设+PHY（LAN8720）的PHY寄存器详解。

关于MII、RMII、GMII、RGMII、PHY、网络变压器、RJ45的硬件总结

Ethernet接口与Interface - 技象科技

跳至内容

技象科技有限公司官方网站

技象天猫旗舰店技象天猫旗舰店

技象京东旗舰店技象京东旗舰店

IoT演示平台IoT演示平台

首页

解决方案

菜单切换

TPUNB®

城域物联网

菜单切换

智慧农业

环境监测

智慧停车

局域物联网

菜单切换

智慧楼宇

智慧养老

智慧照明

智慧能源

工业智控物联网

菜单切换

动环监控

终端自组网

菜单切换

自组网

产品终端

菜单切换

芯片产品

模组产品

网关产品

应用产品

云平台

关于我们

菜单切换

荣誉资质

合作伙伴

公司一览

招聘岗位

新闻资讯

菜单切换

公司新闻

下载专区

联系我们

首页

TPUNB®

物联网解决方案

菜单切换

城域物联网

局域物联网

工业智控物联网

终端自组网

物联网产品终端

菜单切换

芯片产品

模组产品

网关产品

物联网云平台

关于我们

菜单切换

荣誉资质

合作伙伴

公司新闻

联系我们

Main Menu

首页

TPUNB®

物联网解决方案

菜单切换

城域物联网

局域物联网

工业智控物联网

终端自组网

物联网产品终端

菜单切换

芯片产品

模组产品

网关产品

物联网云平台

关于我们

菜单切换

荣誉资质

合作伙伴

公司新闻

联系我们

技象科技首页 / 行业百科 / Ethernet接口与Interface

Ethernet接口与Interface作者：

技象物联网

/ 行业百科 / 电子技术 / 2023年9月28日 00:26:31 2023年9月28日 00:26:31

Ethernet接口和interface是网络设备中常见的两种接口，它们可以帮助用户实现网络连接。Ethernet接口是一种物理接口，可以将设备连接到网络，而interface则是一种软件接口，用于实现网络设备之间的通信。本文将详细介绍Ethernet接口和interface的功能、优势以及如何使用它们来实现网络连接。

Ethernet接口

什么是Ethernet接口

Ethernet接口是一种物理接口，它可以将设备连接到网络，并实现数据传输。它通常位于网络设备的背面，并使用RJ45插头连接到网络线路。Ethernet接口可以支持多种网络协议，例如以太网、Fast Ethernet和Gigabit Ethernet等，并且可以实现最高传输速率为1Gbps的数据传输。

Ethernet接口的优势

Ethernet接口具有多种优势，首先它可以支持多种网络协议，可以实现最高传输速率为1Gbps的数据传输，而且它的插头也非常容易操作，可以轻松实现网络连接。此外，Ethernet接口还具有较高的耐用性，可以持续提供高速稳定的数据传输服务，而且它的价格也比较实惠，可以满足不同用户的需求。

如何使用Ethernet接口

使用Ethernet接口连接网络非常简单，首先，您需要将Ethernet插头插入网络设备的背面，然后将另一端的插头插入网线上，最后检查网络设备是否正常工作，如果一切正常，则网络连接就完成了。

Interface

什么是Interface

Interface是一种软件接口，它可以帮助用户实现网络设备之间的通信。它可以帮助用户更好地管理网络设备，并可以实现多种网络协议之间的转换，从而实现网络设备之间的通信。

Interface的优势

Interface具有多种优势，首先它可以帮助用户更好地管理网络设备，可以实现多种网络协议之间的转换，从而实现网络设备之间的通信。此外，Interface还可以实现高速稳定的数据传输，并且可以轻松实现网络连接，而且它的价格也比较实惠，可以满足不同用户的需求。

如何使用Interface

使用Interface连接网络非常简单，首先，您需要在网络设备上安装Interface软件，然后根据提示进行设置，最后检查网络设备是否正常工作，如果一切正常，则网络连接就完成了。

总结

本文详细介绍了Ethernet接口和interface的功能、优势以及如何使用它们来实现网络连接。Ethernet接口是一种物理接口，可以将设备连接到网络，而interface则是一种软件接口，用于实现网络设备之间的通信。它们都具有较高的耐用性，可以持续提供高速稳定的数据传输服务，而且它们的价格也比较实惠，可以满足不同用户的需求。

Ethernet - Wikipedia

Jump to content

Main menu

move to sidebar

hide

Navigation

Main pageContentsCurrent eventsRandom articleAbout WikipediaContact usDonate

Contribute

HelpLearn to editCommunity portalRecent changesUpload file

Create account

Personal tools

Create account Log in

Pages for logged out editors learn more

ContributionsTalk

Contents

move to sidebar

hide

(Top)

1History

2Standardization

3Evolution

Toggle Evolution subsection

3.1Shared medium

3.2Repeaters and hubs

3.3Bridging and switching

3.4Advanced networking

4Varieties

5Frame structure

6Autonegotiation

7Error conditions

Toggle Error conditions subsection

7.1Switching loop

7.2Jabber

7.3Runt frames

8See also

9Notes

10References

11Further reading

12External links

Toggle the table of contents

Ethernet

76 languages

AfrikaansالعربيةAsturianuAzərbaycancaবাংলাБеларускаяБългарскиBosanskiBrezhonegCatalàČeštinaDanskDeutschEestiΕλληνικάEspañolEsperantoEuskaraفارسیFrançaisGaeilgeGaelgGalegoગુજરાતી한국어हिन्दीHrvatskiBahasa IndonesiaÍslenskaItalianoעבריתಕನ್ನಡҚазақшаKurdîLatinaLatviešuLëtzebuergeschLietuviųLombardMagyarМакедонскиമലയാളംमराठीBahasa MelayuNederlands日本語Norsk bokmålNorsk nynorskਪੰਜਾਬੀپښتوPolskiPortuguêsRomânăРусскийShqipSimple EnglishSlovenčinaSlovenščinaکوردیСрпски / srpskiSrpskohrvatski / српскохрватскиSuomiSvenskaTagalogதமிழ்తెలుగుไทยTürkçeУкраїнськаاردوTiếng ViệtVõro吴语ייִדיש粵語中文

Edit links

ArticleTalk

English

ReadEditView history

Tools

move to sidebar

hide

Actions

ReadEditView history

General

What links hereRelated changesUpload fileSpecial pagesPermanent linkPage informationCite this pageGet shortened URLDownload QR codeWikidata item

Print/export

Download as PDFPrintable version

In other projects

Wikimedia Commons

From Wikipedia, the free encyclopedia

Computer networking technology

An Ethernet port on a laptop computer connected to a twisted pair cable with an 8P8C modular connector

Symbol used by Apple on some devices to denote an Ethernet connection

Ethernet (/ˈiːθərnɛt/ EE-thər-net) is a family of wired computer networking technologies commonly used in local area networks (LAN), metropolitan area networks (MAN) and wide area networks (WAN).[1] It was commercially introduced in 1980 and first standardized in 1983 as IEEE 802.3. Ethernet has since been refined to support higher bit rates, a greater number of nodes, and longer link distances, but retains much backward compatibility. Over time, Ethernet has largely replaced competing wired LAN technologies such as Token Ring, FDDI and ARCNET.

The original 10BASE5 Ethernet uses a thick coaxial cable as a shared medium. This was largely superseded by 10BASE2, which used a thinner and more flexible cable that was both cheaper and easier to use. More modern Ethernet variants use twisted pair and fiber optic links in conjunction with switches. Over the course of its history, Ethernet data transfer rates have been increased from the original 2.94 Mbit/s[2] to the latest 400 Gbit/s, with rates up to 1.6 Tbit/s under development. The Ethernet standards include several wiring and signaling variants of the OSI physical layer.

Systems communicating over Ethernet divide a stream of data into shorter pieces called frames. Each frame contains source and destination addresses, and error-checking data so that damaged frames can be detected and discarded; most often, higher-layer protocols trigger retransmission of lost frames. Per the OSI model, Ethernet provides services up to and including the data link layer.[3] The 48-bit MAC address was adopted by other IEEE 802 networking standards, including IEEE 802.11 (Wi-Fi), as well as by FDDI. EtherType values are also used in Subnetwork Access Protocol (SNAP) headers.

Ethernet is widely used in homes and industry, and interworks well with wireless Wi-Fi technologies. The Internet Protocol is commonly carried over Ethernet and so it is considered one of the key technologies that make up the Internet.

History[edit]

Accton Etherpocket-SP parallel port Ethernet adapter (c. 1990). Supports both coaxial (10BASE2) and twisted pair (10BASE-T) cables. Power is drawn from a PS/2 port passthrough cable.

Ethernet was developed at Xerox PARC between 1973 and 1974[4][5] as a means to allow Alto computers to communicate with each other.[6] It was inspired by ALOHAnet, which Robert Metcalfe had studied as part of his PhD dissertation[7][8] and was originally called the Alto Aloha Network.[6] The idea was first documented in a memo that Metcalfe wrote on May 22, 1973, where he named it after the luminiferous aether once postulated to exist as an "omnipresent, completely-passive medium for the propagation of electromagnetic waves."[4][9][10] In 1975, Xerox filed a patent application listing Metcalfe, David Boggs, Chuck Thacker, and Butler Lampson as inventors.[11] In 1976, after the system was deployed at PARC, Metcalfe and Boggs published a seminal paper.[12][a] Yogen Dalal,[14] Ron Crane, Bob Garner, and Roy Ogus facilitated the upgrade from the original 2.94 Mbit/s protocol to the 10 Mbit/s protocol, which was released to the market in 1980.[15]

Metcalfe left Xerox in June 1979 to form 3Com.[4][16] He convinced Digital Equipment Corporation (DEC), Intel, and Xerox to work together to promote Ethernet as a standard. As part of that process Xerox agreed to relinquish their 'Ethernet' trademark.[17] The first standard was published on September 30, 1980, as "The Ethernet, A Local Area Network. Data Link Layer and Physical Layer Specifications". This so-called DIX standard (Digital Intel Xerox)[18] specified 10 Mbit/s Ethernet, with 48-bit destination and source addresses and a global 16-bit Ethertype-type field.[19] Version 2 was published in November 1982[20] and defines what has become known as Ethernet II. Formal standardization efforts proceeded at the same time and resulted in the publication of IEEE 802.3 on June 23, 1983.[21]

Ethernet initially competed with Token Ring and other proprietary protocols. Ethernet was able to adapt to market needs and with 10BASE2, shift to inexpensive thin coaxial cable and from 1990, to the now-ubiquitous twisted pair with 10BASE-T. By the end of the 1980s, Ethernet was clearly the dominant network technology.[4] In the process, 3Com became a major company. 3Com shipped its first 10 Mbit/s Ethernet 3C100 NIC in March 1981, and that year started selling adapters for PDP-11s and VAXes, as well as Multibus-based Intel and Sun Microsystems computers.[22]: 9 This was followed quickly by DEC's Unibus to Ethernet adapter, which DEC sold and used internally to build its own corporate network, which reached over 10,000 nodes by 1986, making it one of the largest computer networks in the world at that time.[23] An Ethernet adapter card for the IBM PC was released in 1982, and, by 1985, 3Com had sold 100,000.[16] In the 1980s, IBM's own PC Network product competed with Ethernet for the PC, and through the 1980s, LAN hardware, in general, was not common on PCs. However, in the mid to late 1980s, PC networking did become popular in offices and schools for printer and fileserver sharing, and among the many diverse competing LAN technologies of that decade, Ethernet was one of the most popular. Parallel port based Ethernet adapters were produced for a time, with drivers for DOS and Windows. By the early 1990s, Ethernet became so prevalent that Ethernet ports began to appear on some PCs and most workstations. This process was greatly sped up with the introduction of 10BASE-T and its relatively small modular connector, at which point Ethernet ports appeared even on low-end motherboards.[citation needed]

Since then, Ethernet technology has evolved to meet new bandwidth and market requirements.[24] In addition to computers, Ethernet is now used to interconnect appliances and other personal devices.[4] As Industrial Ethernet it is used in industrial applications and is quickly replacing legacy data transmission systems in the world's telecommunications networks.[25] By 2010, the market for Ethernet equipment amounted to over $16 billion per year.[26]

Standardization[edit]

An Intel 82574L Gigabit Ethernet NIC, PCI Express ×1 card

In February 1980, the Institute of Electrical and Electronics Engineers (IEEE) started project 802 to standardize local area networks (LAN).[16][27] The "DIX-group" with Gary Robinson (DEC), Phil Arst (Intel), and Bob Printis (Xerox) submitted the so-called "Blue Book" CSMA/CD specification as a candidate for the LAN specification.[19] In addition to CSMA/CD, Token Ring (supported by IBM) and Token Bus (selected and henceforward supported by General Motors) were also considered as candidates for a LAN standard. Competing proposals and broad interest in the initiative led to strong disagreement over which technology to standardize. In December 1980, the group was split into three subgroups, and standardization proceeded separately for each proposal.[16]

Delays in the standards process put at risk the market introduction of the Xerox Star workstation and 3Com's Ethernet LAN products. With such business implications in mind, David Liddle (General Manager, Xerox Office Systems) and Metcalfe (3Com) strongly supported a proposal of Fritz Röscheisen (Siemens Private Networks) for an alliance in the emerging office communication market, including Siemens' support for the international standardization of Ethernet (April 10, 1981). Ingrid Fromm, Siemens' representative to IEEE 802, quickly achieved broader support for Ethernet beyond IEEE by the establishment of a competing Task Group "Local Networks" within the European standards body ECMA TC24. In March 1982, ECMA TC24 with its corporate members reached an agreement on a standard for CSMA/CD based on the IEEE 802 draft.[22]: 8 Because the DIX proposal was most technically complete and because of the speedy action taken by ECMA which decisively contributed to the conciliation of opinions within IEEE, the IEEE 802.3 CSMA/CD standard was approved in December 1982.[16] IEEE published the 802.3 standard as a draft in 1983 and as a standard in 1985.[28]

Approval of Ethernet on the international level was achieved by a similar, cross-partisan action with Fromm as the liaison officer working to integrate with International Electrotechnical Commission (IEC) Technical Committee 83 and International Organization for Standardization (ISO) Technical Committee 97 Sub Committee 6. The ISO 8802-3 standard was published in 1989.[29]

Evolution[edit]

Internet protocol suite

Application layer

BGP

DHCP (v6)

DNS

FTP

HTTP (HTTP/3)

HTTPS

IMAP

IRC

LDAP

MGCP

MQTT

NNTP

NTP

OSPF

POP

PTP

ONC/RPC

RTP

RTSP

RIP

SIP

SMTP

SNMP

SSH

Telnet

TLS/SSL

XMPP

more...

Transport layer

TCP

UDP

DCCP

SCTP

RSVP

QUIC

more...

Internet layer

ICMP (v6)

NDP

ECN

IGMP

IPsec

more...

Link layer

ARP

Tunnels

PPP

MAC

more...

vte

Ethernet has evolved to include higher bandwidth, improved medium access control methods, and different physical media. The multidrop coaxial cable was replaced with physical point-to-point links connected by Ethernet repeaters or switches.[30]

Ethernet stations communicate by sending each other data packets: blocks of data individually sent and delivered. As with other IEEE 802 LANs, adapters come programmed with globally unique 48-bit MAC address so that each Ethernet station has a unique address.[b] The MAC addresses are used to specify both the destination and the source of each data packet. Ethernet establishes link-level connections, which can be defined using both the destination and source addresses. On reception of a transmission, the receiver uses the destination address to determine whether the transmission is relevant to the station or should be ignored. A network interface normally does not accept packets addressed to other Ethernet stations.[c][d]

An EtherType field in each frame is used by the operating system on the receiving station to select the appropriate protocol module (e.g., an Internet Protocol version such as IPv4). Ethernet frames are said to be self-identifying, because of the EtherType field. Self-identifying frames make it possible to intermix multiple protocols on the same physical network and allow a single computer to use multiple protocols together.[31] Despite the evolution of Ethernet technology, all generations of Ethernet (excluding early experimental versions) use the same frame formats.[32] Mixed-speed networks can be built using Ethernet switches and repeaters supporting the desired Ethernet variants.[33]

Due to the ubiquity of Ethernet, and the ever-decreasing cost of the hardware needed to support it, by 2004 most manufacturers built Ethernet interfaces directly into PC motherboards, eliminating the need for a separate network card.[34]

Shared medium[edit]

Older Ethernet equipment. Clockwise from top-left: An Ethernet transceiver with an in-line 10BASE2 adapter, a similar model transceiver with a 10BASE5 adapter, an AUI cable, a different style of transceiver with 10BASE2 BNC T-connector, two 10BASE5 end fittings (N connectors), an orange vampire tap installation tool (which includes a specialized drill bit at one end and a socket wrench at the other), and an early model 10BASE5 transceiver (h4000) manufactured by DEC. The short length of yellow 10BASE5 cable has one end fitted with an N connector and the other end prepared to have an N connector shell installed; the half-black, half-grey rectangular object through which the cable passes is an installed vampire tap.

Ethernet was originally based on the idea of computers communicating over a shared coaxial cable acting as a broadcast transmission medium. The method used was similar to those used in radio systems,[e] with the common cable providing the communication channel likened to the Luminiferous aether in 19th-century physics, and it was from this reference that the name "Ethernet" was derived.[35]

Original Ethernet's shared coaxial cable (the shared medium) traversed a building or campus to every attached machine. A scheme known as carrier-sense multiple access with collision detection (CSMA/CD) governed the way the computers shared the channel. This scheme was simpler than competing Token Ring or Token Bus technologies.[f] Computers are connected to an Attachment Unit Interface (AUI) transceiver, which is in turn connected to the cable (with thin Ethernet the transceiver is usually integrated into the network adapter). While a simple passive wire is highly reliable for small networks, it is not reliable for large extended networks, where damage to the wire in a single place, or a single bad connector, can make the whole Ethernet segment unusable.[g]

Through the first half of the 1980s, Ethernet's 10BASE5 implementation used a coaxial cable 0.375 inches (9.5 mm) in diameter, later called thick Ethernet or thicknet. Its successor, 10BASE2, called thin Ethernet or thinnet, used the RG-58 coaxial cable. The emphasis was on making installation of the cable easier and less costly.[36]: 57

Since all communication happens on the same wire, any information sent by one computer is received by all, even if that information is intended for just one destination.[h] The network interface card interrupts the CPU only when applicable packets are received: the card ignores information not addressed to it.[c] Use of a single cable also means that the data bandwidth is shared, such that, for example, available data bandwidth to each device is halved when two stations are simultaneously active.[37]

A collision happens when two stations attempt to transmit at the same time. They corrupt transmitted data and require stations to re-transmit. The lost data and re-transmission reduces throughput. In the worst case, where multiple active hosts connected with maximum allowed cable length attempt to transmit many short frames, excessive collisions can reduce throughput dramatically. However, a Xerox report in 1980 studied performance of an existing Ethernet installation under both normal and artificially generated heavy load. The report claimed that 98% throughput on the LAN was observed.[38] This is in contrast with token passing LANs (Token Ring, Token Bus), all of which suffer throughput degradation as each new node comes into the LAN, due to token waits. This report was controversial, as modeling showed that collision-based networks theoretically became unstable under loads as low as 37% of nominal capacity. Many early researchers failed to understand these results. Performance on real networks is significantly better.[39]

In a modern Ethernet, the stations do not all share one channel through a shared cable or a simple repeater hub; instead, each station communicates with a switch, which in turn forwards that traffic to the destination station. In this topology, collisions are only possible if station and switch attempt to communicate with each other at the same time, and collisions are limited to this link. Furthermore, the 10BASE-T standard introduced a full duplex mode of operation which became common with Fast Ethernet and the de facto standard with Gigabit Ethernet. In full duplex, switch and station can send and receive simultaneously, and therefore modern Ethernets are completely collision-free.

Comparison between original Ethernet and modern Ethernet

The original Ethernet implementation: shared medium, collision-prone. All computers trying to communicate share the same cable, and so compete with each other.

Modern Ethernet implementation: switched connection, collision-free. Each computer communicates only with its own switch, without competition for the cable with others.

Repeaters and hubs[edit]

A 1990s ISA network interface card supporting both coaxial-cable-based 10BASE2 (BNC connector, left) and twisted-pair-based 10BASE-T (8P8C connector, right)

Main article: Ethernet hub

For signal degradation and timing reasons, coaxial Ethernet segments have a restricted size.[40] Somewhat larger networks can be built by using an Ethernet repeater. Early repeaters had only two ports, allowing, at most, a doubling of network size. Once repeaters with more than two ports became available, it was possible to wire the network in a star topology. Early experiments with star topologies (called Fibernet) using optical fiber were published by 1978.[41]

Shared cable Ethernet is always hard to install in offices because its bus topology is in conflict with the star topology cable plans designed into buildings for telephony. Modifying Ethernet to conform to twisted pair telephone wiring already installed in commercial buildings provided another opportunity to lower costs, expand the installed base, and leverage building design, and, thus, twisted-pair Ethernet was the next logical development in the mid-1980s.

Ethernet on unshielded twisted-pair cables (UTP) began with StarLAN at 1 Mbit/s in the mid-1980s. In 1987 SynOptics introduced the first twisted-pair Ethernet at 10 Mbit/s in a star-wired cabling topology with a central hub, later called LattisNet.[16][35]: 29 [42] These evolved into 10BASE-T, which was designed for point-to-point links only, and all termination was built into the device. This changed repeaters from a specialist device used at the center of large networks to a device that every twisted pair-based network with more than two machines had to use. The tree structure that resulted from this made Ethernet networks easier to maintain by preventing most faults with one peer or its associated cable from affecting other devices on the network.[citation needed]

Despite the physical star topology and the presence of separate transmit and receive channels in the twisted pair and fiber media, repeater-based Ethernet networks still use half-duplex and CSMA/CD, with only minimal activity by the repeater, primarily generation of the jam signal in dealing with packet collisions. Every packet is sent to every other port on the repeater, so bandwidth and security problems are not addressed. The total throughput of the repeater is limited to that of a single link, and all links must operate at the same speed.[35]: 278

Bridging and switching[edit]

Patch cables with patch fields of two Ethernet switches

Main articles: Network bridge and Network switch

While repeaters can isolate some aspects of Ethernet segments, such as cable breakages, they still forward all traffic to all Ethernet devices. The entire network is one collision domain, and all hosts have to be able to detect collisions anywhere on the network. This limits the number of repeaters between the farthest nodes and creates practical limits on how many machines can communicate on an Ethernet network. Segments joined by repeaters have to all operate at the same speed, making phased-in upgrades impossible.[citation needed]

To alleviate these problems, bridging was created to communicate at the data link layer while isolating the physical layer. With bridging, only well-formed Ethernet packets are forwarded from one Ethernet segment to another; collisions and packet errors are isolated. At initial startup, Ethernet bridges work somewhat like Ethernet repeaters, passing all traffic between segments. By observing the source addresses of incoming frames, the bridge then builds an address table associating addresses to segments. Once an address is learned, the bridge forwards network traffic destined for that address only to the associated segment, improving overall performance. Broadcast traffic is still forwarded to all network segments. Bridges also overcome the limits on total segments between two hosts and allow the mixing of speeds, both of which are critical to the incremental deployment of faster Ethernet variants.[citation needed]

In 1989, Motorola Codex introduced their 6310 EtherSpan, and Kalpana introduced their EtherSwitch; these were examples of the first commercial Ethernet switches.[i] Early switches such as this used cut-through switching where only the header of the incoming packet is examined before it is either dropped or forwarded to another segment.[43] This reduces the forwarding latency. One drawback of this method is that it does not readily allow a mixture of different link speeds. Another is that packets that have been corrupted are still propagated through the network. The eventual remedy for this was a return to the original store and forward approach of bridging, where the packet is read into a buffer on the switch in its entirety, its frame check sequence verified and only then the packet is forwarded.[43] In modern network equipment, this process is typically done using application-specific integrated circuits allowing packets to be forwarded at wire speed.[citation needed]

When a twisted pair or fiber link segment is used and neither end is connected to a repeater, full-duplex Ethernet becomes possible over that segment. In full-duplex mode, both devices can transmit and receive to and from each other at the same time, and there is no collision domain.[44] This doubles the aggregate bandwidth of the link and is sometimes advertised as double the link speed (for example, 200 Mbit/s for Fast Ethernet).[j] The elimination of the collision domain for these connections also means that all the link's bandwidth can be used by the two devices on that segment and that segment length is not limited by the constraints of collision detection.

Since packets are typically delivered only to the port they are intended for, traffic on a switched Ethernet is less public than on shared-medium Ethernet. Despite this, switched Ethernet should still be regarded as an insecure network technology, because it is easy to subvert switched Ethernet systems by means such as ARP spoofing and MAC flooding.[citation needed][45]

The bandwidth advantages, the improved isolation of devices from each other, the ability to easily mix different speeds of devices and the elimination of the chaining limits inherent in non-switched Ethernet have made switched Ethernet the dominant network technology.[46]

Advanced networking[edit]

A core Ethernet switch

Simple switched Ethernet networks, while a great improvement over repeater-based Ethernet, suffer from single points of failure, attacks that trick switches or hosts into sending data to a machine even if it is not intended for it, scalability and security issues with regard to switching loops, broadcast radiation, and multicast traffic.[citation needed]

Advanced networking features in switches use Shortest Path Bridging (SPB) or the Spanning Tree Protocol (STP) to maintain a loop-free, meshed network, allowing physical loops for redundancy (STP) or load-balancing (SPB). Shortest Path Bridging includes the use of the link-state routing protocol IS-IS to allow larger networks with shortest path routes between devices.

Advanced networking features also ensure port security, provide protection features such as MAC lockdown[47] and broadcast radiation filtering, use VLANs to keep different classes of users separate while using the same physical infrastructure, employ multilayer switching to route between different classes, and use link aggregation to add bandwidth to overloaded links and to provide some redundancy.[citation needed]

In 2016, Ethernet replaced InfiniBand as the most popular system interconnect of TOP500 supercomputers.[48]

Varieties[edit]

Main articles: Ethernet physical layer and Ethernet over twisted pair

The Ethernet physical layer evolved over a considerable time span and encompasses coaxial, twisted pair and fiber-optic physical media interfaces, with speeds from 1 Mbit/s to 400 Gbit/s.[49] The first introduction of twisted-pair CSMA/CD was StarLAN, standardized as 802.3 1BASE5.[50] While 1BASE5 had little market penetration, it defined the physical apparatus (wire, plug/jack, pin-out, and wiring plan) that would be carried over to 10BASE-T through 10GBASE-T.

The most common forms used are 10BASE-T, 100BASE-TX, and 1000BASE-T. All three use twisted-pair cables and 8P8C modular connectors. They run at 10 Mbit/s, 100 Mbit/s, and 1 Gbit/s, respectively.[51][52][53]

Fiber optic variants of Ethernet (that commonly use SFP modules) are also very popular in larger networks, offering high performance, better electrical isolation and longer distance (tens of kilometers with some versions). In general, network protocol stack software will work similarly on all varieties.[54]

Frame structure[edit]

A close-up of the SMSC LAN91C110 (SMSC 91x) chip, an embedded Ethernet chip

Main article: Ethernet frame

In IEEE 802.3, a datagram is called a packet or frame. Packet is used to describe the overall transmission unit and includes the preamble, start frame delimiter (SFD) and carrier extension (if present).[k] The frame begins after the start frame delimiter with a frame header featuring source and destination MAC addresses and the EtherType field giving either the protocol type for the payload protocol or the length of the payload. The middle section of the frame consists of payload data including any headers for other protocols (for example, Internet Protocol) carried in the frame. The frame ends with a 32-bit cyclic redundancy check, which is used to detect corruption of data in transit.[55]: sections 3.1.1 and 3.2 Notably, Ethernet packets have no time-to-live field, leading to possible problems in the presence of a switching loop.

Autonegotiation[edit]

Main article: Autonegotiation

Autonegotiation is the procedure by which two connected devices choose common transmission parameters, e.g. speed and duplex mode. Autonegotiation was initially an optional feature, first introduced with 100BASE-TX (1995 IEEE 802.3u Fast Ethernet standard), and is backward compatible with 10BASE-T. The specification was improved in the 1998 release of IEEE 802.3. Autonegotiation is mandatory for 1000BASE-T and faster.

Error conditions[edit]

Switching loop[edit]

Main article: Switching loop

A switching loop or bridge loop occurs in computer networks when there is more than one Layer 2 (OSI model) path between two endpoints (e.g. multiple connections between two network switches or two ports on the same switch connected to each other). The loop creates broadcast storms as broadcasts and multicasts are forwarded by switches out every port, the switch or switches will repeatedly rebroadcast the broadcast messages flooding the network. Since the Layer 2 header does not support a time to live (TTL) value, if a frame is sent into a looped topology, it can loop forever.[56]

A physical topology that contains switching or bridge loops is attractive for redundancy reasons, yet a switched network must not have loops. The solution is to allow physical loops, but create a loop-free logical topology using the SPB protocol or the older STP on the network switches.[citation needed]

Jabber[edit]

A node that is sending longer than the maximum transmission window for an Ethernet packet is considered to be jabbering. Depending on the physical topology, jabber detection and remedy differ somewhat.

An MAU is required to detect and stop abnormally long transmission from the DTE (longer than 20–150 ms) in order to prevent permanent network disruption.[57]

On an electrically shared medium (10BASE5, 10BASE2, 1BASE5), jabber can only be detected by each end node, stopping reception. No further remedy is possible.[58]

A repeater/repeater hub uses a jabber timer that ends retransmission to the other ports when it expires. The timer runs for 25,000 to 50,000 bit times for 1 Mbit/s,[59] 40,000 to 75,000 bit times for 10 and 100 Mbit/s,[60][61] and 80,000 to 150,000 bit times for 1 Gbit/s.[62] Jabbering ports are partitioned off the network until a carrier is no longer detected.[63]

End nodes utilizing a MAC layer will usually detect an oversized Ethernet frame and cease receiving. A bridge/switch will not forward the frame.[64]

A non-uniform frame size configuration in the network using jumbo frames may be detected as jabber by end nodes.[citation needed] Jumbo frames are not part of the official IEEE 802.3 Ethernet standard.

A packet detected as jabber by an upstream repeater and subsequently cut off has an invalid frame check sequence and is dropped.[65]

Runt frames[edit]

Runts are packets or frames smaller than the minimum allowed size. They are dropped and not propagated.[66]

AUTOSAR中定义的Ethernet Interface的API简介 (一) - 知乎

AUTOSAR中定义的Ethernet Interface的API简介 (一) - 知乎首发于00_汽车电子嵌入式软件/AUTOSAR切换模式写文章登录/注册AUTOSAR中定义的Ethernet Interface的API简介 (一)塞鸿北度汽车电子嵌入式软件工程师这篇文章简单介绍AUTOSAR中所定义的Ethernet Interface模块中所包含的API. 这篇文章参照的是R19-11版本的《Specification of Ethernet Interface》并结合个人的一些项目经验。如果你对AUTOSAR架构或者Ethernet感兴趣或者正在准备相关的面试，希望这篇文章可以帮到你。Ethernet Interface是AUTOSAR结构中的一个模块，这个模块的作用主要是: 通过控制下层的驱动软件发送或接收Ethernet frames. 并把发送和接收的结果反馈给上层软件。这里的"下层驱动软件"指的是Ethernet Driver, Ethernet Transceiver Driver 以及 Ethernet Switch Driver. "发送和接收的结果"指的是: 接收到的Ethernet Frame中的Payload中的内容，接收是否成功，发送是否完成等等。"上层软件"则指的是AUTOSAR中所定义的TCP/IP模块以及Ethernet State Manager 模块。通过下层驱动软件获得硬件的状态信息，比如link的状态以及Signal Quality等等，并简单处理。在使用VLAN ID的情况下，把有着不同的VLAN ID的Ethernet Frame分配到不同的Ethernet Controller. 主要用于多个Ethernet Controller同时使用的情况下。接下来开始介绍Ethernet Interface 中需要实现的API:EthIf_MainFunctionRx 这个API会被周期性地调用，用来在Polling模式下接收Ethernet Frame. 在这个被周期性调用的API中会检查是否有Ethernet Frame到来，一旦有Ethernet Frame 到来，会调用 Eth_Receive 去接收。Eth_Receive 是 AUTOSAR 中定义的 Ethernet Driver 中的API, 可以参考我之前的文章《AUTOSAR中定义的Ethernet Driver的API简介 (三)》来了解这个API.EthIf_MainFunctionRx 这个API会被周期性地调用，用来在Polling模式下去确认Ethernet Frame 的发送的情况。在这个周期性调用的API中，Eth_TxConfirmation 会在有Ethernet Frame被发送完成后的情况下被调用。而在Interrupt模式下，Eth_TxConfirmation 则是在发送Ethernet Frame所产生的中断的Interrupt Service Routine中被调用，可以参考我之前的文章《AUTOSAR中定义的Ethernet Driver的API简介 (四)》了解Ethernet Frame的发送过程。EthIf_MainFunctionState 这也是一个会被周期性调用的API，用于周期性地检查Link的状态以及Signal Quality. AUTOSAR定义的Ethernet Transceiver Driver中的API EthTrcv_GetLinkState EthTrcv_GetPhySignalQuality 以及Ethernet Switch Driver中的API EthSwt_GetLinkState EthSwt_GetPortSignalQuality 会被调用。(未完待续)发布于 2021-08-28 17:39嵌入式开发autosar汽车电子赞同 4添加评论分享喜欢收藏申请转载文章被以下专栏收录00_汽车电子嵌入式软件/AUTOSAR汽车电子嵌入式软件开发/AUTO

Attention Required! | Cloudflare

Please enable cookies.

Sorry, you have been blocked

You are unable to access techopedia.com

Why have I been blocked?

What can I do to resolve this?

You can email the site owner to let them know you were blocked. Please include what you were doing when this page came up and the Cloudflare Ray ID found at the bottom of this page.

Cloudflare Ray ID: 86440b7dcb920968

•

Your IP:

Click to reveal

49.157.13.121

•

Performance & security by Cloudflare

What Is Ethernet? Definition, Types, and Uses - Spiceworks

Skip to Main Navigation

Skip to Footer

Home

News & Insights

News & Insights Home

Artificial Intelligence

Innovation

IT Careers & Skills

Cloud

Cyber Security

Future of Work

IT Research

All Categories

Community

Community Home

Cloud

Collaboration

Hardware

Networking

Programming

Security

Software

Storage

Vendors

Virtualization

Windows

All Categories

Events

IT Tools

Cloud Help Desk

Inventory Online

Contracts

Connectivity Dashboard

Pricing

All IT Tools

IT Spend

Newsletters

Reviews

Join

Networking

What Is Ethernet? Definition, Types, and Uses

Ethernet helps plug a desktop or laptop into a local area network (LAN) for speedy data transmission via cables.

Chiradeep BasuMallick

Technical Writer

January 12, 2023

Ethernet is defined as a networking technology that includes the protocol, port, cable, and computer chip needed to plug a desktop or laptop into a local area network (LAN) for speedy data transmission via coaxial or fiber optic cables. This article explains the meaning of Ethernet and how it works, along with its key uses.

Table of Contents

What Is Ethernet?

10 Types of Ethernet

Key Uses of Ethernet

What Is Ethernet?

Ethernet is a networking technology that includes the protocol, port, cable, and computer chip needed to plug a desktop or laptop into a local area network (LAN) for speedy data transmission via coaxial or fiber optic cables.

Ethernet is a communication technology developed in the 1970s by Xerox that links computers in a network via a wired connection. It connects local area network (LAN) and wide area network (WAN) systems (WAN). With LAN and WAN, several devices, such as printers and laptops, may be connected across buildings, residences, and even small communities.

It provides a straightforward user interface that facilitates the connection of several devices, including switches, routers, and PCs. With a router and just a few Ethernet connections, it is possible to construct a local area network (LAN) that enables users to communicate between all connected devices. This is because laptops have Ethernet connectors, into which cables are inserted, and the other end is linked to routers.

Most Ethernet devices are compatible with Ethernet connections and devices that run at slower speeds. However, the connection speed will be determined by the weakest components.

Wireless networks have superseded Ethernet in many locations, yet the latter continues to be more prevalent for wired networking. Wired networks are more reliable and less susceptible to interference than wireless networks. This is the primary reason why so many businesses and organizations continue to adopt Ethernet.

Ethernet celebrated 25 years of existence in 1998; by that time, it had undergone several revisions as technology advanced. Ethernet is continually redesigned as its capabilities expand and evolve. Today, it is among the most widely used network technologies worldwide.

How did Ethernet evolve?

Ethernet was created in the early 1970s at the Xerox Palo Alto Research Center (PARC) by a group that included David Boggs and Robert Metcalfe. In 1983, the Institute of Electrical and Electronics Engineers (IEEE) ratified it as a standard.

Metcalfe developed the idea of Ethernet in a document he wrote for Xerox PARC in 1973, marking the beginning of Ethernet’s development. Metcalfe built Ethernet based on the Aloha system, an earlier networking initiative that started in 1968 at the University of Hawaii. Metcalfe determined in 1973 that the technology had surpassed its initial appellation, Alto Aloha Network, and rebranded it as Ethernet.

Metcalfe and Boggs, together with their colleagues at Xerox, Charles Thacker, and Butler Lampson, would successfully trademark Ethernet technology four years later.

In 1980, Xerox collaborated with Digital Equipment Corporation and Intel to create the first 10 Mbps Ethernet standard. And in the meantime, the IEEE Local and Metropolitan Area Networks (LAN/MAN) Standards Committee set out to produce an equivalent open standard. The LAN/MAN committee established an Ethernet subcommittee with the designation 802.3. The IEEE adopted the first 802.3 standards for thick Ethernet in 1983 and was published formally in 1985.

See More: What Is Software-Defined Networking (SDN)? Definition, Architecture, and Applications

How does Ethernet work?

The Ethernet protocol employs a star topology or linear bus, which is the basis for the IEEE 802.3 standard. In the OSI network structure, this protocol works bot.h the physical layer and data link layer, the first two levels. Ethernet divides the data connection layer into two distinct layers: the logical link control tier and also the medium access control (MAC) tier.

The data connection layer in a network system is primarily concerned with transmitting data packets from one node to the other. Ethernet employs an access mechanism known as CSMA/CD (Carrier Sense Multiple Access/Collision Detection) to enable each computer to listen to the connection before delivering data across the network.

Ethernet also transmits data using two components: packets and frames. The frame contains the sent data payload as well as the following:

Both the MAC and physical addresses of the sender and recipient

Error correction data for identifying transmission faults

Information on Virtual LAN (VLAN) tagging, as well as the quality of service (QoS)

Each frame is encapsulated in packets that comprise many bytes of data to set up the connection and identify the frame’s commencement point.

See More: What Is IPv6 (Internet Protocol Version 6)? Definition, Features, and Uses

Key components of an Ethernet connection

An Ethernet connection encompasses the following:

The Ethernet protocol: This protocol was developed in the 1970s by Xerox. It is a series of standards that governs how data is sent between Ethernet components as explained before.

The Ethernet port: Ethernet ports (commonly known as jacks or sockets) are openings on computer network infrastructure into which one may plug in Ethernet cables. It supports cables with RJ-45 connectors. The Ethernet connector on the majority of computers serves to connect the equipment to a wired connection. The Ethernet port of a computer is linked to an Ethernet network adapter, also known as an Ethernet card, mounted on the motherboard. A router may contain numerous Ethernet ports to support various wired network devices.

Ethernet network adapter: An Ethernet adapter is a chip or card that fits into a slot on the motherboard and allows a computer to connect to a local area network (LAN). In the past, these were always used with desktop computers. Ethernet is now integrated into the chipsets of laptop and desktop motherboards.

An Ethernet cable: Ethernet cable, often known as a network cable, links your computer to a modem, router, or network switch. The Ethernet cable consists of the RJ45 connection, the internal cabling, and a plastic jacket.

See More: What Is MPLS (Multi-Protocol Label Switching)? Definition, Working, and Architecture

10 Types of Ethernet

The key types of Ethernet connectivity are as follows:

Types of Ethernet Connections

1. Ethernet connections that use coaxial cables

A coaxial cable transmits electrical signals at high frequencies with minimal loss. Ethernet types 10Base2 and 10Base5 are now used. A copper conductor is surrounded by a dielectric insulator often constructed of PVC or Teflon. The dielectric insulator is encircled by a braided conductive metallic shield that minimizes electromagnetic interference of the metal as well as outside interference. Lastly, the metallic shield is covered with a PVC, or other fire-resistant plastic wrapping called a sheath. 10 Mbps is its highest transmission speed.

This Ethernet type can be further classified into networks that use one of the following cable types:

Tri-axial: Using an extra copper braid shield, this variant of Ethernet offers additional bandwidth and interference resistance. It is often used to link cameras and cable televisions.

RG-6: This kind of coaxial cable can be used when signal quality needs improvement. They include a thicker dielectric insulator and are employed in wireless broadband, cable television, etc.

Hardline: This cable variant is utilized in Ethernet networks that demand a strong signal. They are employed in telephone and internet connections.

2. Connections via fiber optic cables

These connections employ optical fibers with glass cores wrapped by several sheets of cladding material, often PVC or Teflon. Since it sends data as light signals, there are no interference difficulties with fiber optics.

Fiber optics can transfer signals over far greater distances than twist pairs and coaxial cables. It employs 10BaseF, 100BaseFX, 100BaseBX, 100BaseSX, 1000BaseFx, 1000BaseSX, and 1000BaseBx variations of Ethernet. Consequently, it can transmit information at a rapid speed. This Ethernet type may also be subdivided into networks using the following:

Single-mode fiber (SMF): It is utilized for long-distance communication and employs a single beam of light to deliver data.

Multi-mode fiber (MMF): It utilizes numerous light beams to convey data and is less expensive than other alternatives.

3. Ethernet connections via twisted pair cables

Twisted pair is a copper wire cable consisting of two insulated copper wires wrapped around to prevent interference and crosstalk. It employs 10BASE-T, 100BASE-T, and a few additional Ethernet variants of more recent origin. It utilizes RJ-45 plugs. This Ethernet type may be among the following variations:

Ethernets that use shielded twisted pair (STP) cables: This foil shield offers protection against interference flowing into or out of the cable. Consequently, they are used across more considerable distances and at higher transmission rates.

Ethernets that use unshielded twisted pair (UTP) cables: Unshielded twisted pair cable is now one of the most frequently deployed cables in computer networks. UTP comprises two twisted, insulated copper wires; twisting cables help limit interference.

4. Fast Ethernet

It is an Ethernet network capable of 100 Mbit/s data transmission. It may use twisted pairs or fiber optic cables. (The earlier 10 Mbit/s Ethernet is still deployed and utilized but lacks the bandwidth required for specific network video scenarios.)

Most network-connected devices, like laptops and network cameras, include a 100BASE-TX/10BASE-T Ethernet interface, often referred to as a 10/100 interface, that supports both 10 Mbit/s and Fast Ethernet. Cat-5 cable is the type of twisted pair cable which enables Fast Ethernet.

5. Gigabit Ethernet

Gigabit Ethernet, which might alternatively be based on twisted pair or fiber optic cable, provides a data transfer rate of one gigabit per second (1 Gbit/s) and is gaining in popularity. It is anticipated to supersede Fast Ethernet as the de facto norm in the near future.

Cat-5e is the kind of twisted pair cable which enables Gigabit Ethernet, in which all four types of twisted wires are used to accomplish high data speeds. Cat-5e cables or higher are suggested for networked video systems. Most interfaces are interoperable with 10 and 100 Mbit/s Ethernet and therefore are frequently referred to as 10/100/1000 interfaces.

See More: What is an Intranet? Meaning, Features, and Best Practices

6. 10 Gigabit Ethernet

The newest iteration of Ethernet, 10 Gigabit Ethernet, offers a data throughput of 10 Gbit/s (10,000 Mbit/s) via an optic fiber or twisted pair connection. 10GBASE-LX4, 10GBASE-ER, or 10GBASE-SR built on an optical fiber connection could reach up to 10,000 meters in distance (6.2 miles). The twisted pair option requires a cable of exceptional quality (Cat-6a or Cat-7). Ethernet 10 Gbit/s is mainly utilized for backbone networks in high-end operations that demand significant data speeds.

7. Switch-based Ethernet

This network configuration includes a hub or a switch. In addition, a standard network cable is employed as opposed to a twisted pair cable. A network switch’s primary role is to transfer information/data from one device to another on the same network. Consequently, a network switch efficiently completes this operation since data is transported from one machine to another without harming other networking hardware within the same environment.

This form of Ethernet network has a star topology centered on a switch. A network switch employs a filtering and switching process comparable to gateways, where these methods have been around for an extended period.

8. Wired Ethernet, which uses cables

This is the most prevalent type of wired LAN or WAN communication. A modem is directly attached to an Ethernet cable, and the cable’s opposite end is linked to a machine (laptop or desktop). This cable needs to be at least Cat5 or above. Due to the direct connection, the speed is also much higher than wireless networks. In reality, this is an excellent Internet connection choice for individual users.

This is also feasible for several users, like in a small company network. One to fifteen devices may be connected to such a network across a range of up to 10 kilometers. While wired Ethernet is virtually extinct, it is still advantageous for smaller groups since it is considerably faster and more secure than wireless networks and can load and transmit large amounts of data, such as films and audio, and live stream them without interruption.

9. Wireless Ethernet – i.e., without cables

A wireless network relies on high-frequency radio signals and does not require cables to connect a receiving device, such as a laptop, to the network. In this method, often known as Wi-Fi, data is transferred using wireless signals instead of a cable. Consequently, it is more adaptable than wired networks, and the device will connect if it is within a certain range or on the router and modem’s periphery.

If a modem and a router are present, one must connect the modem to the router via a category 5 (Cat5) or category 6 (Cat6) Ethernet connection. The item that is virtually linked receives a signal from the routers. This network is simple to set up, although there may be wifi signal concerns.

10. SOHO Ethernet LAN

SOHO refers to a tiny office or home office. This is the simplest Ethernet LAN configuration. To construct this LAN, an Ethernet LAN Switch is utilized. Ethernet LAN Switches have several ports. An Ethernet cable links an endpoint or user device to one of these ports.

Today, Internet connectivity is an essential component of every network. To take advantage of this requirement, suppliers currently offer integrated networking connections that function as both routers and Ethernet switches. These devices typically contain four-eight LAN access points. Additionally, specific variants have wireless LAN entry (or access) points.

Key Uses of Ethernet

Ethernet is now a near-ubiquitous technology in today’s hyper-connected digital world. This is because it:

Uses of Ethernet

Improves consumer internet experiences: When their wireless Wi-Fi data connection speed is insufficient, many in their homes deploy Ethernet connections. Ethernet is typically used to link several devices in a local area network (LAN) and a wide area network (WAN).

Offers high bandwidth connections: Ethernet offers data transfer rates of 10, 100, 1000, 10000, 40000, and 100000 megabits per second (Mbps). When Ethernet was originally created, bands were defined in megabits per second (Mbps), but they are currently calculated in gigabits per second (Gbps).

Provides different options of speed, based on budget, region, and requirements: Standard Ethernet’s top speed is 10Mbps, whereas fast Ethernet’s is 100Mbps, Gigabit Ethernet’s is 1Gbps, while 10 Gigabit Ethernet is 10Gbps.

Strikes a balance between cost and performance: Ethernet is widely used due to its affordable price and compatibility with any subsequent network device. Ethernet speed was approximately 10Mbps in 1983 and now exceeds 400Gbps. Ethernet is extensively used by companies, hospitals, schools, universities, and gamers due to its fast speed, network security, and dependability.

Amplifies the capabilities of your Wi-Fi network: In recent years, Wi-Fi has become increasingly popular. Wi-Fi has improved speeds and offered extensive coverage due to technological improvements. Wi-Fi transmissions can only simultaneously support a limited number of devices. In older buildings with frequent Wi-Fi dead zones, Ethernet connections are essential.

Enforces greater security: Ethernet has the advantage of being more secure than Wi-Fi. Anyone within a Wi-Fi hotspot’s range may access data transferred over the radio. Because radio signals deliver the information, it is vulnerable to theft. In contrast, data supplied by Ethernet can only be accessible on the local area network.

Supports direct current (DC) power transmission: As its names suggest, Power over Ethernet (POE) is the provisioning of energy supply over Ethernet connections. It powers many devices, including CCTV cameras and wireless access points. One of the primary advantages of Power over Ethernet is that a distinct power source is unnecessary. This is especially useful for placing devices in locations that are far from the nearest power source.

See More: Wifi 5 vs. Wifi 6: Understanding the 10 Key Differences

Takeaway

Even in the era of high-speed wireless connectivity – particularly with the emergence of Wi-Fi 6 – Ethernet remains relevant. For many regions, it is still the best way to get Internet access, and most homes have an Ethernet connection linked to their router or hub. The market for Ethernet switches is constantly growing, despite being around for many years. For enterprises, Ethernet forms a crucial part of the networking infrastructure. By understanding how Ethernet works, you can optimize the power of wired internet connections

Did you find this article helpful in learning about Ethernet? Tell us on FacebookOpens a new window , TwitterOpens a new window , and LinkedInOpens a new window . We’d love to hear from you.

Follow

imtoken官网正版下载|ethernet interface

imtoken官网正版下载|ethernet interface

以太网（Ethernet） - 知乎

Ethernet接口和Interface：深入了解接口的结构和功能 - 技象科技

以太网知识01 Media Independent Interface （MII) 媒体独立接口 - 知乎

高性能网络 — SmartNIC、DPU 设备演进与运行原理 - 知乎

Attention Required! | Cloudflare

Network 之二 Ethernet（以太网）中的 MAC、MII、PHY 详解_ethernet phy-CSDN博客

Ethernet接口与Interface - 技象科技

Ethernet - Wikipedia

AUTOSAR中定义的Ethernet Interface的API简介 (一) - 知乎

Attention Required! | Cloudflare

What Is Ethernet? Definition, Types, and Uses - Spiceworks

最近的新闻

您可能喜欢的文章

tokenpocket最新版本下载|怎么给ethernet设备分配名称

im钱包官方下载|ethics of his conduct

TP钱包资金池玩法介绍及操作指南

忘记比特币钱包私钥怎么办？恢复和解决方法详

TP钱包自定义排序教程：简单操作一步搞定

哥伦比亚人喜欢使用哪种区块链钱包？

TP钱包的币多久能到账？一文告诉你