WO2011057447A1 - Router and cluster router - Google Patents

Router and cluster router Download PDF

Info

Publication number
WO2011057447A1
WO2011057447A1 PCT/CN2009/074937 CN2009074937W WO2011057447A1 WO 2011057447 A1 WO2011057447 A1 WO 2011057447A1 CN 2009074937 W CN2009074937 W CN 2009074937W WO 2011057447 A1 WO2011057447 A1 WO 2011057447A1
Authority
WO
WIPO (PCT)
Prior art keywords
infmiband
router
infiniband
switch
network
Prior art date
Application number
PCT/CN2009/074937
Other languages
French (fr)
Chinese (zh)
Inventor
梁冰
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2009/074937 priority Critical patent/WO2011057447A1/en
Priority to CN2009801476192A priority patent/CN102150400A/en
Publication of WO2011057447A1 publication Critical patent/WO2011057447A1/en

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/35—Switches specially adapted for specific applications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/35—Switches specially adapted for specific applications
    • H04L49/356—Switches specially adapted for specific applications for storage area networks
    • H04L49/358—Infiniband Switches
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/60—Software-defined switches
    • H04L49/602—Multilayer or multiprotocol switching, e.g. IP switching

Definitions

  • the invention relates to the field of data communication, in particular to a router and a cluster router based on Infmiband technology.
  • Ethernet technology has basically covered LAN, MAN and WAN.
  • Ethernet is never able to achieve zero packet loss, especially when the network is crowded, the latency is also very large.
  • Infmiband technology is a general-purpose network protocol, which is characterized by extremely low delay and zero packet loss rate. If multiple servers are networked using Infmiband technology, you can achieve extremely high parallel computing performance.
  • the interfaces provided by current core routers include Ethernet (including 10GE, 10 Gigabit Ethernet/GE, Gigabit Ethernet/FE, Fast Ethernet, etc.), ATM (Asynchronous Transfer Mode), etc.
  • Ethernet including 10GE, 10 Gigabit Ethernet/GE, Gigabit Ethernet/FE, Fast Ethernet, etc.
  • ATM Asynchronous Transfer Mode
  • the bandwidth on the Ethernet side can reach 1G to 10G
  • the bandwidth on the Infmiband network side can reach 40G to 160G.
  • the bandwidth of the Ethernet interface of the gateway device is often smaller than the bandwidth of the Infmiband interface. The device must become the bottleneck for communication between Ethernet and Infmiband networks.
  • the cluster router solution is proposed in the prior art, as shown in FIG. 2, two or more devices are to be used.
  • Ordinary core routers a common core router can be used as a forwarding box
  • Logical router because the interconnect technologies provided by different vendors are different, they are all based on proprietary protocols. Therefore, cluster routers composed of devices provided by different vendors may have incompatibility problems (such as center frame and different Between the forwarding boxes), which causes bottlenecks in network node forwarding performance.
  • the embodiment of the present invention provides a router, which can solve the problem of network bandwidth bottleneck that occurs when an Infmiband network is connected to an IP backbone network.
  • a router includes a switch chip and at least two line cards connected thereto, wherein the at least two line cards include an Infmiband line card supporting Infmiband technology.
  • the router provided by the embodiment of the present invention implements the Infmiband line card supporting the Infmiband technology in the router, so that the conversion between the Ethernet technology and the Infmiband technology is implemented inside the router, bypassing the bandwidth bottleneck caused by the external gateway device, thereby solving the problem.
  • the embodiment of the present invention further provides a cluster router, which can solve the problem that the network node forwarding performance bottleneck is caused by the incompatibility between the devices existing in the cluster router.
  • a cluster router includes a switch and at least two routers, wherein
  • the switch is an Infmiband switch supporting Infmiband technology
  • the router includes a switch chip, at least two line cards connected to the switch chip,
  • the switch chip is a switch chip supporting a proprietary protocol
  • the router also includes a cascading card connected between the Infmiband switch and the switching chip for performing a transfer between a signal in the Infmiband format and a signal in a proprietary protocol format.
  • the cluster router provided by the embodiment of the present invention replaces the existing private protocol-based switching technology by using the standard Infmiband technology, and converts the signals sent by the routers based on the private protocol into a unified Infmiband format signal, and then performs the signal. Communication between each router and the switch, thereby avoiding the problem of incompatibility between the internal devices of the cluster router caused by the proprietary protocol, and also solving the problem of network node forwarding performance bottleneck.
  • FIG. 1 is a schematic diagram of a structure in which an Infmiband network accesses an IP backbone network through a gateway device in the prior art;
  • FIG. 2 is a schematic structural diagram of a cluster router in the prior art
  • FIG. 3 is a schematic structural diagram of a router according to Embodiment 1 of the present invention.
  • FIG. 4 is a schematic structural diagram of an Infmiband line card in a router according to Embodiment 1 of the present invention
  • FIG. 5 is a schematic structural diagram of an Infmiband network accessing an IP backbone network by using a router provided in Embodiment 1 of the present invention
  • FIG. 6 is a schematic structural diagram of a cluster router according to Embodiment 2 of the present invention.
  • FIG. 7 is a schematic structural diagram of a single router in Embodiment 2 of the present invention.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the router provided in this embodiment includes a switch chip 31 and at least two line cards connected to the switch chip 31 through the backplane 32.
  • the at least two line cards include Infmiband support.
  • the above routers implement the Infmiband line card supporting Infmiband technology inside the router, so that the conversion between Ethernet technology and Infmiband technology is implemented inside the router, bypassing the bandwidth bottleneck caused by the external gateway device, thus solving the Infmiband network and IP.
  • the network bandwidth bottleneck that occurs when the backbone network is connected.
  • the Infmiband line card 33 in the above router can be implemented by the following structure:
  • the Infmiband line card 33 includes: an Infmiband optical module 331, an Infmiband framing (Framer) module 332, a network processor (NP) 333, and a Fabric Interface Controller (FIC) 334;
  • an Infmiband optical module 331 an Infmiband optical module 331, an Infmiband framing (Framer) module 332, a network processor (NP) 333, and a Fabric Interface Controller (FIC) 334;
  • NP network processor
  • FIC Fabric Interface Controller
  • the Infmiband optical module is connected to an external Infmiband network for photoelectrically converting the received code stream.
  • the Infmiband framing module 332 is configured to parse the received Infmiband code stream into a message, or encapsulate the received message into an Infmiband code stream;
  • the network processor 333 is configured to obtain a destination address of the packet according to the content of the packet, and exchange the access chip 334, and connect to the switch chip 31, for reassembling, storing, and sorting the received packet. .
  • the Infmiband optical module 331 converts the received Infmiband format optical code stream. Generating the code stream, and then parsing the code stream into a message by the Infmiband framing module 332; then, the network processor 333 obtains the destination address of the message from the content of the message, that is, the Which port of the line card is to be forwarded to which the message is to be forwarded; the switching access chip 334 reassembles, stores and sorts the text and sends it to the switch chip 31, which is used by the switch chip 31. Good text is sent to the destination line card.
  • the switching access chip 334 When the signals of other formats are converted into signals of the Infmiband format and sent out through the Infmiband line card, the switching access chip 334 first receives the message sent by the switching chip 31, and reassembles the message. And storing, sorting, and sending to the network processor 333.
  • the network processor 333 determines, according to the content of the message, which port of the line card is required for the message to be output. Then, the Infmiband framing module 332 encapsulates the message.
  • the Infmiband code stream is sent to the corresponding port according to the destination port acquired by the network processor 333. Then, the Infmiband code stream is converted into the Infmiband code stream by the Infmiband optical module 331 and passed through the Infmiband interface. Send it out.
  • the network structure of the Infmiband network connected to the IP backbone network through the router is as shown in FIG. 5; the router here has a traditional Ethernet interface POS interface/ATM interface for connecting IP.
  • the Infmiband interface is also provided to connect to the private network such as the HPC, and the Infmiband network is directly connected to the IP backbone network through the router.
  • the router provided in the embodiment of the present invention belongs to the core router, and the switching capability thereof can be achieved. T (10 12 ) level, so it can completely avoid the problem of network bandwidth bottleneck caused by existing network equipment.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • the cluster router provided in this embodiment includes a switch 61 and at least two routers 62;
  • the switch 61 is an Infmiband switch supporting Infmiband technology;
  • the router 62 further includes: a switch chip 621, at least two line cards 622 connected to the switch chip 621, the switch chip 621 is a switch chip supporting a private protocol, wherein the private protocol refers to different vendors Development and adoption of communication standards;
  • the router 62 further includes a cascading card 623 connected between the Infmiband switch 61 and the switch chip 621 for performing switching between signals in the Infmiband format and signals in the private protocol format.
  • the above-mentioned cluster routers use the standard Infmiband technology to replace the existing private protocol-based switching technology, and convert the signals sent by the routers based on the private protocol into a unified Infmiband format signal, and then perform various routers and switches. Inter-communication, so as to avoid the incompatibility between the internal devices of the cluster router caused by the proprietary protocol, and also solve the problem of network node forwarding performance bottleneck.
  • the router 62 includes: a switch chip 621 and at least two line cards 622 connected thereto, and an Infmiband framer 6231 and at least one Infmiband optical module connected to the Infmiband framer 6231. 6232, where the combination of the Infmiband framer 6231 and the Infmiband optical module 6232 is the above-mentioned cascade card 623;
  • the Infmiband framer 6231 is configured to convert the received signal in the Infmiband format into a signal of a proprietary protocol format supported by the current router, or convert the received signal in the proprietary protocol format into a signal in the Infmiband format;
  • the Infmiband optical module 6232 is connected between the Infmiband switch 61 and the Infmiband framer 6231 for performing photoelectric conversion.
  • the Infmiband optical module 6232 is connected to the Infmiband switch 61 through an optical fiber.
  • the signal transmission process is as follows: One of the routers 62 connected to the Infmiband switch 61 receives a signal based on the private protocol 1 through one of the line cards 622 it contains, the private protocol 1 being a private communication standard supported by the current router; the line card 622 The signal is transmitted to the Infiniband framer 6231 via the switch chip 621, and the signal based on the proprietary protocol 1 is converted into an electrical signal of the Infmiband format by the Infiniband framer 6231, and then the Infmiband optical module 6232 outputs the electrical signal in the Infiniband format. Converting the optical signal into the Infiniband format and transmitting it to the switch 61 through the optical fiber;
  • the switch forwards the optical signal of the Infmiband format to another router 62 according to the destination address information carried in the optical signal of the Infmiband format; the Infiniband optical module 6232 of the other router 62 receives the Infiniband format.
  • the optical signal is converted into an electrical signal in the Infiniband format, and the electrical signal of the Infmiband format is converted into a signal based on the proprietary protocol 2 by the Infmiband framer 6231, which is the private communication supported by the other router 62. Standard;
  • the signal based on the proprietary protocol 2 is then passed through the switch chip 621 to a line card 622 connected thereto.
  • the cluster router provided by the embodiment of the present invention replaces the existing private protocol-based switching technology by using the standard Infmiband technology, and converts the signals sent by the routers based on the private protocol into the unified Infiniband format signal, and then performs the signal. Communication between each router (that is, the forwarding frame in the cluster router) and the switch (that is, the central box in the cluster router), thereby avoiding the problem of incompatibility between the internal devices of the cluster router due to the proprietary protocol, and also solving the problem. Network nodes forward performance bottlenecks.
  • the cost of a typical cluster router is mostly used for interconnected physical layer optical modules. Since traditional cluster routers use optical modules based on proprietary protocols, they cannot be mass-produced uniformly, and the industrial chain cannot be formed, resulting in an expensive optical module portion of the router.
  • the present invention uses the standard in the Infmiband network. Optical modules are interconnected and are much lower in cost than traditional cluster routers. Moreover, since the cluster router provided in this embodiment is no longer limited by the proprietary protocol, the scale that can be implemented is much larger than that of the traditional cluster router.
  • ROM/RAM read-only memory
  • diskette diskette
  • optical disk etc.
  • a computer device which may be a personal computer, server, or network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The embodiments of the present invention provide a router, which relates to the field of data communications and can resolve the problem of network bandwidth bottleneck appearing when an InfiniBand network connects to an IP backbone network. The router includes one switch chip and at least two line cards connected to the switch chip, wherein said at least two line cards include an InfiniBand line card supporting InfiniBand technology. The embodiments of the present invention also provide a cluster router, which can resolve the problem of network nodes forwarding performance bottleneck caused by the non-compatibility among the devices existing inside the cluster router.

Description

路由器及集群路由器 技术领域  Routers and cluster routers
本发明涉及数据通讯领域 ,尤其涉及一种基于 Infmiband技术的路由器及 集群路由器。  The invention relates to the field of data communication, in particular to a router and a cluster router based on Infmiband technology.
背景技术 Background technique
网络发展到今天， 以太网技术基本上已经完全覆盖了局域网、 城域网和 广域网。 然而， 以太网始终无法实现零丢包， 尤其当网络拥挤时， 延迟也会 非常大。 虽然对于很多传统业务， 比如收发邮件、 浏览网页等， 其应用不会 受到很大影响； 但是对于今后的业务， 比如视频点播、 互动游戏等， 延迟和 丢包会影响到用户的体验。  With the development of the network to this day, Ethernet technology has basically covered LAN, MAN and WAN. However, Ethernet is never able to achieve zero packet loss, especially when the network is crowded, the latency is also very large. Although for many traditional services, such as sending and receiving mail, browsing the web, etc., its application will not be greatly affected; but for future business, such as video on demand, interactive games, etc., delay and packet loss will affect the user experience.
而且， 在某些用于连接特殊设备（比如网络存储器、 高性能服务器等） 的专网领域， 由于有其特殊的要求， 而以太网技术不能满足其需求， 因此还 需要有其他的技术存在， 比如基于 Fiber-Channel (光纤通道）技术的网络存 储方案以及基于 Infmiband技术的 HPC ( High Performance Compute, 高性能 计算） 网络仍然占据相当大的市场。  Moreover, in some private network areas for connecting special devices (such as network storage, high-performance servers, etc.), because of their special requirements, and Ethernet technology can not meet their needs, other technologies are needed. For example, network storage solutions based on Fiber-Channel technology and HPC (High Performance Compute) networks based on Infmiband technology still occupy a considerable market.
其中， Infmiband技术是一种通用的网络协议， 其显著特征就是极低的延 迟和零丢包率。 如果多个服务器采用 Infmiband技术进行联网， 则可以获得极 高的并行计算性能。  Among them, Infmiband technology is a general-purpose network protocol, which is characterized by extremely low delay and zero packet loss rate. If multiple servers are networked using Infmiband technology, you can achieve extremely high parallel computing performance.
然而， 目前的核心路由器提供的接口包括以太网 （包括 10GE, 万兆以太 网 /GE , 千兆以太网/ FE , 快速以太网等）、 ATM ( Asynchronous Transfer Mode , 异步传输模式）等， 至今还没有 Infmiband技术对应的接口， 因此 IP ( Internet Protocol, 因特网协议）骨干网无法通过核心路由器直接与 Infmiband 网络建 立连接， 而只能通过一个外置的网关设备进行转接。 然而， 如图 1 所示， 以 太网侧的带宽可以达到 1G到 10G, 而 Infmiband网络侧的带宽可以达到 40G 到 160G , 网关设备的以太接口的带宽往往小于 Infmiband接口的带宽， 网关 设备必然成为以太网与 Infmiband网络之间进行通信的瓶颈。 However, the interfaces provided by current core routers include Ethernet (including 10GE, 10 Gigabit Ethernet/GE, Gigabit Ethernet/FE, Fast Ethernet, etc.), ATM (Asynchronous Transfer Mode), etc. There is no interface corresponding to the Infmiband technology, so the IP (Internet Protocol) backbone network cannot directly establish a connection with the Infmiband network through the core router, but can only be transferred through an external gateway device. However, as shown in Figure 1, the bandwidth on the Ethernet side can reach 1G to 10G, and the bandwidth on the Infmiband network side can reach 40G to 160G. The bandwidth of the Ethernet interface of the gateway device is often smaller than the bandwidth of the Infmiband interface. The device must become the bottleneck for communication between Ethernet and Infmiband networks.
此外， 由于单设备的核心路由器已经无法满足带宽需求， 因此为了緩解 曰益增长的网络带宽要求， 现有技术中提出了集群路由器的方案， 如图 2所 示， 即将两台或者两台以上的普通核心路由器（一台普通核心路由器可以作 为一个转发框）通过特殊的技术内部互联起来， 使得这些核心路由器能够进 行设备间协同工作和并行处理， 实现***容量的平滑扩展， 并且对外只表现 为一台逻辑路由器。 但是， 由于不同厂商提供的集群路由器所采用的互联技 术各不相同， 都是基于私有协议， 因此由不同厂商提供的设备来组成的集群 路由器内部可能会存在无法兼容的问题（比如中心框和不同的转发框之间）， 导致出现网络节点转发性能的瓶颈。  In addition, since the core router of a single device cannot meet the bandwidth requirement, in order to alleviate the network bandwidth requirement for increasing profit, the cluster router solution is proposed in the prior art, as shown in FIG. 2, two or more devices are to be used. Ordinary core routers (a common core router can be used as a forwarding box) are interconnected through special technologies, enabling these core routers to work together and parallelize devices to achieve a smooth expansion of system capacity. Logical router. However, because the interconnect technologies provided by different vendors are different, they are all based on proprietary protocols. Therefore, cluster routers composed of devices provided by different vendors may have incompatibility problems (such as center frame and different Between the forwarding boxes), which causes bottlenecks in network node forwarding performance.
发明内容 Summary of the invention
本发明的实施例提供一种路由器， 能够解决 Infmiband网络与 IP骨干网 进行连接时所出现的网絡带宽瓶颈的问题。  The embodiment of the present invention provides a router, which can solve the problem of network bandwidth bottleneck that occurs when an Infmiband network is connected to an IP backbone network.
为达到上述目的 , 本发明的实施例采用如下技术方案：  In order to achieve the above object, the embodiment of the present invention adopts the following technical solutions:
一种路由器， 包括一个交换芯片以及与其相连的至少两个线卡， 其中， 所述至少两个线卡中包括支持 Infmiband技术的 Infmiband线卡。  A router includes a switch chip and at least two line cards connected thereto, wherein the at least two line cards include an Infmiband line card supporting Infmiband technology.
本发明实施例提供的路由器，通过在路由器内部设置支持 Infmiband技术 的 Infmiband线卡， 使以太技术与 Infmiband技术之间的转换在路由器内部实 现，绕开了外部网关设备所造成的带宽瓶颈，从而解决了现有技术中 Infmiband 网络与 IP骨干网进行连接时所出现的网络带宽瓶颈的问题。  The router provided by the embodiment of the present invention implements the Infmiband line card supporting the Infmiband technology in the router, so that the conversion between the Ethernet technology and the Infmiband technology is implemented inside the router, bypassing the bandwidth bottleneck caused by the external gateway device, thereby solving the problem. The problem of the network bandwidth bottleneck that occurs when the Infmiband network is connected to the IP backbone network in the prior art.
本发明的实施例还提供一种集群路由器， 能够解决集群路由器内部存在 的设备间无法兼容而导致网络节点转发性能瓶颈的问题。  The embodiment of the present invention further provides a cluster router, which can solve the problem that the network node forwarding performance bottleneck is caused by the incompatibility between the devices existing in the cluster router.
为达到上述目的， 本发明的实施例采用如下技术方案：  In order to achieve the above object, the embodiment of the present invention adopts the following technical solutions:
一种集群路由器， 包括一个交换机和至少两个路由器， 其中，  A cluster router includes a switch and at least two routers, wherein
所述交换机为支持 Infmiband技术的 Infmiband交换机；  The switch is an Infmiband switch supporting Infmiband technology;
所述路由器包括一个交换芯片、 与该交换芯片相连的至少两个线卡， 所 述交换芯片为支持私有协议的交换芯片； The router includes a switch chip, at least two line cards connected to the switch chip, The switch chip is a switch chip supporting a proprietary protocol;
该路由器还包括级联卡，该级联卡连接在所述 Infmiband交换机和所述交 换芯片之间，用于完成 Infmiband格式的信号与私有协议格式的信号之间的转 接。  The router also includes a cascading card connected between the Infmiband switch and the switching chip for performing a transfer between a signal in the Infmiband format and a signal in a proprietary protocol format.
本发明实施例提供的集群路由器，利用标准的 Infmiband技术来取代现有 的基于私有协议的交换技术， 将各个基于私有协议的路由器要向外发送的信 号转换成统一的 Infmiband格式的信号，然后进行各个路由器与交换机之间的 通信， 从而避免由于私有协议而导致的集群路由器内部设备间无法兼容的问 题， 同时也就解决了网絡节点转发性能瓶颈的问题。  The cluster router provided by the embodiment of the present invention replaces the existing private protocol-based switching technology by using the standard Infmiband technology, and converts the signals sent by the routers based on the private protocol into a unified Infmiband format signal, and then performs the signal. Communication between each router and the switch, thereby avoiding the problem of incompatibility between the internal devices of the cluster router caused by the proprietary protocol, and also solving the problem of network node forwarding performance bottleneck.
附图说明 DRAWINGS
为了更清楚地说明本发明实施例或现有技术中的技术方案， 下面将对实 施例描述中所需要使用的附图作筒单地介绍， 显而易见地， 下面描述中的附 图仅仅是本发明的一些实施例， 对于本领域普通技术人员来讲， 在不付出创 造性劳动性的前提下 , 还可以根据这些附图获得其他的附图。  In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments will be briefly described below. Obviously, the drawings in the following description are merely the present invention. Some of the embodiments can be obtained by those skilled in the art from the drawings without any inventive labor.
图 1为现有技术中 Infmiband网络通过网关设备接入 IP骨干网的结构示 意图；  1 is a schematic diagram of a structure in which an Infmiband network accesses an IP backbone network through a gateway device in the prior art;
图 2为现有技术中集群路由器的结构示意图；  2 is a schematic structural diagram of a cluster router in the prior art;
图 3为本发明实施例一中的路由器的结构示意图；  3 is a schematic structural diagram of a router according to Embodiment 1 of the present invention;
图 4为本发明实施例一中的路由器中 Infmiband线卡的结构示意图； 图 5为 Infmiband网络通过本发明实施例一中提供的路由器接入 IP骨干 网的结构示意图；  4 is a schematic structural diagram of an Infmiband line card in a router according to Embodiment 1 of the present invention; FIG. 5 is a schematic structural diagram of an Infmiband network accessing an IP backbone network by using a router provided in Embodiment 1 of the present invention;
图 6为本发明实施例二中的集群路由器的结构示意图；  6 is a schematic structural diagram of a cluster router according to Embodiment 2 of the present invention;
图 7为本发明实施例二中的单台路由器的结构示意图。  FIG. 7 is a schematic structural diagram of a single router in Embodiment 2 of the present invention.
具体实施方式 detailed description
下面将结合本发明实施例中的附图， 对本发明实施例中的技术方案进行 清楚、 完整地描述， 显然， 所描述的实施例仅仅是本发明一部分实施例， 而 不是全部的实施例。 基于本发明中的实施例， 本领域普通技术人员在没有作 出创造性劳动前提下所获得的所有其他实施例， 都属于本发明保护的范围。 The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and Not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.
下面结合附图对本发明实施例提供的路由器及集群路由器进行详细描 述。  The router and the cluster router provided by the embodiments of the present invention are described in detail below with reference to the accompanying drawings.
实施例一：  Embodiment 1:
如图 3所示， 本实施例提供的路由器， 包括一个交换芯片 31 , 以及通过 背板 32与该交换芯片 31相连的至少两个线卡； 其中， 所述至少两个线卡中 包括支持 Infmiband技术的 Infmiband线卡 33。  As shown in FIG. 3, the router provided in this embodiment includes a switch chip 31 and at least two line cards connected to the switch chip 31 through the backplane 32. The at least two line cards include Infmiband support. Technical Infmiband line card 33.
上述路由器， 通过在路由器内部设置支持 Infmiband技术的 Infmiband线 卡， 使以太技术与 Infmiband技术之间的转换在路由器内部实现， 绕开了外部 网关设备所造成的带宽瓶颈， 因此可以解决 Infmiband网络与 IP骨干网进行 连接时所出现的网络带宽瓶颈的问题。  The above routers implement the Infmiband line card supporting Infmiband technology inside the router, so that the conversion between Ethernet technology and Infmiband technology is implemented inside the router, bypassing the bandwidth bottleneck caused by the external gateway device, thus solving the Infmiband network and IP. The network bandwidth bottleneck that occurs when the backbone network is connected.
进一步地， 如图 4所示， 上述路由器中的 Infmiband线卡 33 , 可以通过 以下结构来实现：  Further, as shown in FIG. 4, the Infmiband line card 33 in the above router can be implemented by the following structure:
所述 Infmiband线卡 33 包括： Infmiband光模块 331、 Infmiband成帧 ( Framer )模块 332、 网络处理器（ Network Processor, NP ) 333及交换接入 芯片 （Fabric Interface Controller, FIC ) 334; 其中，  The Infmiband line card 33 includes: an Infmiband optical module 331, an Infmiband framing (Framer) module 332, a network processor (NP) 333, and a Fabric Interface Controller (FIC) 334;
Infmiband光模块 331， 与外部 Infmiband网络连接，用于对接收到的码流 进行光电转换；  The Infmiband optical module is connected to an external Infmiband network for photoelectrically converting the received code stream.
Infmiband成帧模块 332,用于将接收到的 Infmiband码流解析成报文，或 者将接收到的报文封装成 Infmiband码流；  The Infmiband framing module 332 is configured to parse the received Infmiband code stream into a message, or encapsulate the received message into an Infmiband code stream;
网络处理器 333， 用于根据所述报文的内容获取到报文的目的地址； 交换接入芯片 334, 与所述交换芯片 31连接， 用于对接收到的报文进行 重组、 存储和排序。  The network processor 333 is configured to obtain a destination address of the packet according to the content of the packet, and exchange the access chip 334, and connect to the switch chip 31, for reassembling, storing, and sorting the received packet. .
具体地，在外部的 Infmiband格式的信号通过所述 Infmiband线卡 33输入 到路由器时， Infmiband光模块 331将接收到的 Infmiband格式的光码流转换 成电码流， 然后由 Infmiband成帧模块 332将所述电码流解析成报文； 之后， 网络处理器 333根据所述报文的内容， 从中获取到所述报文的目的地址， 也 就是所述报文将要被转发到哪个线卡的哪个端口进行输出；交换接入芯片 334 则将所述 文进行重組、 存储和排序后发送给所述交换芯片 31， 由所述交换 芯片 31将所述 4良文发送往目的线卡。 Specifically, when an external Infmiband format signal is input to the router through the Infmiband line card 33, the Infmiband optical module 331 converts the received Infmiband format optical code stream. Generating the code stream, and then parsing the code stream into a message by the Infmiband framing module 332; then, the network processor 333 obtains the destination address of the message from the content of the message, that is, the Which port of the line card is to be forwarded to which the message is to be forwarded; the switching access chip 334 reassembles, stores and sorts the text and sends it to the switch chip 31, which is used by the switch chip 31. Good text is sent to the destination line card.
而在将其他格式的信号转换成 Infmiband格式的信号并通过 Infmiband线 卡向外发送时， 首先所述交换接入芯片 334接收到所述交换芯片 31发送来的 报文， 并对报文进行重组、存储和排序后发送给网络处理器 333; 网络处理器 333根据所述报文的内容， 判断出所迷报文需要过线卡的哪个端口进行输出； 然后， Infmiband成帧模块 332将报文封装成 Infmiband格式的电码流， 并根 据网络处理器 333获取到的目的端口将 Infmiband电码流发送到对应的端口； 之后 , 由 Infmiband光模块 331将 Infmiband电码流转换成 Infmiband光码流， 并通过 Infmiband接口发送出去。  When the signals of other formats are converted into signals of the Infmiband format and sent out through the Infmiband line card, the switching access chip 334 first receives the message sent by the switching chip 31, and reassembles the message. And storing, sorting, and sending to the network processor 333. The network processor 333 determines, according to the content of the message, which port of the line card is required for the message to be output. Then, the Infmiband framing module 332 encapsulates the message. The Infmiband code stream is sent to the corresponding port according to the destination port acquired by the network processor 333. Then, the Infmiband code stream is converted into the Infmiband code stream by the Infmiband optical module 331 and passed through the Infmiband interface. Send it out.
需要注意的是,在上述方案的实现过程中，将网络处理器 333与 Infmiband 成帧模块 332进行连接以后， 还需要对网络处理器 333中的软件部分进行升 级， 这样网络处理器 333和 Infmiband成帧模块 332才能更好地兼容。  It should be noted that, in the implementation of the foregoing solution, after the network processor 333 is connected to the Infmiband framing module 332, the software part of the network processor 333 needs to be upgraded, so that the network processor 333 and the Infmiband are Frame module 332 is better compatible.
利用本实施例中提供的路由器，将 Infmiband网络通过所述路由器连入 IP 骨干网后的网络结构如图 5 所示； 这里的路由器对外除了具有传统的以太接 口 POS接口 /ATM接口用以连接 IP骨干网外， 还提供了 Infmiband接口用以 连接 HPC等专用网络， 实现了将 Infmiband网络直接通过路由器接入 IP骨干 网； 由于本发明实施例中所提供的路由器属于核心路由器， 其交换能力可达 到 T ( 1012 )级， 因此完全可以避免现有的网络设备所带来的网络带宽瓶颈的 问题。 With the router provided in this embodiment, the network structure of the Infmiband network connected to the IP backbone network through the router is as shown in FIG. 5; the router here has a traditional Ethernet interface POS interface/ATM interface for connecting IP. In addition to the backbone network, the Infmiband interface is also provided to connect to the private network such as the HPC, and the Infmiband network is directly connected to the IP backbone network through the router. The router provided in the embodiment of the present invention belongs to the core router, and the switching capability thereof can be achieved. T (10 12 ) level, so it can completely avoid the problem of network bandwidth bottleneck caused by existing network equipment.
实施例二：  Embodiment 2:
如图 6所示， 本实施例提供的集群路由器， 包括一个交换机 61和至少两 个路由器 62; 其中， 所述交换机 61为支持 Infmiband技术的 Infmiband交换机； As shown in FIG. 6, the cluster router provided in this embodiment includes a switch 61 and at least two routers 62; The switch 61 is an Infmiband switch supporting Infmiband technology;
所述路由器 62进一步包括： 一个交换芯片 621、 与该交换芯片 621相连 的至少两个线卡 622, 所述交换芯片 621为支持私有协议的交换芯片，其中所 述私有协议指的是不同厂商内部发展、 采用的通信标准；  The router 62 further includes: a switch chip 621, at least two line cards 622 connected to the switch chip 621, the switch chip 621 is a switch chip supporting a private protocol, wherein the private protocol refers to different vendors Development and adoption of communication standards;
该路由器 62还包括级联卡 623， 该级联卡 623连接在所述 Infmiband交 换机 61和所述交换芯片 621之间， 用于完成 Infmiband格式的信号与私有协 议格式的信号之间的转接。  The router 62 further includes a cascading card 623 connected between the Infmiband switch 61 and the switch chip 621 for performing switching between signals in the Infmiband format and signals in the private protocol format.
上述集群路由器，利用标准的 Infmiband技术来取代现有的基于私有协议 的交换技术， 将各个基于私有协议的路由器要向外发送的信号转换成统一的 Infmiband格式的信号， 然后进行各个路由器与交换机之间的通信， 从而避免 由于私有协议而导致的集群路由器内部设备间无法兼容的问题， 同时也就解 决了网络节点转发性能瓶颈的问题。  The above-mentioned cluster routers use the standard Infmiband technology to replace the existing private protocol-based switching technology, and convert the signals sent by the routers based on the private protocol into a unified Infmiband format signal, and then perform various routers and switches. Inter-communication, so as to avoid the incompatibility between the internal devices of the cluster router caused by the proprietary protocol, and also solve the problem of network node forwarding performance bottleneck.
进一步地， 如图 7所示， 上述路由器 62包括： 交换芯片 621及与其相连 的至少两个线卡 622，还有 Infmiband成帧器 6231及于该 Infmiband成帧器 6231 相连的至少一个 Infmiband光模块 6232, 这里的 Infmiband成帧器 6231 和 Infmiband光模块 6232的组合即为上述级联卡 623; 其中，  Further, as shown in FIG. 7, the router 62 includes: a switch chip 621 and at least two line cards 622 connected thereto, and an Infmiband framer 6231 and at least one Infmiband optical module connected to the Infmiband framer 6231. 6232, where the combination of the Infmiband framer 6231 and the Infmiband optical module 6232 is the above-mentioned cascade card 623;
所述 Infmiband成帧器 6231 , 用于将接收到的 Infmiband格式的信号转换 成当前路由器所支持私有协议格式的信号， 或者将接收到的私有协议格式的 信号转换成 Infmiband格式的信号；  The Infmiband framer 6231 is configured to convert the received signal in the Infmiband format into a signal of a proprietary protocol format supported by the current router, or convert the received signal in the proprietary protocol format into a signal in the Infmiband format;
所述 Infmiband光模块 6232, 连接在所述 Infmiband交换机 61和所述 Infmiband成帧器 6231之间， 用于完成光电转换。  The Infmiband optical module 6232 is connected between the Infmiband switch 61 and the Infmiband framer 6231 for performing photoelectric conversion.
此外， 所述 Infmiband光模块 6232通过光纤与所述 Infmiband交换机 61 进行连接。  In addition, the Infmiband optical module 6232 is connected to the Infmiband switch 61 through an optical fiber.
在本实施例提供的集群路由器内部， 如果用户需要通过一台路由器中的 一个线卡接口将某些信息传送给另一台路由器中某一线卡的接口处， 则其信 号传递过程大致如下： 与 Infmiband交换机 61相连的其中一台路由器 62通过其包含的其中一个 线卡 622接收到一基于私有协议 1的信号， 所述私有协议 1是当前路由器所 支持的私有通信标准； 所述线卡 622将所述信号通过交换芯片 621 传送给 Infiniband成帧器 6231，由 Infiniband成帧器 6231将该基于私有协议 1的信号 转换成 Infmiband格式的电信号， 然后 Infmiband光模块 6232将该 Infiniband 格式的电信号转换成 Infiniband格式的光信号， 并通过光纤传递到交换机 61 处； In the cluster router provided in this embodiment, if a user needs to transmit certain information to an interface of a line card of another router through one line card interface of one router, the signal transmission process is as follows: One of the routers 62 connected to the Infmiband switch 61 receives a signal based on the private protocol 1 through one of the line cards 622 it contains, the private protocol 1 being a private communication standard supported by the current router; the line card 622 The signal is transmitted to the Infiniband framer 6231 via the switch chip 621, and the signal based on the proprietary protocol 1 is converted into an electrical signal of the Infmiband format by the Infiniband framer 6231, and then the Infmiband optical module 6232 outputs the electrical signal in the Infiniband format. Converting the optical signal into the Infiniband format and transmitting it to the switch 61 through the optical fiber;
交换机根据所述 Infmiband格式的光信号中携带的目的地址信息，将所述 Infmiband格式的光信号转发给另一路由器 62; 所述另一路由器 62 中的 Infiniband光模块 6232将接收到的 Infiniband格式的光信号转换成 Infiniband 格式的电信号，并通过 Infmiband成帧器 6231将该 Infmiband格式的电信号转 换成基于私有协议 2的信号， 所述私有协议 2为所述另一路由器 62所支持的 私有通信标准； 之后， 将所述基于私有协议 2的信号通过交换芯片 621传递 到与其连接的一个线卡 622上。  The switch forwards the optical signal of the Infmiband format to another router 62 according to the destination address information carried in the optical signal of the Infmiband format; the Infiniband optical module 6232 of the other router 62 receives the Infiniband format. The optical signal is converted into an electrical signal in the Infiniband format, and the electrical signal of the Infmiband format is converted into a signal based on the proprietary protocol 2 by the Infmiband framer 6231, which is the private communication supported by the other router 62. Standard; The signal based on the proprietary protocol 2 is then passed through the switch chip 621 to a line card 622 connected thereto.
本发明实施例提供的集群路由器，利用标准的 Infmiband技术来取代现有 的基于私有协议的交换技术， 将各个基于私有协议的路由器要向外发送的信 号转换成统一的 Infiniband格式的信号， 然后进行各个路由器（即集群路由器 中的转发框） 与交换机（即集群路由器中的中心框）之间的通信， 从而避免 由于私有协议而导致的集群路由器内部设备间无法兼容的问题， 同时也就解 决了网络节点转发性能瓶颈的问题。  The cluster router provided by the embodiment of the present invention replaces the existing private protocol-based switching technology by using the standard Infmiband technology, and converts the signals sent by the routers based on the private protocol into the unified Infiniband format signal, and then performs the signal. Communication between each router (that is, the forwarding frame in the cluster router) and the switch (that is, the central box in the cluster router), thereby avoiding the problem of incompatibility between the internal devices of the cluster router due to the proprietary protocol, and also solving the problem. Network nodes forward performance bottlenecks.
此外， 一般集群路由器的成本 4艮大部分用于互联的物理层光模块上。 传 统的集群路由器由于采用的都是基于私有协议的光模块， 因此无法统一地大 规模生产， 也就无法形成产业链， 致使路由器中光模块部分的价格昂贵； 而 本发明使用 Infmiband网络里的标准光模块进行互联，在成本上远低于传统的 集群路由器。 而且， 由于本实施例中提供的集群路由器不再受到私有协议的 限制， 因此其可实现的规模也远大于传统集群路由器。 通过以上实施方式的描述， 本领域的技术人员可以清楚地了解到本发明 可借助软件加必需的硬件平台的方式来实现， 当然也可以全部通过硬件来实 施。 基于这样的理解， 本发明的技术方案对背景技术做出贡献的部分可以以 软件产品的形式体现出来， 该计算机软件产品可以存储在存储介质中， 如In addition, the cost of a typical cluster router is mostly used for interconnected physical layer optical modules. Since traditional cluster routers use optical modules based on proprietary protocols, they cannot be mass-produced uniformly, and the industrial chain cannot be formed, resulting in an expensive optical module portion of the router. The present invention uses the standard in the Infmiband network. Optical modules are interconnected and are much lower in cost than traditional cluster routers. Moreover, since the cluster router provided in this embodiment is no longer limited by the proprietary protocol, the scale that can be implemented is much larger than that of the traditional cluster router. Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary hardware platform, and of course, all can be implemented by hardware. Based on such understanding, the portion of the technical solution of the present invention that contributes to the background art may be embodied in the form of a software product, which may be stored in a storage medium, such as
ROM/RAM,磁碟、 光盘等， 包括若干指令用以使得一台计算机设备（可以是 个人计算机， 服务器， 或者网络设备等）执行本发明各个实施例或者实施例 的某些部分所述的方法。 ROM/RAM, diskette, optical disk, etc., comprising instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods described in various embodiments or portions of the embodiments of the present invention. .
以上所述， 仅为本发明的具体实施方式， 但本发明的保护范围并不局限 于此， 任何熟悉本技术领域的技术人员在本发明揭露的技术范围内， 可轻易 想到的变化或替换， 都应涵盖在本发明的保护范围之内。 因此， 本发明的保 护范围应以权利要求的保护范围为准。  The above is only the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any change or replacement that can be easily conceived by those skilled in the art within the technical scope of the present invention is All should be covered by the scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

权利要求 书 Claim
1、 一种路由器， 包括一个交换芯片以及与其相连的至少两个线卡， 其特征 在于， 所述至少两个线卡中包括支持 Infmiband技术的 Infiniband线卡。  A router comprising a switch chip and at least two line cards connected thereto, wherein the at least two line cards comprise an Infiniband line card supporting Infmiband technology.
2、 根据权利要求 1所述的路由器， 其特征在于， 所述 Infmiband线卡包括： Infiniband光模块， 与外部 Infiniband网络连接， 用于对接收到的码流进行 光电转换；  The router according to claim 1, wherein the Infmiband line card comprises: an Infiniband optical module, and is connected to an external Infiniband network, for performing photoelectric conversion on the received code stream;
Infiniband成帧模块， 用于将接收到的 Infmiband码流解析成报文， 或者将 接收到的报文封装成 Infmiband码流；  An Infiniband framing module, configured to parse the received Infmiband code stream into a message, or encapsulate the received message into an Infmiband code stream;
网络处理器模块， 用于根据所述报文的内容获取到报文的目的地址； 交换接入芯片， 与所述交换芯片连接， 用于对接收到的报文进行重组、 存 储和排序。  The network processor module is configured to obtain a destination address of the packet according to the content of the packet, and exchange the access chip, and connect with the switch chip, to reassemble, store, and sort the received packet.
3、 一种集群路由器， 包括一个交换机和至少两个路由器， 其特征在于， 所述交换机为支持 Infiniband技术的 Infmiband交换机；  A cluster router, comprising a switch and at least two routers, wherein the switch is an Infmiband switch supporting Infiniband technology;
所述路由器包括一个交换芯片、 与该交换芯片相连的至少两个线卡， 所述 交换芯片为支持私有协议的交换芯片；  The router includes a switch chip, at least two line cards connected to the switch chip, and the switch chip is a switch chip supporting a private protocol;
该路由器还包括级联卡 ,该级联卡连接在所述 Infmiband交换机和所述交换 芯片之间， 用于完成 Infiniband格式的信号与私有协议格式的信号之间的转接。  The router further includes a cascading card connected between the Infmiband switch and the switch chip for performing switching between signals in the Infiniband format and signals in the proprietary protocol format.
4、 根据权利要求 3所述的集群路由器， 其特征在于， 所述级联卡包括一个 Infiniband成帧器以及与该 Infmiband成帧器相连的至少一个 Infiniband光模块； 其中，  The cluster router according to claim 3, wherein the cascading card comprises an Infiniband framer and at least one Infiniband optical module connected to the Infmiband framer;
所述 Infmiband成帧器用于将接收到的 Infmiband格式的信号转换成当前路 由器所支持私有协议格式的信号， 或者将接收到的私有协议格式的信号转换成 Infiniband格式的信号 ^  The Infmiband framer is configured to convert the received signal in the Infmiband format into a signal of a proprietary protocol format supported by the current router, or convert the received signal in the proprietary protocol format into a signal in the Infiniband format.
所述 Infmiband光模块， 连接在所述 Infmiband交换机和所述 Infmiband成 帧器之间， 用于完成光电转换。  The Infmiband optical module is connected between the Infmiband switch and the Infmiband framer for performing photoelectric conversion.
5、 根据权利要求 4所述的集群路由器， 其特征在于， 所述 Infmiband光模 块通过光纤连接到所述 Infiniband交换机。 The cluster router according to claim 4, wherein the Infmiband optical mode The block is connected to the Infiniband switch via fiber optics.
PCT/CN2009/074937 2009-11-13 2009-11-13 Router and cluster router WO2011057447A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2009/074937 WO2011057447A1 (en) 2009-11-13 2009-11-13 Router and cluster router
CN2009801476192A CN102150400A (en) 2009-11-13 2009-11-13 Router and cluster router

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/074937 WO2011057447A1 (en) 2009-11-13 2009-11-13 Router and cluster router

Publications (1)

Publication Number Publication Date
WO2011057447A1 true WO2011057447A1 (en) 2011-05-19

Family

ID=43991169

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/074937 WO2011057447A1 (en) 2009-11-13 2009-11-13 Router and cluster router

Country Status (2)

Country Link
CN (1) CN102150400A (en)
WO (1) WO2011057447A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102542524A (en) * 2011-12-31 2012-07-04 曙光信息产业股份有限公司 Cluster workstation and method for realizing same
CN106878180A (en) * 2016-12-23 2017-06-20 新华三技术有限公司 Cluster routers route control method, device and cluster routers

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103684855A (en) * 2013-11-29 2014-03-26 中国电子科技集团公司第三十研究所 Intelligent control method for cluster routers

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1492643A (en) * 2002-10-21 2004-04-28 华为技术有限公司 Device and method for realizing Infini Band data transmission in optical network
CN1494003A (en) * 2002-10-30 2004-05-05 华为技术有限公司 Device and method for realizing interface conversion
CN1864392A (en) * 2003-10-14 2006-11-15 国际商业机器公司 Method and apparatus for translating data packets from one network protocol to another
CN101355430A (en) * 2007-07-27 2009-01-28 华为技术有限公司 Exchange frame, cluster router
US20090141734A1 (en) * 2007-12-04 2009-06-04 Brown Aaron C Method and system for a converged infiniband over ethernet network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1492643A (en) * 2002-10-21 2004-04-28 华为技术有限公司 Device and method for realizing Infini Band data transmission in optical network
CN1494003A (en) * 2002-10-30 2004-05-05 华为技术有限公司 Device and method for realizing interface conversion
CN1864392A (en) * 2003-10-14 2006-11-15 国际商业机器公司 Method and apparatus for translating data packets from one network protocol to another
CN101355430A (en) * 2007-07-27 2009-01-28 华为技术有限公司 Exchange frame, cluster router
US20090141734A1 (en) * 2007-12-04 2009-06-04 Brown Aaron C Method and system for a converged infiniband over ethernet network

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102542524A (en) * 2011-12-31 2012-07-04 曙光信息产业股份有限公司 Cluster workstation and method for realizing same
CN106878180A (en) * 2016-12-23 2017-06-20 新华三技术有限公司 Cluster routers route control method, device and cluster routers
CN106878180B (en) * 2016-12-23 2020-01-03 新华三技术有限公司 Cluster router route control method and device and cluster router

Also Published As

Publication number Publication date
CN102150400A (en) 2011-08-10

Similar Documents

Publication Publication Date Title
US20210051045A1 (en) Communication switching apparatus for switching data in multiple protocol data frame formats
US8369347B2 (en) Fiber channel over Ethernet and fiber channel switching based on Ethernet switch fabrics
US7830875B2 (en) Autonegotiation over an interface for which no autonegotiation standard exists
KR100750880B1 (en) Switching system and its mechanism which enables data-switching of variable-length packets and heterogeneous network packets
US20160337146A1 (en) Method of data delivery across a network fabric in a router or ethernet bridge
US20150334034A1 (en) Methods and systems for accessing remote digital data over a wide area network (wan)
WO2014161291A1 (en) Data scheduling and switching method, apparatus, and system
WO2014206103A1 (en) Data packet switching method and device, access switch and switching system
KR20040024969A (en) Method and apparatus for integration processing of different network protocols and multimedia traffics
CN100561990C (en) A kind of digital home gateway device and processing method thereof
JPH07321842A (en) Apparatus for interfacing packet switching network to plurality of data terminals,module for interfacing plurality of end points to system that switches frame relay packet,and method for interfacing terminal to system that switches data packet
WO2011120214A1 (en) Cluster router and cluster routing method
US11509579B2 (en) Gateway for remote direct memory access over converged ethernet
CN116760911A (en) Heterogeneous protocol conversion system and method
WO2011057447A1 (en) Router and cluster router
WO2013159501A1 (en) Data transmission system, data interface device and data transmission method used among multiple servers
WO2022147792A1 (en) Switching system, switching network and switching node
Ocheltree et al. A comparison of fibre channel and 802 MAC services
WO2015010514A1 (en) Signal conversion method, apparatus and system
CN109803322B (en) Data frame forwarding system and method
KR100441882B1 (en) Packet processor supporting CSIX-L1 interface
CN114066707A (en) General graphic processing system, computing device and distributed system
WO2009086778A1 (en) Rate adapting method and apparatus, exchange board and line card
CN118075364A (en) Method and device for realizing conversion bridge between FC switch and Ethernet switch
CN118075216A (en) Full-switching circuit and method

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200980147619.2

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09851202

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09851202

Country of ref document: EP

Kind code of ref document: A1