CN1409903A - Encapsulation protocol for linking storage area networks over packet-based network - Google Patents

Encapsulation protocol for linking storage area networks over packet-based network Download PDF

Info

Publication number
CN1409903A
CN1409903A CN 00817156 CN00817156A CN1409903A CN 1409903 A CN1409903 A CN 1409903A CN 00817156 CN00817156 CN 00817156 CN 00817156 A CN00817156 A CN 00817156A CN 1409903 A CN1409903 A CN 1409903A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
network
method
protocol
system
nodes
Prior art date
Application number
CN 00817156
Other languages
Chinese (zh)
Inventor
罗伯特·A·雷诺兹
约翰·B·希切顿
肯尼思·D·斯梅兹
Original Assignee
交叉路径系统公司
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

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. local area networks [LAN], wide area networks [WAN]
    • H04L12/46Interconnection of networks
    • H04L12/4633Interconnection of networks using encapsulation techniques, e.g. tunneling
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L29/00Arrangements, apparatus, circuits or systems, not covered by a single one of groups H04L1/00 - H04L27/00 contains provisionally no documents
    • H04L29/02Communication control; Communication processing contains provisionally no documents
    • H04L29/06Communication control; Communication processing contains provisionally no documents characterised by a protocol
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/10Network-specific arrangements or communication protocols supporting networked applications in which an application is distributed across nodes in the network
    • H04L67/1097Network-specific arrangements or communication protocols supporting networked applications in which an application is distributed across nodes in the network for distributed storage of data in a network, e.g. network file system [NFS], transport mechanisms for storage area networks [SAN] or network attached storage [NAS]
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/202Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Application independent communication protocol aspects or techniques in packet data networks
    • H04L69/30Definitions, standards or architectural aspects of layered protocol stacks
    • H04L69/32High level architectural aspects of 7-layer open systems interconnection [OSI] type protocol stacks
    • H04L69/322Aspects of intra-layer communication protocols among peer entities or protocol data unit [PDU] definitions
    • H04L69/329Aspects of intra-layer communication protocols among peer entities or protocol data unit [PDU] definitions in the application layer, i.e. layer seven

Abstract

本发明揭示了一种为在两个或更多个节点(110,112,115)之间通过一个基于分组的网络(140)进行数据传输而封装SCSI协议的方法和系统。 The present invention discloses a method and system for data transmission of packet-based network (140) via a between two or more nodes (110,112,115) encapsulated SCSI protocol. 本发明的方法包括在该网络的每个节点执行的下列步骤:标识在这个网络上的所有其他可用节点和连接到这些节点上的远地设备;表示一个或多个所连接的远地设备,使得它们对于本节点的本地主机成为可用的;封装在一个或多个本地主机与一个或多个远地设备之间的I/O阶段;以及为在一个或多个主机与一个或多个设备之间的后续I/O重复封装步骤。 The method of the present invention comprises the steps of each of the nodes of the network to perform: identifying all the other available nodes in this network and the nodes connected to the remote device; represents one or more remote devices connected, for a local host so that they become available to the local node; enclosed between one or more local host and a plurality of remote devices or I / O phase; and as one or more hosts with one or more devices between the subsequent I / O encapsulation step is repeated. 封装在一个本地主机与一个远地设备之间的I/O阶段的步骤还可以包括封装任务管理功能、错误恢复功能和普通的I/O处理功能。 Encapsulating step the I / O phase between a local host and a remote device may further comprise packaging task management, error recovery and normal I / O processing. 每个节点(130,120,150)可以是一个光纤通道-SCSI路由器。 Each node (130,120,150) may be a Fiber Channel -SCSI router.

Description

通过一个基于分组的网络链接存储区域网络的封装协议 By encapsulating a packet-based network link protocol storage area network

技术领域 FIELD

本发明一般与数据和信息通信系统以及它们的运行有关,具体地说,与存储区域联网技术有关。 The present invention relates generally to data communications and information systems and their related operation, in particular, with a storage area networking technologies. 特别是,本发明涉及光纤通道存储区域网(SAN)和通过一个基于分组的网络链接存储区域网络的封装协议(encapsulation protocol)。 In particular, the present invention relates to Fiber Channel storage area network (SAN) and via a packet-based network protocol encapsulated link storage area network (encapsulation protocol).

存储区域网SAN是供数据存储专用的计算机网,可以帮助解决其中的一些问题。 Storage area network SAN is a dedicated computer for data storage networks, can help solve some of these problems. 存储区域网采用一种不同的称为光纤通道(FC)的更高性能计算机协议来传输数据。 A storage area network using a different protocol of higher performance computer called Fiber Channel (FC) to transmit data. 存储区域网还消除了服务器与存储设备之间一对一连接的限制,允许多个服务器与多个存储设备连接,共享对多个存储设备的访问。 A storage area network also removes the restriction between servers and storage devices connected to one, and allows multiple servers connected to a plurality of storage devices, shared access to multiple storage devices. 这种通过存储区域网得以实现的多对多连接,再加上光纤通道协议,使数据传输过程可以更快速、更高效和更可靠,也更容易管理。 This is achieved by a storage area network connecting many to many, plus the fiber channel protocol, data transmission process may be faster, more efficient and more reliable, easier to manage. 此外,存储区域网可以使数据备份通过它而不是通过主计算机网来实现,因此大大地减少了主计算机网上的拥塞,从而使主计算机网可以有效得多地进行日常运作。 In addition, the storage area network data backup can be achieved through it instead of the host computer through the network, thus greatly reducing the host computer Internet congestion, so that the main computer network can be much more effective in daily operations.

然而,市面上所售大多数存储设备仍配置小型计算机系统接口。 However, the market for sale Most storage device is still configure small computer system interface. 此外,大多数机构已经在采用小型计算机系统接口的存储设备和服务器上作了相当大的投资。 In addition, most agencies have been employing considerable investment on storage devices and servers Small Computer System Interface. 因此,为了使一个光纤通道存储区域网的设备能与采用SCSI的存储设备共用,在这些设备之间必须安装一些存储路由器(storage router)。 Accordingly, in order to make a fiber channel storage area network devices to SCSI storage devices using the shared between these devices must be installed in some of the storage router (storage router). 具体地说,存储路由器对于将数据备份过程从一个主计算机网移到存储区域网上来说是必不可少的,因为大多数数据备份存储设备采用的是SCSI接口,只能通过存储路由器连接到存储区域网上。 Specifically, the storage router for backing up data from a host computer the process moves to the storage area network line is essential, since most data backup storage device uses a SCSI interface, can only be connected to the storage router via a storage regional online. 随着新的计算机协议的引入,存储路由器日益成为采用不同协议的服务器、存储设备和存储区域网设备之间进行快速、无缝通信所必需的。 With the introduction of new computer protocol, storage router becoming performed between servers, storage devices and storage area network devices using different protocols fast, seamless necessary for communication.

然而,典型的SAN是为一个特定的机构或特定的场所服务的本地光纤通道网。 However, typical SAN local Fiber Channel network to a specific mechanism or a specific location and services. 这些SAN可以相当大,但是不能跨越很长的距离,因为这些SAN都有着由支持光纤通道所需的基础设施给定的距离限制。 The SAN can be quite large, but can not span long distances, because they have the required by the SAN Fiber Channel infrastructure to support a given distance limitations. 例如,光纤通道标准规定了进行跨距最大能达到10km而在有些情况下能达到30km的通信的方法。 For example, Fiber Channel standard specifies the maximum span for 10km reach in some cases to achieve a method of communication of 30km. 然而,为了能这样,实现光纤通道网的机构通常必须拥有光纤或者从其他方租用专用光纤,这可能是非常昂贵的,在大多数情况下会受到成本限制。 However, in order to so, the Fiber Channel network to achieve agency usually must have a dedicated fiber optical fiber or leased from other parties, which can be very expensive, and in most cases it would be cost-prohibitive.

这是因为用来承载光纤通道业务的光纤只能承载光纤通道协议的业务。 This is because the fiber channel used to carry optical traffic bearer services only the Fiber Channel protocol. 它们不能与其他协议共享。 They can not be shared with other protocols. 因此,更为经济的是采用一种能被诸如由电话公司拥有的那些可以承载ATM业务、SONET业务和IP业务的现有网络支持的协议进行远距离传输数据。 Thus, a more economical use a can, such as those owned by the telephone company can carry ATM traffic, support of the existing network services and IP services SONET protocol for remote data transmission. 所以,SAN通常局限于它们可以服务的地理区域(即,它们局限于本地运作)。 So, SAN are usually limited geographic area they can service (ie, they are limited to local operations). 此外,地域不同的两个或更多个SAN不能以它们似乎相互都是本地SAN那样无缝地相互连接,因为连接它们的基础设施不存在或者费用过高。 In addition, different regions of two or more SAN can not they seem to like each other are local SAN seamlessly connected to each other, because they are connected to the infrastructure does not exist or too costly.

发明概述因此,需要有一种连接距离大于光纤通道协议当前可支持距离的多个本地SAN的方法和系统,为实现联合或全局存储区域网提供支持。 Summary of the Invention Accordingly, there is need for a connector a distance greater than the Fiber Channel protocol currently supports a plurality of local SAN distance method and system, support for joint or global storage area network.

更进一步地说,需要有一个封装协议,能够利用现有的通信网,通过一个诸如IP、ATM、SONET之类的基于分组的网络协议或其他这样的现有通信协议来连接多个存储区域网。 More particularly, there is a need for an encapsulation protocol, the existing communication network can be utilized, such as by an IP, ATM, SONET and the like of a packet-based network protocol, or other such conventional connecting a plurality of communication protocols to the storage area network .

再进一步说,需要有一个封装协议,可以通过一个基于分组的网络以无缝方式链接多个存储区域网,使得这些SAN工作起来就好象它们都是本地的SAN。 Still further, there is a need for the encapsulation protocol, can seamlessly link the plurality of storage area network via a packet-based network, such that they are working together as if these local SAN SAN.

本发明提供了一种通过一个基于分组的网络链接多个SAN的封装协议的方法和系统,这种方法和系统基本上消除或者减少了与大跨距情况下采用光纤通道协议相关的缺陷和问题。 The present invention provides a packet-based network via a plurality of links SAN protocol encapsulation method and system, a method and system which substantially eliminates or reduces the Fiber Channel protocol and the span is large and where the drawbacks associated problems . 具体地说,本发明提供了一种无缝互联一些地域不同的SAN的方法,使得它们工作起来就好象它们都是本地的SAN。 More specifically, the present invention provides a seamless interconnection of different local SAN Some methods, such that together they work as if they are local SAN.

具体地说,本发明提供了一种为在两个或更多个节点之间通过一个基于分组的网络进行数据传输而封装SCSI协议的方法和系统。 More specifically, the present invention provides a method and system for data transmission via a packet-based network between two or more nodes encapsulated SCSI protocol. 本发明的方法包括在该网络内的每个节点执行的下列步骤:标识在这个网络上所有其他可用节点和连接到这些节点上的远地设备;表示一个或多个所连接的远地设备,使得它们对于本节点的本地主机成为可用的;封装在一个或多个本地主机与一个或多个远地设备之间的I/O阶段(I/O phase);以及对于在一个或多个主机与一个或多个远地设备之间的后续I/O重复封装步骤。 The method of the present invention comprises the steps within the network each node is performed: this identifies all network nodes connected to other available devices on remote nodes; represents one or more remote devices connected, for a local host so that they become available to the local node; enclosed between one or more local host and a plurality of remote devices or I / O phase (I / O phase); and for one or more host subsequent to one or more remote devices among I / O encapsulation step is repeated. 封装在一个本地主机与一个远地设备之间的I/O阶段还可以包括封装任务管理功能、错误恢复功能和普通的I/O处理功能。 Package I / O phase between a local host and a remote device may further comprise packaging task management, error recovery and normal I / O processing. 每个节点可以是一个光纤通道-SCSI路由器。 Each node may be a Fiber Channel -SCSI router.

本发明有一个重要的技术优点,所提供的方法和系统可以连接距离大于当前在光纤通道协议下所允许距离的多个本地SAN,这种方法和系统将为实现联合或全局存储区域网提供所需的支持。 There is an important technical advantage of the present invention, methods and systems provide a connection distance may be greater than the current in the Fiber Channel protocol allows a plurality of local distances a SAN, such methods and systems would be combined to achieve a storage area network or global provide need of support.

更进一步地说,本发明有一个重要的技术优点,所提供的封装协议能够利用现有的通信网通过一个诸如IP、ATM、SONET之类的基于分组的网络协议或其他这样的现有通信协议连接多个存储区域网。 More particularly, there is an important technical advantage of the present invention, provided encapsulation protocol can utilize existing communication network through a packet-based network protocols such as IP, ATM, SONET and the like, or other such existing communication protocols connecting the plurality of storage area networks.

本发明还有一个重要的技术优点,所提供的封装协议可以通过一个基于分组的网络以无缝方式链接多个存储区域网,使得这些SAN工作起来就象是一个整体SAN。 Another important technical advantage of the present invention, the encapsulation protocol provides a seamless manner can link multiple storage area network via a packet-based network, such that the SAN work together as a whole is SAN.

发明详细说明附图例示了本发明的优选实施例,图中同样的数字用来标注同样和相应的部分。 DETAILED DESCRIPTION accompanying drawings illustrate preferred embodiments of the present invention embodiment, the same numerals are used in FIG same and corresponding parts are denoted.

本发明提供了一种为在一个或多个节点之间通过一个基于分组的网络的传输进行封装SCSI协议的方法和系统,这种方法和系统可以利用现有的通信网高效、经济地连接多个或许地域不同的SAN,使得它们工作起来就象是一个存储区域网。 The present invention provides a method to encapsulate SCSI protocol transport packets through a network based on one or more nodes among the methods and systems, such methods and systems may utilize the existing communication network efficiently and economically connect multiple perhaps a different geographical SAN, is that they work together as a storage area network. 因此,本发明的方法和系统能有效地克服现有的光纤通道网的距离限制,使SAN模式可以扩展到横跨许多英里的多个SAN。 Thus, the method and system of the present invention can effectively overcome distance limitations of existing Fiber Channel network, so that patterns can be extended to a plurality of SAN SAN across many miles. 例如,本发明可以将一个在洛杉矶的公司SAN与另一个在纽约或东京的公司SAN链接起来。 For example, the invention may be a SAN company in Los Angeles and another in New York or Tokyo-based company SAN linked. 在存储恢复的情况下,本发明允许一个备份库驻留在远地,从而在本地由于象火灾或水灾那样的灾祸而受破坏的情况下也能保证数据的完整性。 In the case of storing the recovery, the present invention allows a remote backup repository resides in, since the lower case so as local fire disaster such as a flood or disrupted and also ensures data integrity. 实现本发明的SAN没有必要局限于只是本地使用。 SAN implementation of the invention is not necessarily limited to only local use.

为了连接跨距比在光纤通道协议下所允许跨距大的一些本地SAN,本发明定义了一个封装协议(EP),可以对光纤通道协议(FCP)进行操作,使它可以通过诸如异步传输模式(ATM)或以太网之类的任何基于分组的传输设施进行传播。 In order to connect more than span the Fiber Channel protocol allows a large span of some of the local SAN, the present invention defines a protocol encapsulation (EP), can operate on Fiber Channel Protocol (FCP), it may be an asynchronous transfer mode, such as by (ATM) or Ethernet or the like propagate any packet-based transmission facilities. 图1为例示本发明的方法和系统的在典型的SAN环境内的一个实施例的简化框图。 A simplified block diagram illustrating the method and system of the present invention in FIG. 1 in a typical embodiment of the SAN environment. 图1的网络100包括远地主机110,远地主机110可以是本地光纤通道SAN,可以访问另一个用于例如磁带备份和磁盘镜象的本地光纤通道SAN 115。 Network 100 of FIG. 1 includes a remote host 110, remote host 110 may be a SAN native Fiber Channel, for example, you can access another local Fiber Channel SAN 115 in the mirror disk and tape backup. 光纤通道主机110和光纤通道主机115分别连接到节点120、130和150上,这些节点可以是光纤通道-SCSI路由器,诸如由Crossroads Systems公司制造和销售的光纤通道-SCSI路由器。 Fiber Channel host 110 and the Fiber Channel host 115 are connected to the node 120, 130 and 150, these nodes can be Fiber Channel -SCSI routers, such as routers Fiber Channel -SCSI Crossroads Systems manufactured and sold by the company. 节点120、130和150可以是SAN110和115与网络100其余部分的接口。 Nodes 120, 130 and 150 may be SAN110 100 and 115 with the rest of the network interface.

包括节点120、130和150的光纤通道-SCSI路由器都能实现EP层,使得光纤通道协议可以在基于分组的WAN(广域网)140上无缝流通。 Includes nodes 120, 130 and 150 of the fiber channel router can achieve -SCSI EP layer, such that the Fiber Channel protocol packet-based WAN (Wide Area Network) 140 flow seamlessly. WAN 140表示物理上的基于分组的传输设施,诸如ATM或以太网之类。 WAN 140 represents a packet-based transmission facilities, such as an ATM or Ethernet or the like on a physical. WAN 140可以是一个专用链路或交换网络。 WAN 140 may be a switching network or a dedicated link. 主机110和115通过光纤通道链路190连接到它们各自的节点上。 Host 110 and 115 connected to their respective nodes 190 via a Fiber Channel link. 节点120、130和150各通过网络链路192连接到WAN 140上。 Nodes 120, 130 and 150 are each connected to the WAN 140 through a network link 192. 光纤通道链路190按照给定应用的需要可以是铜线或光纤链路。 Fiber Channel links 190 desired for a given application may be copper wire or fiber optic links. 网络链路192按需要同样可以是铜线或光纤链路。 Network link 192 may also be needed copper wire or fiber optic links.

远地主机110和本地主机115可以包括多个目标设备和多个始发设备。 Remote host 110 and the local host 115 may include a plurality of target devices and a plurality of originating devices. 例如,SAN 115包括光纤通道集线器(交换机)160、磁带库170和磁盘180。 For example, SAN 115 includes a fiber channel hub (switch) 160, a library 170 and disk 180. 虽然只示出了磁带库170和磁盘180,但可以有多个始发和目标设备连接到光纤通道集线器160上,再通过它连接到光纤通道-SCSI路由器150上。 Although only the disk library 170 and 180, but a plurality of the originating and target devices connected to a Fiber Channel hub 160, and then connected to a Fiber Channel -SCSI router 150 through it. 光纤通道SAN 115因此可以包括多个主机和多个始发设备。 Thus Fiber Channel SAN 115 may include a plurality of hosts and plurality of the originating device.

用一个扩展设施协议(extender protocol),例如本发明的封装协议,将两个或更多个SAN连接在一起需要将一个SAN上的每个SCSI设备的本地地址映射为一个中间地址以通过扩展设施,再将每个中间地址映射为在一个远地SAN上的一个远地地址。 Expanded facility with a protocol (extender protocol), e.g. encapsulation protocol of the present invention, two or more connected SAN local address needs to each SCSI device on a SAN is mapped to an intermediate address by extending the facility , then the address mapping for each of the intermediate in a remote SAN is a remote address. 这种映射使得在一个SAN上的始发设备可以访问在一个远地SAN上的SCSI设备,就好象它们都是在这个始发设备所连接的本地SAN上的SCSI设备。 This mapping allows the originating device on a SAN SAN can be accessed on a remote SCSI devices, if they were all on the local SAN originating device attached SCSI devices. 这样,在一个远地SAN上的设备可以表示为对在其他SAN上的始发设备可用的设备。 Thus, a device on the remote SAN may be represented as originating devices on other SAN devices available. 在------递交的相关专利申请No.------“------”中揭示了一种在两个或更多个由一个SAN扩展设施(例如本发明的封装协议)连接的SAN之间映射SCSI设备地址的方法和系统。 In the related patent applications filed ------ No .------ "------" discloses a two by one or more SAN expanded facility (e.g., according to the present invention method and system for mapping between the SCSI device address SAN encapsulation protocol) connection. 该申请在此列作参考全面引用,因为它揭示了一种可兼容的在两个SAN之间映射SCSI设备地址的方法,可以与本发明一起使用。 Which is incorporated herein by reference in full by reference columns, since it discloses a method for mapping SCSI device address between two compatible SAN one kind can be used with the present invention.

本发明的方法和系统提供了一种规定通过扩展协议进行通信的方法,也就是一种将光纤通道数据映射为一个扩展协议,使得它可以解码恢复到光纤通道协议的方法,以便与一个在远地SAN上的目标设备通信。 The method and system of the present invention provides a predetermined communication protocol by extending the method, i.e. a method of mapping data to a Fiber Channel protocol extension, the decoding method makes it possible to recover the Fiber Channel protocol, a far so as communication target device on the SAN. 因此,本发明规定了一种可以用来将光纤通道封装入一个基于分组的网络的协议。 Accordingly, the invention provides a method may be used to package into the Fiber Channel protocol is a packet-based network. 这是通过将需发送的数据从需扩展的协议(光纤通道协议)变换到扩展设施协议(对于具体应用的基于分组的协议),再在远地SAN处变换回原协议(光纤通道协议)来实现的。 This is transformed to the expanded facility protocol (Fiber Channel Protocol) data to be transmitted from the need to expand (for protocol packet based on the specific application), then in the remote SAN the transformation back to the original protocol (Fiber Channel Protocol) achieved. 本发明的封装协议可以配置成变换一个给定协议的特定命令,而且随着光纤通道协议扩展提供新的功能,该封装协议可以扩展到包括一些新的命令。 Encapsulation protocol of the present invention may be configured to a specific command given protocol conversion, and with the fiber channel protocol extensions provide new features, the encapsulation protocol can be extended to include some new commands. 在本发明中引用的任何光纤通道命令或消息都是例示性的而不是排它性的。 Any Fiber Channel command or message cited in this disclosure are illustrative rather than an exclusive.

本发明的方法和系统的EP层包括两个部分:Aspen节点管理(ANM)和光纤通道协议-封装协议(FCP-EP)。 EP layer method and system of the present invention comprises two parts: Aspen Node Manager (the ANM) and Fiber Channel Protocol - encapsulation protocol (FCP-EP). 这些名称只是任意选来进行说明的。 These names are just arbitrarily selected for illustration. ANM可以认为是控制机制。 ANM control mechanisms can be considered. 它定义了一个客户机服务器环境,使得具有多个始发设备和多个目标设备的多个SAN可以作为在例如图1的WAN 140上的各个节点同时受到管理。 It defines a client-server environment, such that a plurality of SAN having a plurality of originating devices and multiple target devices can be used as, for example, each node on the WAN 140 of FIG. 1 are managed simultaneously. 在多节点的环境内,有一个节点被指定为服务器节点,所有的节点(包括服务器节点)都是客户机。 In a multi-node environment, one node is designated as a server node, all nodes (including the server node) is the client. 以下消息用来定义和管理ANM控制机制:●客户机节点注册(Client Node Register):由一个客户机节点用来向服务器节点注册。 To define and manage the message ANM control mechanisms: ● a client node (Client Node Register): a node for registering a client to a server node. 这个消息注册了客户机的EP地址,使得服务器可以向这个客户机开放将来的连接(如果必要的话)。 The news registered client EP address so that the server can open future connections to this client (if necessary).

●客户机节点目标注册(Client Node Target Register):由一个客户机节点用来向服务器节点注册FC目标设备。 ● target client node registration (Client Node Target Register): consists of a client node to register FC target device to the server node. 这个消息注册了在本地FC网络上发现的所有FC目标设备。 The news all registered FC target devices found on the local FC network.

●目标表更新(Target Table Update);由服务器节点发送,用来更新一个客户机的全局目标设备表。 ● table update target (Target Table Update); sent by the server node, the target device to update the global table a client. 目标表更新含有对于所有注册的客户机的所有项目,但是不规定客户机所用的表的格式。 Updating the target table contains all items for all registered client, but the client does not specify the format used in the table. 每个客户机可以用一种使每个客户机节点对表的使用最佳化的格式来格式化目标表。 Each client can be in a client node so that each of the preferred format of the table to the target table format.

●目标地址冻结(Target Address Freeze):由服务器节点发送,用来通知一个客户机节点将要发送一个目标表更新消息,应该挂起所有FCP-EP出站消息,直到接收到该目标表更新消息。 ● freeze destination address (Target Address Freeze): sent by the server node to notify a client node to send a target table update message, it should suspend all FCP-EP outbound message until receiving the update message to the target table.

●节点确认(Node Acknowledgment):是一个对ANM消息的确认帧,用于流量控制。 ● acknowledgment node (Node Acknowledgment): is an acknowledgment frame for the ANM message for flow control.

FCP-EP可被看作一种数据机制。 FCP-EP data can be seen as a mechanism. 它规定了通过WAN 140传输的FCP帧的分拆和重组。 It specifies spin recombinant WAN 140 and transmitted via the FCP frame. FCP-EP层本身可以是一个基于分组的协议,具有与SCSI-FCP序列的净荷定义十分类似的净荷定义。 FCP-EP layer itself may be a packet-based protocol, the payload has defined SCSI-FCP sequence very similar to the definition of the payload. 在分拆时,可以将来自本地SAN的FCP序列分成一些较小的、与要采用的低层传输协议(例如,ATM或以太网)可兼容的分组。 When spin FCP sequence, from the local SAN may be divided into a number of smaller, the underlying transport protocols (e.g., ATM or Ethernet) packet compatible to be used. 消息标识符用来保留在分拆期间的序列信息。 Message identifier is used to retain the split information during the sequence. 然后,将经分拆的序列组合成一些帧,通过物理链路传输。 Then, the composition was split into a sequence of frames, by a physical link transmission. 一旦通过了链路,另一个节点接收到这些帧后,就利用消息标识符重组FCP序列。 Once received frame over the link to the other node, a message identifier on the use of recombinant FCP sequences. 这些分组于是就可以按与它们原来生成时相同的格式传输给本地SAN。 These packets may then be transmitted to the local SAN press when they generate the same original format.

以下消息用来定义和管理FCP-EP数据机制:●SCSI命令(SCSI Command):含有一个新的FCP命令的信息。 The following message is used to define and manage data FCP-EP mechanisms: ● SCSI command (SCSI Command): contains information about a new FCP command.

●SCSI命令加数据(SCSI Command Plus Data):含有一个新的FCP命令的信息和一个用于所规定的FCP命令的数据块。 ● SCSI command with data (SCSI Command Plus Data): FCP command contains a new block of information and data for FCP command specified.

●SCSI数据请求(SCSI Data Request):含有一个写请求的数据传输量有关的信息。 ● SCSI data request (SCSI Data Request): contains information about the amount of data transfer is a write request. 这个消息请求由一个数据量字段规定的一定量的数据,数据开始点为一个相对偏置量字段规定的相对偏置量。 The amount of the request message by a predetermined amount data field data, the start point is offset relative to a predetermined amount of relative offset field.

●SCSI数据(SCSI Data):含有一个读、写要用的数据块。 ● SCSI data (SCSI Data): comprises a reading and writing use data blocks.

●SCSI数据响应(SCSI Data Response):含有一个数据块和FCP响应信息。 ● SCSI response data (SCSI Data Response): a block containing response information and FCP.

●SCSI响应(SCSI Response):含有FCP响应信息。 ● SCSI response (SCSI Response): comprising FCP response information.

●SCSI异常中止请求(SCSI Abort Request):用来通知一个I/O的异常中止条件。 ● SCSI abort request (SCSI Abort Request): for notifying an I / O abort condition is.

●SCSI数据确认(SCSI Data Acknowledgement):是一个对FCP-EP消息的确认帧,用于流量控制。 ● SCSI data acknowledgment (SCSI Data Acknowledgement): is an acknowledgment frame for FCP-EP message for flow control.

本发明的方法和系统可以在一个诸如图1中的路由器120、130和150(节点120、130和150)之类的光纤通道-SCSI路由器内实现。 The method and system of the present invention may be implemented in a router, such as a Fiber Channel -SCSI in FIG. 1 routers 120, 130 and 150 (nodes 120, 130 and 150) or the like. 本发明可以实现为纯粹是一些存储在光纤通道-SCSI路由器中的存储器内的软件指令,从而可以很方便地升级为具有创建的新功能的新版本。 The present invention may be implemented purely in software instructions stored in the memory of some fiber channel -SCSI router, so that can be easily upgraded with new features of the new version created. 实现本发明并不要求在现有的光纤通道-SCSI路由器的硬件上有什么改变。 Implement the present invention does not require any change in existing hardware Fiber Channel -SCSI router. 存储本发明的软件指令的存储器可以是RAM(随机存取存储器)或ROM(只读存储器),或者其他存储装置。 The memory stores software instructions of the present invention may be a RAM (Random Access Memory) or a ROM (Read Only Memory) or other storage devices.

本发明的一种实施例可以包括一个动态发现机制,借助于这个机制,在一个多节点实施例中的每个节点(路由器)可以与所有的其他节点通信,以便例如最初发现在系统内存在的其他这样的节点。 One inventive embodiment can include a dynamic discovery mechanism, by means of this mechanism, each node (router) in the embodiment in more than one node can communicate with all other nodes, for example, originally discovered in the system memory other such nodes. 这个动态发现机制可用来使每个节点通过一个公共服务器与每个其他节点通信。 This dynamic discovery mechanism may be used to each node via a common server nodes communicate with each other. 网络内的每个节点因此可以发现存在的其他节点,从而可以与它可用的、接有目标设备的其他节点通信。 Each node within the network can be found in the presence of other nodes, so that it may be available, other nodes connected to the communication target device. 每个节点可以从每个其他节点接收该信息,从而可以向它们各自连接的主机或者配置每个路由器(节点)的技术人员指出在所有可用节点上可用的所有目标设备。 Each node can receive this information from every other node, so that the host can be connected to their respective configuration or each router (node) in the art indicates that all target devices available on all available nodes.

至少有一个路由器(节点)必须被指定为服务器,通过这个服务器可以实现上述发现功能。 At least one router (node) must be designated as the server, this can be achieved by the discovery server. 也可以将其他路由器例如在公共服务器出现故障和进行错误恢复的情况下指定为服务器,或者就指定为公共服务器的一个备份。 The case of other routers can also be a server failure, for example, in public and error recovery of the designation as a designated backup server is a public server, or. 服务器“模块”实际上不一定要驻留在SAN的一个光纤通道-SCSI路由器内。 Server "module" does not actually have to reside within a Fiber Channel SAN -SCSI router. 它也可以是一个仅提供服务器功能的单独设备。 It may also be a function of only providing server separate device. 然而,一个光纤通道-SCSI路由器可以含有和提供服务器功能。 However, a fiber channel may contain -SCSI router and providing server functionality. 如果服务器功能由一个单独的设备提供,这个单独的服务器可以集成入一个不在SAN内的标准网络设备,而SAN的路由器可以与这个服务器通信,接收所需的信息。 If the server functionality is provided by a separate device, the separate server may be integrated into a standard network device is not within the SAN, and the SAN router can communicate with the server to receive the information needed. 因此,这个服务器就是一个实在的“服务器”,而不是在一个光纤通道-SCSI路由器内的一个附加功能。 Therefore, this server is a real "servers" rather than an additional feature in a Fiber Channel -SCSI router.

可以将本发明的服务器功能与在IP网络内的DNS服务器作一下类比。 Server function of the present invention may be in the DNS server for the IP network at the analogy. DNS服务器存在于网络基础设施内,知道怎样与主服务器通信。 DNS server exists in the network infrastructure, we know how to communicate with the master server. 有一个规定的协议,一个IP网络内的主服务器可以用这个协议相互发现。 There is a predetermined protocol, a master server within an IP network using this protocol can be found in each. 本发明的动态发现机制实质上提供了相同的功能,因此可以集成入网络,而使光纤通道-SCSI路由器只是简单的客户机节点,它们可以相互通信和与服务器通信,获得有关在网络上的任何其他节点的信息。 Dynamic discovery mechanism of the present invention essentially provides the same functionality, can be integrated into the network, the routers simply -SCSI Fiber Channel client nodes, they can communicate with each other and communicate with the server, obtain any information about the network on information on other nodes.

本发明的封装协议的方法和系统可以通过现有的因特网基础设施和其他现有的网络协议来应用。 The method and system of the present invention is an encapsulation protocol can be applied by existing Internet infrastructure and other existing network protocols. 例如,扩展设施协议可以是一个典型的IP网络协议、ATM网或千兆比特以太网协议,或者任何允许数据分组在节点之间流通的协议。 For example, the expanded facility may be a typical protocol of the IP network protocol, ATM or Gigabit Ethernet network protocol, or any protocol allows packets of data flow between nodes. 本发明的方法和系统通过封装光纤通道SCSI,提供了一种使数据可以在一个扩展网络两端的任何SCSI协议SAN之间传输的方法。 The method and system of the present invention by encapsulating Fiber Channel SCSI, there is provided a so data can be transferred between any SCSI protocol across a network SAN extension method. 本发明的方法和系统规定了一种发现在一个网络设施内可用的所有节点的动态方式,使得在具有多个目标设备和始发设备的多个节点(路由器)之间可以建立1对n或者n对n的关系。 The method and system of the present invention provides a dynamic manner found available within a network all the nodes of the facility, so that one pair can be established between a plurality of n nodes (routers) having a plurality of target devices and the originating device or n-to-n relationship. 每个节点(router)都是一个对应的SAN的访问点。 Each node (Router) is an access point corresponding to the SAN.

通过封装光纤通道协议消息,本发明的方法和系统可通过中间的WAN 140传输标准化的消息,使它们在一个远地SAN内可以得到解码和起作用而不会丢失或受到破坏。 By encapsulating the Fiber Channel protocol messages, the method and system of the present invention may be transmitted through the intermediary of a standardized message WAN 140, such that they can be decoded at a remote place and function within the SAN without loss or damage. 因此,标准化的消息可以通过WAN140从一个本地SAN到达一个远地SAN而不需要专用协议格式。 Thus, standardized message can be reached by a local WAN140 from a remote SAN SAN without the need for a dedicated protocol format.

在操作时,一个实现本发明的光纤通道-SCSI路由器最初在一个存储区域网110或115内进行联机时,它向指定的服务器注册,向网络标识本身,并通过交换信息接收有关这个网络上存在的每个其他节点以及每个节点可通达的SCSI目标设备的信息。 In operation, an implementation of the present invention, Fiber Channel -SCSI router is initially performed on-line in a storage area network 110 or 115, it registers to the specific server, identifies itself to the network, and receive information about the existence on the network by exchanging information information on each of the other nodes and each node SCSI target device can be accessible. 同样,所指定的服务器可以检测出一个光纤通道-SCSI路由器由于维护或发生故障而从网络撤出的事件。 Similarly, the designated server can detect a Fiber Channel -SCSI router failure occurs due to maintenance or withdrawal from the network event. 可以通过光纤通道SCSI协议内的脉搏(heartbeat)消息来提供此功能。 This functionality may be provided by a pulse (Heartbeat) message within a Fiber Channel SCSI protocol. 对于一个实现本发明的存储区域网可以含有的节点的数量,在理论上目前并没有什么限制。 For a number of nodes of the storage area network according to the present invention may be implemented to contain, in theory no limit to the current. 不过,节点的数量受硬件的限制,特别是受一个光纤通道-SCSI路由器内的存储器的容量的限制。 However, the number of nodes is limited hardware, particularly memory capacity is limited within a Fiber Channel -SCSI router.

本发明并不局限于用于每个存储区域网都采用相同的光纤通道协议的情况。 The present invention is not limited to the case of each of the storage area network use the same fiber channel protocol. 例如,图1中的主机115和主机110可以采用不同的协议,采用了本发明的封装协议它们仍能正确地工作。 For example, in FIG. 1 and the host 115 the host 110 may use different protocols, using the protocol of the present invention which the package can still work correctly. 通过将光纤通道SCSI封装成中间形态,本发明为在例如扩展设施一侧的光纤通道网与扩展设施另一侧的并行SCSI网络之间出现的转换提供了足够的信息。 By Fiber Channel SCSI encapsulated into an intermediate form, the present invention is, for example, parallel SCSI conversion occurs between the network side of the other side extension facility Fiber Channel network and expanded facility provides sufficient information. 利用在封装的光纤通道协议的消息内的足够信息,本发明可以执行在任何两种类型的SCSI网络之间的转换。 With sufficient information within the message encapsulated fiber channel protocol, and the present invention can perform the conversion between any two types of SCSI network.

概括地说,本发明的方法的一个实施例包括下列步骤:将一个光纤通道-SCSI路由器挂到网络上联机,这个路由器与一个指定的服务器通信,向网络标识本身,而且接收网络上其他可用节点的信息。 Broadly speaking, one embodiment of the method of the present invention comprises the steps of: a Fiber Channel -SCSI router linked to the network connection, the router server in communication with a specified identifier to the network itself, but other available nodes on the network receives Information. 这个光纤通道-SCSI路由器还将接收有关连接到各个其他网络节点上的可用设备的信息。 The Fiber Channel -SCSI also receive information about the router is connected to each of the other network nodes available devices. 这个过程发生在每个节点挂到存储区域网络上进行联机时。 This process occurs in each node to be linked on-line storage area network. 这样,本发明的方法和系统在每个节点确定哪些其他节点可用和哪些设备连接到这些节点上。 Thus, the method and system of the invention can be used to determine which other nodes and which devices are connected to these nodes in each node.

本发明的方法将在每个节点上的每个可用设备表示为一个具有一个或几个LUN的单独目标设备或者表示为一组具有一个或几个LUN的目标设备,并使得所表示的设备成为对本地存储区域网上的所有其他主机可用。 The method of the present invention will be used for each device on each node is represented as a device having a single or several target LUN, or as a set of one or more target device having the LUN, and so the device becomes expressed All other hosts on the local storage area available online. 这意味着每个联机的节点表示所登记的其他节点上的设备,使得这些设备成为对于它自己的主机是可用的。 This means that each node represents the device online on other nodes registered, making these devices to become its own host is available.

来自连接到一个联机的光纤通道-SCSI路由器上的一些主机的通信于是可以被封装,通过可用设备经由WAN 140扩展设施发送出去,类似地,经封装的消息可以在本地节点接收后,传输给它的本地主机。 After some of the host connected to the communication from the router on a Fiber Channel -SCSI online then be packaged, transmitted via the WAN 140 through the expanded facility available equipment, likewise, can be received in the local node the encapsulated message, transmitted to it local host. 因此,本发明可以封装在一个本地主机与对于这个主机成为可通达的一些远地设备之间的I/O阶段。 Accordingly, the present invention may be encapsulated in a local host and I / O stage may be reached between the number of remote devices to become the master. 在一个本地主机与一个远地设备之间封装的这些I/O阶段可以包括例如命令阶段、数据阶段和响应阶段。 Between a local host and a remote device packages, the I / O stage may comprise, for example, command phase, a data phase and a response phase.

因此,一个主机可以将一个命令由与它配合的光纤通道-SCSI路由器(节点)封装后,通过扩展设施(WAN 140)发送给一个与这个命令所要发向的预定设备配合的远地节点。 Therefore, the host can be a package of a command with its Fiber Channel -SCSI router (node), and transmitted to the predetermined device a command to be sent to the remote node with the expanded facility (WAN 140). 这个远地节点将对该命令解封装后转送给这个预定设备。 This remote node for the command transferred to the predetermined apparatus decapsulated. 这个设备然后可以回送数据或响应,或者数据和响应,经该远地节点封装后发送给与发送命令的始发设备(主机)配合的节点。 This device may then send back a response or data, or data, and in response, the originating transmitting device (host) for sending a command given by the node with the node is remote package. 与该发起主机配合的节点接收到响应或数据或者这两者,就对其进行解封装后转送给这个主机。 After the originating node receives a response of the host or data fitting, or both, would be forwarded to the host decapsulates. 命令、数据和/或响应的次序可以包含一个需封装的I/O阶段。 Order commands, data, and / or responses may comprise a package for an I / O stage. 本发明提供了封装协议但是并没有以任何其他方式改变主机与设备之间发送的数据。 The present invention provides a packaging protocol, but does not change the data transmission between the host and the device in any other way.

根据后续I/O的需要重复封装在一个本地主机与一个远地设备之间的I/O阶段的步骤。 Repeat steps I / O phase between a local host and a remote device package needed subsequent I / O's. 可以在连接到多个与多个设备通信的节点的多个主机之间执行封装,所有的这些主机和节点分别属于多个本地或远地SAN。 Encapsulation may be performed between a plurality of hosts connected to a plurality of nodes in communication with a plurality of devices, all of which belong to a plurality of nodes and hosts a local or remote SAN. 这样,本发明的方法和系统建立了一种初始网络配置,在一个本地SAN的多个本地路由器与在一个或多个远地SAN的多个远地路由器之间或者在同一个SAN内的其他本地路由器之间维持通信。 Thus, the method and system of the present invention is to establish an initial network configuration, the other between a plurality of local SAN local router and a plurality of remote routers or more remote SAN or SAN in the same maintaining communication between the local router.

封装I/O的步骤还可以包括封装给定的光纤通道、SCSI或其他协议的特定命令和消息。 Package I / O may further comprise the step of commands and messages specific to a given package Fiber Channel, SCSI, or other protocols. 例如,可以封装任务管理功能、错误恢复功能和其他I/O处理功能。 For example, task management package, error recovery, and other I / O processing. 此外,本发明的方法和系统可以方便地扩展到为一些可以添加到现有协议上的附加命令和消息提供同样的封装功能。 Further, the method and system of the invention can easily be extended to a number of additional commands may be added to the existing protocols and messages to provide the same function package.

本发明提供了将一个SAN模式扩展到距离远大于当前光纤通道协议所允许跨距的多个SAN的能力。 The present invention provides an extended distance model SAN Fiber Channel protocol is much greater than the current allowable span of a plurality of SAN capacity. 本发明提供了互联一些处在诸如不同城市之类的不同地理位置的SAN的能力,使得它们能以无缝方式进行工作,就好象它们构成了一个单独的本地SAN那样。 The present invention provides a number of different locations such as in different cities like SAN interconnect ability, so that they can operate in a seamless manner, as if they constitute a single local SAN like. 此外,就存储恢复来说,本发明允许一个备份库驻留在远地,从而在本地由于某种故障或灾祸而受破坏的情况下也能保证数据的完整性。 Furthermore, storage recovery, the present invention allows a remote backup repository resides in, so in the case of failure due to some local or disrupted scourge also ensures data integrity.

虽然以上结合例示性的实施例对本发明作了详细说明,但应理解,所作的说明只是例示性的而不是限制性的。 Although the above embodiments in conjunction with an exemplary embodiment of the present invention were described in detail, it should be understood that the description made merely illustrative and not restrictive. 因此,可以理解,本发明的这些实施例的细节的各种改变和本发明的其他实施方式都是显而易见的,可由熟悉本技术领域的人员参考本说明实现。 Thus, it is understood that such detail is an embodiment of the present invention, various modifications and other embodiments of the invention are apparent by those skilled in the art with reference to this description achieved. 所有这样的改变和其他的实施方式都应在所附权利要求书给出的本发明的实质和范围之内。 All such modifications and other embodiments of the appended claims should be within the spirit and scope of the invention being set forth.

Claims (45)

  1. 1.一种为在两个或更多个节点之间通过一个基于分组的网络进行数据传输而封装SCSI协议的方法,所述方法包括在每个节点执行的下列步骤:(a)标识所述网络上所有其他可用节点和连接到每个所述可用节点上的远地设备;(b)表示一个或多个所述远地设备,使得它们对于一个或多个本地主机成为可用的;(c)封装在一个或多个所述本地主机与一个或多个所述远地设备之间的一个输入/输出(I/O)阶段;以及(d)为后续的I/O阶段重复步骤(c)。 1. A method for data transmission via a packet-based network between two or more nodes encapsulated SCSI protocol, said method comprising the following steps performed in each node: (a) identifying the All other nodes on the network is available and connected to each of the available remote node apparatus; (b) represents one or more remote device, such that they for one or more local hosts becomes available; (c ) enclosed between an input of one or more of the local hosts and one or more remote device / output (I / O) stage; and (d) for subsequent I / O stage of repeating steps (c ).
  2. 2.权利要求1的方法,其中所述输入/输出阶段包括一个命令阶段、一个数据阶段和一个响应阶段。 2. The method of claim 1, wherein said input / output stage includes a command phase, a data phase and a response phase.
  3. 3.权利要求1的方法,其中封装所述I/O阶段包括封装光纤通道或SCSI协议的一个专用命令。 The method of claim 1, wherein the package I / O phase includes a special command to package the Fiber Channel or SCSI protocol.
  4. 4.权利要求3的方法,其中所述专用命令是一个任务管理功能、错误恢复功能或其他I/O处理功能。 The method of claim 3, wherein said specific command is a task management, error recovery, or other I / O processing.
  5. 5.权利要求1的方法,其中所述两个或更多个节点中每一个以通信方式分别连接到一个存储区域网(SAN)上。 The method of claim 1, wherein said two or more nodes each communicatively connected to a storage area network (SAN) on.
  6. 6.权利要求5的方法,其中所述两个或更多节点中每一个都是在各自的SAN与所述基于分组的网络之间的一个接口。 The method of claim 5, wherein said two or more nodes are each in a respective SAN based on an interface between the packet network.
  7. 7.权利要求5的方法,其中所述SAN之一是一个备份库。 The method of claim 5, wherein one of the SAN is a backup repository.
  8. 8.权利要求1的方法,其中每个所述节点都是一个光纤通道-SCSI路由器。 1 wherein each of said nodes is a Fiber Channel -SCSI claim router.
  9. 9.权利要求1的方法,其中所述SCSI协议是一个光纤通道SCSI协议。 9. The method of claim 1, wherein the fiber channel SCSI protocol is a SCSI protocol.
  10. 10.权利要求1的方法,其中所述基于分组的网络是一个异步传输模式(ATM)网、以太网、IP网或SONET网。 10. The method of claim 1, wherein the packet-based network is an asynchronous transfer mode (ATM) network, Ethernet, IP networks or SONET networks.
  11. 11.权利要求1的方法,其中所述基于分组的网络是一个广域网(WAN)。 11. The method of claim 1, wherein said network is a packet-based wide area network (WAN).
  12. 12.权利要求1的方法,其中所述基于分组的网络是一个专用链路。 12. The method of claim 1, wherein said packet-based network is a dedicated link.
  13. 13.权利要求1的方法,其中所述基于分组的网络是一个交换网。 13. The method of claim 1, wherein the packet-based network is a switched network.
  14. 14.权利要求1的方法,其中所述表示步骤还包括下列步骤:将一个或多个连接到一个节点上的所述远地设备各自的本地地址映射为一个中间地址;以及在另一个节点将每个所述中间地址映射为一个远地地址。 14. The method of claim 1, wherein the representation further comprises the step of the steps of: connecting to one or more of said remote devices on a respective local node address mapping for an intermediate address; and in another node each of said intermediate address mapping for a remote address.
  15. 15.权利要求1的方法,其中所述封装步骤还包括下列步骤:将所述I/O阶段从所述SCSI协议变换到一个与所述基于分组的网络相关的协议;以及在一个远地节点将所述I/O阶段变换回所述SCSI协议。 15. The method of claim 1, wherein said encapsulating step further comprises the step of: the I / O conversion stage to a SCSI protocol from the packet-based network associated with the protocol; and a remote node the I / O phase transformation back to the SCSI protocol.
  16. 16.权利要求15的方法,其中与所述基于分组的网络相关的所述协议是一个异步传输模式(ATM)协议、以太网协议、IP协议或SONET协议。 15 The method of Ethernet protocol, IP protocol or SONET protocol as claimed in claim 16, wherein said protocol associated with the packet-based network is an asynchronous transfer mode (ATM) protocol.
  17. 17.权利要求1的方法,其中所述标识步骤还包括通过一个公共服务器动态地发现所有其他可用节点和连接到所述可用节点上的设备。 17. The method of claim 1, wherein said step of identifying further includes a discovery and all other available nodes connected to the nodes available through a public server apparatus dynamically.
  18. 18.权利要求17的方法,其中所述两个或更多个节点中至少有一个被指定为所述公共服务器。 18. The method of claim 17, wherein said two or more nodes are designated as at least one of said common server.
  19. 19.权利要求17的方法,其中所述公共服务器是一个与所述节点分开的单独设备。 19. The method of claim 17, wherein the public server is a separate device separate from the node.
  20. 20.权利要求17的方法,所述方法还包括通过一个脉搏消息在所述公共服务器确定是否有一个节点从所述网络撤出。 20. The method of claim 17, the method further comprises determining whether there is a node in the network withdrawn from the common server via a pulse message.
  21. 21.权利要求1的方法,其中所述基于分组的网络是任何允许数据分组在节点之间流通的网络。 21. The method of claim 1, wherein said packet-based network to allow any flow of data packets between network nodes.
  22. 22.权利要求1的方法,其中所述两个或更多个节点可以各自采用不同网络协议以通信方式连接到一个SAN上。 22. The method of claim 1, wherein said two or more nodes are each using different network protocols may be communicatively coupled to a SAN.
  23. 23.一种为在两个或更多个节点之间通过一个基于分组的网络进行数据传输而封装SCSI协议的系统,所述系统在每个节点包括:(a)用来标识所述网络上所有其他可用节点和连接到每个所述可用节点上的远地设备的指令;(b)用来表示一个或多个所述远地设备的指令,使得它们对于一个或多个本地主机成为可用的;(c)用来封装在一个或多个所述本地主机与一个或多个所述远地设备之间的一个输入/输出(I/O)阶段的指令;以及(d)用来为后续I/O阶段重复步骤(c)的指令。 23. A method for data transmission via a packet-based network between two or more nodes encapsulated SCSI protocol, the system comprising at each node: (a) for identifying the network and all other available nodes connected to the remote command device on each of the available node; (b) is used to represent one or more of said remote command device, such that they for one or more local hosts becomes available the; (c) used in one or more of a local host and a remote device between the input / stages or more of said instruction output (I / O) package; and (d) used to repeat step instructions subsequent I / O stage (c) is.
  24. 24.权利要求23的系统,其中所述输入/输出阶段包括一个命令阶段、一个数据阶段和一个响应阶段。 24. The system of claim 23, wherein said input / output stage includes a command phase, a data phase and a response phase.
  25. 25.权利要求23的系统,其中所有的指令存储在每个所述节点内的存储器中。 25. The system of claim 23, wherein the memory stores all of the instructions in each said node.
  26. 26.权利要求23的系统,其中所述用来封装所述I/O阶段的指令包括用来封装光纤通道或SCSI协议的一个专用命令的指令。 26. The system of claim 23, wherein the package for the I / O stage comprises instructions specific commands used to encapsulate a fiber channel or SCSI protocol.
  27. 27.权利要求26的系统,其中所述专用命令是一个任务管理功能、错误恢复功能或其他I/O处理功能。 27. The system of claim 26, wherein said specific command is a task management, error recovery, or other I / O processing.
  28. 28.权利要求23的系统,所述系统还包括所述两个或更多个节点以通信方式分别连接的存储区域网(SAN)。 28. The system as claimed in claim 23, said system further comprises a storage area network (SAN) of said two or more nodes communicatively connected respectively.
  29. 29.权利要求28的系统,其中所述两个或更多个节点中每一个都是在各自的SAN与所述基于分组的网络之间的一个接口。 29. The system of claim 28, wherein said two or more nodes are each in a respective SAN based on an interface between the packet network.
  30. 30.权利要求28的系统,其中所述SAN中至少有一个是一个备份库。 30. The system of claim 28, wherein at least one of the SAN is a backup repository.
  31. 31.权利要求23的系统,其中每个所述节点都是一个光纤通道-SCSI路由器。 System 23 wherein each of said nodes is a fiber channel 31. The router as claimed in claim -SCSI.
  32. 32.权利要求23的系统,其中所述SCSI协议是一个光纤通道SCSI协议。 32. The system of claim 23, wherein the fiber channel SCSI protocol is a SCSI protocol.
  33. 33.权利要求23的系统,其中所述基于分组的网络是一个异步传输模式(ATM)网、以太网、IP网或SONET网。 33. The system of claim 23, wherein said packet-based network is an asynchronous transfer mode (ATM) network, Ethernet, IP networks or SONET networks.
  34. 34.权利要求23的系统,其中所述基于分组的网络是一个广域网(WAN)。 34. The system of claim 23, wherein said network is a packet-based wide area network (WAN).
  35. 35.权利要求23的系统,其中所述基于分组的网络是一个专用链路。 35. The system of claim 23, wherein said packet-based network is a dedicated link.
  36. 36.权利要求23的系统,其中所述基于分组的网络是一个交换网。 36. The system of claim 23, wherein said packet-based network is a switched network.
  37. 37.权利要求23的系统,其中所述用来表示的指令还包括:用来将一个或多个连接到一个节点上的所述远地设备各自的本地地址映射为一个中间地址的指令;以及用来在另一个节点将每个所述中间地址映射为一个远地地址的指令。 37. The system as claimed in claim 23, wherein said instruction is used to indicate further comprising: means for connection to one or more of said remote devices on a local node of a respective instruction address mapping for an intermediate address; and for mapping each said intermediate instruction address is the address of a far node in another.
  38. 38.权利要求23的系统,其中所述用来封装的指令还包括:用来将所述I/O阶段从所述SCSI协议变换到一个与所述基于分组的网络相关的协议的指令;以及用来在一个远地节点将所述I/O阶段变换回所述SCSI协议的指令。 38. The system of claim 23, wherein the instructions for encapsulating further comprises: for the I / O conversion stage from the SCSI command protocol to a packet-based protocol associated with the network; and to a remote node, the I / O command back to the phase conversion in SCSI protocol.
  39. 39.权利要求38的系统,其中与所述基于分组的网络相关的所述协议是一个异步传输模式(ATM)协议、以太网协议、IP协议或SONET协议。 The system 38 of the Ethernet protocol, IP protocol or SONET protocol as claimed in claim 39., wherein said protocol associated with the packet-based network is an asynchronous transfer mode (ATM) protocol.
  40. 40.权利要求23的系统,所述系统还包括一个公共服务器,而且其中,所述用来标识的指令还包括用来通过所述公共服务器动态地发现所有其他可用节点和连接到所述可用节点上的设备的指令。 40. The instruction of claim 23, said system further comprising a common server, and wherein said means for identifying further comprises means for discovery and all other available nodes connected to the common node through the available server dynamically the instruction device.
  41. 41.权利要求40的系统,其中所述两个或更多个节点中至少有一个被指定为所述公共服务器。 41. The system of claim 40, wherein said two or more nodes are designated as at least one of said common server.
  42. 42.权利要求40的系统,其中所述公共服务器是一个与所述节点分开的单独设备。 42. The system of claim 40, wherein the public server is a separate device separate from the node.
  43. 43.权利要求40的系统,所述系统还包括用于一个脉搏消息的指令,通过所述脉搏消息在所述公共服务器确定是否有一个节点从所述网络撤出。 43. The system as claimed in claim 40, said system further comprises a pulse message instructions for determining whether there is a node in the network withdrawn from the server via the common pulse message.
  44. 44.权利要求23的系统,其中所述基于分组的网络是任何允许数据分组在节点之间流通的网络。 44. The system of claim 23, wherein said packet-based network to allow any flow of data packets between network nodes.
  45. 45.权利要求23的系统,其中所述两个或更多个节点可以各自采用不同网络协议以通信方式连接到一个SAN上。 45. The system of claim 23, wherein said two or more nodes are each using different network protocols may be communicatively coupled to a SAN.
CN 00817156 1999-11-12 2000-11-13 Encapsulation protocol for linking storage area networks over packet-based network CN1409903A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16519499 true 1999-11-12 1999-11-12

Publications (1)

Publication Number Publication Date
CN1409903A true true CN1409903A (en) 2003-04-09

Family

ID=22597864

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 00817156 CN1409903A (en) 1999-11-12 2000-11-13 Encapsulation protocol for linking storage area networks over packet-based network

Country Status (5)

Country Link
EP (1) EP1240575A2 (en)
JP (1) JP2003517765A (en)
CN (1) CN1409903A (en)
CA (1) CA2391353A1 (en)
WO (1) WO2001044891A3 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101116069B (en) 2005-03-28 2011-01-26 思科技术公司 Zone based quality of service in a fibre channel fabric
CN1806466B (en) 2003-04-16 2011-05-18 英特尔公司 Architecture, method and system of multiple high-speed servers for WDM based photonic burst-switched networks
CN101494643B (en) 2008-01-24 2012-03-28 中国移动通信集团公司 Method, device and system for exiting distributed wireless access system

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2372600B (en) * 2001-02-27 2003-02-19 3Com Corp Network area storage block and file aggregation
US7114009B2 (en) 2001-03-16 2006-09-26 San Valley Systems Encapsulating Fibre Channel signals for transmission over non-Fibre Channel networks
US6968369B2 (en) * 2001-09-27 2005-11-22 Emc Corporation Remote data facility over an IP network
US7020814B2 (en) * 2003-03-18 2006-03-28 Cisco Technology, Inc. Method and system for emulating a Fiber Channel link over a SONET/SDH path
US6925533B2 (en) 2003-05-07 2005-08-02 International Business Machines Corporation Virtual disk image system with local cache disk for iSCSI communications
US7586942B2 (en) 2003-12-09 2009-09-08 Dell Products L.P. Identifying host computers at a physical layer
US7634582B2 (en) 2003-12-19 2009-12-15 Intel Corporation Method and architecture for optical networking between server and storage area networks
US7447788B2 (en) 2004-01-27 2008-11-04 Dell Products L.P. Providing host information to devices in multi SCSI transport protocols
US8886778B2 (en) * 2005-04-29 2014-11-11 Netapp, Inc. System and method for proxying network management protocol commands to enable cluster wide management of data backups
JP5203041B2 (en) * 2008-05-22 2013-06-05 エヌイーシーコンピュータテクノ株式会社 Network system, a network connection method, the connection device, connecting the card

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5519695A (en) * 1994-10-27 1996-05-21 Hewlett-Packard Company Switch element for fiber channel networks
US5768623A (en) * 1995-09-19 1998-06-16 International Business Machines Corporation System and method for sharing multiple storage arrays by dedicating adapters as primary controller and secondary controller for arrays reside in different host computers
JPH09251437A (en) * 1996-03-18 1997-09-22 Toshiba Corp Computer device and continuous data server device
US5959994A (en) * 1996-08-19 1999-09-28 Ncr Corporation ATM/SONET network enhanced as a universal computer system interconnect
US5925119A (en) * 1997-03-28 1999-07-20 Quantum Corporation Computer architecture for automated storage library
US5941972A (en) * 1997-12-31 1999-08-24 Crossroads Systems, Inc. Storage router and method for providing virtual local storage

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1806466B (en) 2003-04-16 2011-05-18 英特尔公司 Architecture, method and system of multiple high-speed servers for WDM based photonic burst-switched networks
CN101116069B (en) 2005-03-28 2011-01-26 思科技术公司 Zone based quality of service in a fibre channel fabric
CN101494643B (en) 2008-01-24 2012-03-28 中国移动通信集团公司 Method, device and system for exiting distributed wireless access system

Also Published As

Publication number Publication date Type
EP1240575A2 (en) 2002-09-18 application
CA2391353A1 (en) 2001-06-21 application
WO2001044891A2 (en) 2001-06-21 application
JP2003517765A (en) 2003-05-27 application
WO2001044891A3 (en) 2002-01-24 application

Similar Documents

Publication Publication Date Title
US6560236B1 (en) Virtual LANs
US6865610B2 (en) Wire protocol for a media server system
US6456603B1 (en) Method of supporting communications mobility in a telecommunications system
US6061807A (en) Methods systems and computer products for error recovery of endpoint nodes
US6359894B1 (en) Remote communications server system
US7197047B2 (en) Method and apparatus for transferring data between IP network devices and SCSI and fibre channel devices over an IP network
US6873620B1 (en) Communication server including virtual gateway to perform protocol conversion and communication system incorporating the same
US7035281B1 (en) Wireless provisioning device
US6738824B1 (en) Dial-out link selection via static route redistribution
US6157649A (en) Method and system for coordination and control of data streams that terminate at different termination units using virtual tunneling
US6061728A (en) Arrangement for controlling network proxy device traffic on a transparently-bridged local area network using a master proxy device
US20060262734A1 (en) Transport protocol connection synchronization
US5774660A (en) World-wide-web server with delayed resource-binding for resource-based load balancing on a distributed resource multi-node network
US7386628B1 (en) Methods and systems for processing network data packets
US20080159260A1 (en) Fibre channel over ethernet frame
US6229809B1 (en) Method and system for combining computer network protocols
US6871347B2 (en) Method and apparatus for facilitating load balancing across name servers
US20020101875A1 (en) Spanning tree alternate routing bridge protocol
US6178181B1 (en) Mapping function and method of transmitting signaling system 7(SS7) telecommunications messages over data networks
US5619645A (en) System isolation and fast-fail
US6920116B1 (en) System and method for automatically configuring network service entity identifiers utilizing a Gb-over-IP interface in a GPRS network
US20050008013A1 (en) Embedded management channel for SONET path terminating equipment connectivity
US20020114328A1 (en) Apparatus for linking a SAN with a LAN
US6320874B1 (en) Establishing and terminating connections in a mixed protocol network
EP0567217A2 (en) System of extending network resources to remote networks

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C02 Deemed withdrawal of patent application after publication (patent law 2001)