KR100416509B1 - Method for controlling switch connection to accomodate the optical network in open switching system - Google Patents

Abstract

The present invention relates to a switch connection control method and apparatus therefor in an open switching system for accommodating an optical network, comprising: (a) detecting a link failure from information due to optical loss; (C) determining a recovery range of the link failure; (d) resetting the link connection according to the recovery range, so that an open system mainly for resilience in an optical network is provided. Even if the optical network accommodates the switch, the switch connections and services are handled smoothly.

Description

Method for controlling switch connection and accomodate the optical network in open switching system in open switching system for accommodating optical network

The present invention relates to the field of switch connection control of an exchange system, and more particularly, to a switch control method and apparatus therefor in an open exchange system for optical network acceptance using a general switch management protocol (GSMP). .

Conventionally, in order to establish a connection to an optical switch, when an optical lightpath setup is required from an upper router control, a network device (NE) receiving this command receives a WADM (Wavelength Add / Drop Multiplexor) controls on / off of the optical path. This process is based on user network interface (UNI, User Network Interface) protocol that the upper control plan is matched with client (host or router), that is, it does not share resource information in the network. There are constraints and performance may vary depending on the capabilities of the underlying switches (eg routers, DWDM systems, ADMs, PXCs, OXCs, etc.). That is, according to the lower switch, the upper IP controller may perform a different switch setting method and control method to the lower NE so that the quick switch setting is not performed.

The present invention has been made in an effort to provide a GSMP switch connection control method and apparatus using a fast recovery capability processing procedure to accommodate an optical network in an open switching system to solve the above problems.

Another object of the present invention is to provide a computer-readable recording medium having recorded thereon a program for executing the method on a computer.

1 is a related block diagram for processing switch connection control using GSMP of an open optical switching system according to the present invention.

2 is a diagram illustrating a flow in a switch connection control method in an open switching system for accommodating an optical network according to the present invention.

FIG. 3 is a diagram illustrating a flow of detecting a link failure in an optical cross connector (OXC).

FIG. 4 is a diagram illustrating a flow for fault notification informing a node capable of restoring a failure.

FIG. 5 is a diagram illustrating a flow of determining a recovery range of a failure through a recognized failure and determining a bypass route through an Open Shortest Path First (OSPF) or an Intermediate System to Intermediate System (IS-IS) protocol.

6 is a diagram illustrating a flow of resetting a switch connection using GSMP.

Switch connection control method in an open switching system for accommodating an optical network according to the present invention for achieving the above object, (a) detecting a link failure from information due to optical loss; (b) recovering the link failure (C) determining a recovery range of the link failure; (d) resetting a link connection according to the recovery range.

Switch connection control apparatus in an open exchange system for accommodating an optical network according to the present invention for achieving the above object, the optical service control unit for performing the IP service, and a switch master unit for controlling the switch connection through a general switch management protocol In the open switching system, the optical service control unit, a routing unit for storing IP routing information; a signaling unit for setting a channel for transmitting and receiving the IP routing information; a link management protocol unit for managing the channel; switch for the IP service And a general switch management protocol control unit for connecting the GS master slave protocol processing unit to analyze the type of the connection management message received from the general switch management protocol control unit; A switch connection control unit managing a connection state of the switch according to the type of the connection management message analyzed; And a switch port state management unit for transmitting shape information of the open exchange system to the general switch management protocol control unit.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the present invention.

The explosive growth of IP services requires performance, survivability, and scalability. To meet these demands, the Internet Engineering Task Force (IETF) recommends Multiprotocol Label Switching (MPLS), and many Request For Comments (RFCs) and drafts are available. MPLS has many advantages, such as reducing the load on the core network, ensuring scalability with increasing IP services, and fast packet forwarding due to a simple labeling mechanism. default) has established itself as a useful protocol to implement MPLS. MPLS can be applied when the underlying network is Asynchronous Transfer Mode (ATM), Frame Relay (FR), Point-to-Point Protocol (PPP), and the like. In the case of a real SONET / SDH network, due to many problems such as scalability and cost of equipment, standardization of the optical network is currently performed by IETF, Optical Interworking Forum (OIF), and Optical Domain (ODSI). Service Interconnect (ITU-T) and International Telecommunications Union-Telecommunications sector (ITU-T). In order to construct an optical network such as Dense Wavelength Division Multiplexing (DWDM) as a substructure, it is difficult to realize the pure optical layer alone, considering the upper control plane and the lower switch plane. Network implementation is required.

The IETF Common Control and Measurement Plane (WAMP) Working Group is currently working on Generalized Multi-Protocol Label Switching (GMPLS) for generalized control plans. GMPLS is independent of any elements, including routers, switches, DWDM systems, Add-Drop Multiplexers (ADMs), photonic cross-connects (PXCs), and OXCs that make up a data or transport network. A protocol for predicting resources and surviving networks. As such, there is a need for a standard open interface that matches a switch plan such as multiplexers and switches of an optical layer, which is a control layer and a lower physical layer. OIF, including ITU-T, is moving towards using the General Switch Management Protocol (GSMP), a standard open interface proposed by the Multi-service Switching Forum (MSF). The IETF's GSMP WG is also working on further work to accommodate optical networks, but details are not covered. Therefore, in the present invention, even when the optical network is accommodated in an open system using the switch control method of the GSMP for the optical network network, the switch connection and the service are smoothly focused on the restoration capability that should be the most important in the optical network. It is intended to be processed.

1 is a block diagram illustrating a switch connection control process using a general switch management protocol (hereinafter referred to as GSMP) of an open optical switching system according to the present invention.

An optical switching system is a system for processing IP services, and any switch fabric (eg, Wavelength Division Multiplexor (WDM), Photonic Cross Connect (PXC), Digital Cross Connect (DXC), Optical Cross Connector (OXC)) at the bottom. If this is implemented (in the present invention, OXC capable of Optical Electronic Optical Conversion (OEO) is taken as an example), by a generalized multi-protocol label switching (hereinafter referred to as GMPLS) by a control plane Receive switch connection information and perform switch connection using GSMP.

Optical Service Controller (OSC, 110) performs IP service through GMPLSU (GMPLS Unit), GMPLSU is a routing module for storing and advertising IP routing protocol information (Routing Module, RTEM, 111) , A signaling module (SIGM, 113) and a channel (control channel and channel) for setting a control channel for transmitting and receiving IP routing protocol information between switch nodes and a data channel through which data flows; It consists of a link management unit (LMPM) that performs data channel) management. In addition, the optical sevis control unit 110 includes a GSMP control unit 117 serving as a master of the GSMP for the switch connection.

The switch master unit (SMU) 120 is a GSMP slave processor that controls the connection of a switch through the GSMP protocol. The connection management messages received from the GSMP controller 117 (generally called a GSMP master) are connected to each other. ) SMP slave protocol processing unit 121 analyzes which type represents the connection setting, release, verify or modify the switch, and the switch connection controller 123 switches according to the transmitted message type. After searching / adding / deleting / updating the switch connection status, when requesting switch connection / disconnection to the device interface (131) to connect the switch, the device interface (131) sends it to the optical wavelength of the switch. Translate to Add / Drop to connect the switch. The switch port state management unit 125 transmits the system shape information to the GSMP control unit 117 and performs asynchronous events such as link failure.

2 is a diagram illustrating a flow in a switch connection control method in an open switching system for accommodating an optical network according to the present invention.

In step 210, a failure detection for detecting a failure of an optical channel or a link is detected. Link failure should basically be detected at the nearest layer where the failure occurred. In the optical network, the optical layer, which is a physical layer, corresponds to this, and one of the failure measures uses a loss of light (LOL). This is because the detection mechanism for the channel / link has not yet been considered.

In order to recover the link failure, a fault notification is performed (step 220). Notification of failures may be considered in terms of direct and indirect failure notices. The direct notification is when the detection and notification of failure occur at the same node at the same time, and the indirect case is when the failure is reported using information reported from downstream nodes. This can be done using the GMPLS protocol, Link Management Protocol (hereinafter referred to as LMP), which means that attenuation or loss of light at the input port indicates a failure of the link / channel. ) And through the control channel already established, the upper node can also search for the input link / channel related to the reported output link / channel and report the failure to the next higher node to determine the node where the actual failure started. . The above items are related to the data channel through which the actual data flows, and if the control channel fails, it is allocated when setting the path using Notify Message of RSVP-TE (Resource reserVation Protocol for Traffic Engineering) instead of LMP. ) Use the recoverable IP address (the address assigned to the actual OXC) to inform the node of the failure. The node notified of the failure uses the refresh message to reset the actual failure channel and update the existing information. In the case of LMP, when down of a control channel is reported to an upper node by setting a ControlChannelDown flag using another effective control channel between two nodes, the upper node no longer sends a Hello message between the two nodes. This allows the control channel to fail more quickly than the actual internal RSVP signal message or other Interior Gateway Protocol (IGP) messages, which can cause switch over to another valid control channel before the LSP (Label Switched Path) level failure occurs. It is possible that no reset of the LSP level is required.

When a failure is notified, the perceived failure determines the extent of recovery of the failure and an alternate route is determined using the routing information advertised through the OSPF or IS-IS protocols. Step 230). Determination of the recovery range of failure may determine whether an unnumbered link, link, or channel failure is performed based on an interface identifier between nodes based on the failure message reported by the lower node. This failure can be switched over if there is a link that is not yet in use between the same two nodes where this failure occurred. If the failure is for a channel and there is no redundant link, it will bypass the destination. Reset the signaling at the same time as recalculating the route. To know the extra link information, IGMP-related extended length length value (TLV) is used in GMPLS. That is, among the extended information, LPT (Link Protection Type) TLV indicates protection capability on a link, which advertises the characteristics of the link with information required for path computation at the source node. Types include Extra Traffic, Unprotected, Shared, Dedicated 1: 1, Dedicated 1 + 1, Enhanced, Reserved.If shared or Dedicated 1: 1 is used, a dedicated disjoint link that protects this link is actually used. link is not currently used as a data link passing through this node because it is not advertised as routing information and thus is not used when calculating a path at the source. Therefore, using this link to recover from a failure by resetting an unnumbered link, link, or channel that failed, can save much more latency than notifying and recovering from the failure to the source node. However, if there are no dedicated links or channels for protection, then the recalculation of the paths for rerouting must be done and a signal setup must be made.

Finally, the reconnection of the switch is established using the actual GSMP (step 240). Failures reported either directly or indirectly and LSPs configured on the failed link must be reestablished. This is because the service connection block of the switch controller part that manages the connection removes the existing LSP information, configures new LSP information, and requests a service connection to the GSMP master block, and the GSMP master block transmits the underlying transmission network (Optical). Networks can be subdivided according to optical switching methods, ie DWDM, Optical Bursting Switching (OBS), Optical Packet networks, etc., which affect the recovery procedure by the control scheme of each packet considering them. Regardless, if the service connection command is given to the GSMP slave block, the GSMP slave block actually requires an LSP connection to the relevant switch.

FIG. 3 is a diagram illustrating a flow of detecting a link failure in an optical cross connector (OXC).

Since link failure should basically be detected at the layer closest to where the failure occurred, the optical communication network receives a loss of light from the device interface 131 corresponding to the optical layer (physical layer) (step 311). The switch port state management unit 125 identifies the failure of the wavelength or the link (step 312). The perceived wavelength is updated by the switch port state manager 125 to the current state (step 313), and correlates with it, i.e. if the failed wavelength is at the output port, Extract the link, port, and wavelength (step 314). The failure type is set to 1 to classify the failure type as hardware because it recognizes that the failure is not caused by a failure message from a lower node, but occurs in its own node (step 315). Information with the specified failure link type is transmitted to the upper optical service control unit 10 (step 318) to perform subsequent processing.

FIG. 4 is a diagram illustrating a flow for fault notification informing a node capable of restoring a failure.

In case of indirectly reporting the occurrence of the failure from the subordinate node or receiving the failure notification directly from the GSMP slave of its own node (step 421), the reported failure information of the link, port, and channel (mapped to the relevant wavelength) is displayed. In operation 422, the current state of input ports, links, and channels correlated with each other is examined (operation 423). If the correlation channel is confirmed to be a failure (step 424), if it is a failure, the fact is reported to a higher node (step 425), and the failure is not a single channel but a bundle type failure (e.g., unnumbered, link), repeat this until all channels belonging to it are reported. If it is not a failure, notify the module that handles localization (step 426).

If a failure message of the control channel from the lower node is reported (step 427), the routing session and signaling control performed in the control channel are terminated (step 428), and the current active (active) Switch over to the other control channel (step 429).

FIG. 5 is a diagram illustrating a flow of determining a recovery range of a failure through a recognized failure and determining a bypass route through an Open Shortest Path First (OSPF) or an Intermediate System to Intermediate System (IS-IS) protocol.

When receiving a local resolution message indicating that a failure has occurred between the main node and a subordinate node from the optical service control unit 110 (step 531), the link is used for protection to recover the actual failed link or channel. The failure type (type) of the link is retrieved (step 532). If the fail type is unnumbered or a link type, an extra associated protection link may be used (step 539). However, if a failure is specified to be of channel type, then the link containing the channel must be known. Relevant protection information is extracted from the managed IGP routing information using the recognized link number (step 533). If the extracted protection type information is dedicated (dedicated) 1: 1 or shared (step 534), and if the dedicated 1: 1 or shared, the associated link, port, channel is extracted (535). step). Update the newly extracted related links, ports, and channels to the existing ones. In step 536, a signal message is transmitted to the GSMP controller 117 to reestablish the connection using this new information. If there is no redundant link or channel, the path must be recalculated (step 537) to the destination by the path calculation algorithm supported by the current node because it now needs to find a new path from the current node. After the path is calculated, the path is set up using the signaling protocol supported by the current node to set the path, and then the data channel is set according to the switch capability for data flow (step 538).

6 is a diagram illustrating a flow of resetting a switch connection using GSMP.

In GSMP, several types of messages are handled. In case of link / channel failure, the failed connection must be deleted and a newly reestablished connection must be established. 6 shows a process of receiving a connection establishment / release message required by the GSMP control unit 117 (step 641) in the GSMP slave protocol processing unit 121. First, when the message type is connection release (delete), Based on the input port information shown in the message, the output port information is extracted, and a switch release request is requested to the device interface 131. The device interface 131 recognizes this as a wavelength drop in the Optical switch and controls the switch to drop the associated wavelength.

If it is a connection setup, extract the switch port to be physically connected and the optical wavelength that maps to the required channel (step 642), and check that the switch port is in the normal state (step 643). Since it cannot be set, the error is transmitted to the upper control block through the GSMP controller 117 (step 646). If it is normal, check whether the wavelength of the input port to be reset is set (step 644), and if it is already set, it is not a wavelength that can be reset, so the error is transmitted to the upper control block through the GSMP controller 117. (Step 646). In addition, if the wavelength of the output port is already set, it is not a wavelength that can be reset, and thus transmits an error to the upper control block through the GSMP controller 117 (step 646). If the connection to be established is not already established, the additional information on the input / output port of the connection to be set is extracted and configured (step 647), and the device interface 131 requests a connection establishment (step 648). The device interface 131 recognizes this as an Add command for the wavelength of optical switching and controls the switch.

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, hard disk, floppy disk, flash memory, optical data storage device, and also carrier waves (for example, transmission over the Internet). It also includes the implementation in the form of. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

As described above, according to the present invention, GMPLS is actually a Packet Switch Capable (PSC), Layer-2 Switch Capable (L2SC), Time-Division Multiplex Capable (TDM), Lambda Switch Capable interfaces (LSC), or Fiber (FSC). As a generalized control plane protocol designed to accommodate many interfaces such as switch capable interfaces, the upper control plane and lower data that can handle any switch capable in the lower part There is a need for a protocol to match data planes, which requires GSMP similar to the processing in open switching systems in existing ATM networks. In the optical network, this open switch control protocol can be used to meet various service needs of various vendors and network operators without being limited by the lower optical switch capability, thereby meeting various service needs of various vendors and network operators. In order to recover from failures that are most important in optical networks, the GSMP accommodates the information processing functions for restoration, thereby efficiently recovering.

Claims (16)

(a) detecting a link failure from information due to optical loss; (b) announcing the link failure to recover the link failure; (c) determining a recovery range of the link failure; And and (d) resetting the link connection according to the recovery range. The method of claim 1, wherein step (a) (a1) receiving information on optical loss and recognizing a port and a wavelength at which a link failure occurs; (a2) extracting a correlation port and a wavelength for the recognized port failure port and wavelength; And (a3) a switch connection control method for accommodating an optical network in an open switching system, comprising the step of designating the extracted information and the failure type. The method of claim 2, And updating the state of the recognized wavelength after the step (a1). The method of claim 1, wherein step (b) In step (a), if information on link failure is received in the same node, (b1) recognizing a port and a channel where a link failure occurs; And (b2) a switch connection control method for accommodating an optical network in an open switching system, comprising the step of examining the correlated ports and channels of the recognized ports and channels. The method of claim 4, wherein in step (b2), If it is determined that link failure occurs in the correlation port and channel, switch information control method for accommodating an optical network in an open switching system, characterized in that it further comprises the step of transmitting the information to the upper node. The method of claim 4, wherein in step (b2), If it is confirmed that the link port and the channel is not a link failure, the switch connection control method for accommodating an optical network in the open switching system, characterized in that it further comprises the step of transmitting a message for local resolution processing. The method of claim 1, wherein step (b) In step (a), if information on a link failure is received at a lower node, (b1`) terminating the routing session and signaling control performed in the channel where the link failure has occurred; And and (b2`) exchanging a channel in which the link failure has occurred to a valid control channel. The method of claim 1, wherein step (c) (c1) if it is determined that a failure type is designated by receiving the message for local resolution processing of claim 6, extracting relevant protection information from internal gateway protocol routing information; (c2) if the related protection information is dedicated or shared, extracting a related fork and a channel; (c3) a switch connection control method for accommodating an optical network in an open switching system, comprising transmitting a signaling message including the extracted information. The method of claim 8, wherein in step (c2), (c2-1) calculating a path to a destination to which data is transmitted, if the associated protection information is not dedicated or shared; and (c2-1) setting the calculated path using a signaling protocol and setting a channel for data flow. The method of claim 1, wherein step (c) If it is determined that the failure type is not specified by receiving the message for the local solution processing of claim 6, the optical network in the open switching system, characterized in that it further comprises the step of connecting using the redundant protection link. Switch connection control method for accommodating. The method of claim 1, wherein step (d) (d1) receiving a connection establishment message to extract a switch port to be connected and a wavelength mapped to the required channel; (d2) checking the state of the switch port; (d3) if the state of the switch port is confirmed to be normal in the step (d2), checking whether the wavelength of the input / output port to be reset is set; (d4) extracting additional information of the input / output port if the wavelength of the input / output port is not set in the step (d3); And (d5) a switch connection control method for accommodating an optical network in an open switching system, comprising the step of requesting the setting of the switch according to the additional information. In the open exchange system having an optical service control unit performing an IP service, a switch master unit for controlling a switch connection through a general switch management protocol, The optical service control unit, A routing unit storing IP routing information; A signaling unit configured to set a channel for transmitting and receiving the IP routing information; A link management protocol unit for managing the channel; And It includes a general switch management protocol control unit for connecting a switch for the IP service, The switch master unit, A SMP slave protocol processor for analyzing a type of a connection management message received from the general switch management protocol controller; A switch connection control unit managing a connection state of the switch according to the type of the connection management message analyzed; And And a switch port state management unit for transmitting the shape information of the open switching system to the general switch management protocol control unit. The method of claim 12, wherein the signaling unit Switch control apparatus for accommodating an optical network in an open switching system, characterized in that for setting the control channel for controlling the transmission and reception of the routing information and the data channel for flowing the information. The method of claim 12, A type of connection management message is a switch connection control device for accommodating an optical network in an open exchange system, characterized in that the setting of the connection, disconnection, verification of the switch or change of the switch. The method of claim 12, wherein the switch port state management unit Switch connection control device for accommodating an optical network in the open switching system, characterized in that for transmitting the shape information of the open switching system to the general switch management protocol control unit and processes an asynchronous event including a link failure. A computer-readable recording medium having recorded thereon a program for executing the method of claim 1 on a computer.