JP2010278845A - Packet non-interrupt transmission system, packet non-interrupt switching device, and packet non-interrupt switching method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem wherein delay fluctuations occur, when switching a route in a conventional packet non-interrupt transmission system. <P>SOLUTION: In the packet non-interrupt transmission system provided with a transmission side for duplicating packets to which order identifiers are added and transmitting them via a plurality of routes and a reception side for selecting the packet arriving early from the packets to which the same identifier is added and which are received from among the plurality of routes and transferring it to the downstream, the packet non-interrupt transmission system includes a switch instruction means for giving an instruction to switch the packet to be transferred to the downstream on the reception side from the reception packet of a first route to the reception packet of a second route; and a packet delay means for gradually delaying the packet of the first route of a switching source; until the delay time becomes equal to or shorter than a preset threshold; when the delay time until the same packet as the packet received from the first route is received from the second route is longer than the threshold at the point of time at which the instruction is given from the switch instruction means and after. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a packet uninterruptible transmission system, a packet uninterruptible switching device, and a packet uninterruptible switching method that use a technique for duplicating a packet and transmitting it through a plurality of different paths.

  In recent years, as a high-reliability technology for packet communication, a packet on which an order identifier is inserted is duplicated on the transmitting side and transmitted in multiple routes, and the receiving side determines the arrival order of packets arriving from each route and arrives early. A packet transmission method in which packets are transmitted downstream has been considered (see, for example, Patent Document 1). In such a packet transmission method, even when a failure occurs in a part of the route and the packet of the route is not reached, if at least one route of the plurality of routes is normal, the route is switched to a normal route. Since packet loss can be avoided, there is an advantage that highly reliable packet communication can be realized.

Japanese Patent No. 4074268

  The packet uninterrupted transmission system as described above selects a packet that has arrived from a route of another system and forwards it downstream when a route failure occurs or when a route is switched by an instruction from the operation side. However, since the lengths of the transmission paths of the respective systems are different, a delay corresponding to the transmission path length difference occurs in the arrival time of the packet having the same identifier flowing through the transmission paths of the plurality of systems. As a result, a fluctuation that greatly changes the packet interval is caused at the time of path switching. In particular, in packet communication that requires real-time transmission such as streaming transmission, there is a problem that the occurrence of such fluctuation due to delay must be suppressed as much as possible.

  An object of the present invention is to provide a packet non-disruptive transmission system, a packet non-disruptive switching device, and a packet non-disruptive switching method that can minimize the occurrence of fluctuations during path switching.

  The packet non-disruptive transmission system according to claim 1 of the present invention includes a transmitting side that duplicates a packet to which an order identifier is added and transmits the same by a plurality of routes, and a packet that has the same identifier that is received from a plurality of routes. In the packet uninterrupted transmission system having a receiving side that selects an early arrival packet from the receiving side and transfers it downstream, the packet that is transferred downstream on the receiving side is changed from the received packet of the first route to the received packet of the second route. A switching instruction means for giving an instruction to switch, and a delay time until the same packet received from the first path after the time point instructed by the switching instruction means is set in advance from the second path Packet delay means for gradually delaying the packet of the switching source first path until the delay time becomes equal to or less than the threshold when the threshold is longer than the threshold And wherein the digit.

  As a result, when switching from short-path packet traffic with a short transmission path to long-path packet traffic with a long transmission path and a short delay time, the packet interval at the time of switching changes rapidly. The occurrence of fluctuation due to the difference in delay time between the short system and the long system can be suppressed.

  The packet non-disruptive transmission system according to claim 2 of the present invention is the packet non-disruptive transmission system according to claim 1, wherein the packet delay means includes the first route after the time point instructed by the switching instruction means. Delay determining means for determining a delay time until the same packet as the packet received from the second path is received from the second path; temporarily holding the received packet in the order of reception; When the delay adjustment memory to be read and the delay time determined by the delay determination means are larger than 0, the received packet of the first route of the switching source is read from the delay adjustment memory until the delay time becomes 0 or less. It is characterized by comprising delay control means for increasing the delay amount in proportion to time.

  As a result, when switching packet traffic from the short system to the long system, the packet interval gradually changes in proportion to the time, so even if the delay time difference between the short system and the long system is large, the occurrence of fluctuations can be suppressed. it can.

  The packet uninterrupted transmission system according to claim 3 of the present invention is the packet uninterrupted transmission system according to claim 1, wherein the packet delay means is configured to send the first route after the time point instructed by the switching instruction means. Delay determining means for determining a delay time until the same packet as the packet received from the second path is received from the second path; temporarily holding the received packet in the order of reception; When the delay adjustment memory to be read and the delay time determined by the delay determination means are larger than 0, the received packet of the first route of the switching source is read from the delay adjustment memory until the delay time becomes 0 or less. It is characterized by comprising delay control means for increasing the delay amount step by step by a predetermined time.

  As a result, when switching the packet traffic from the short system to the long system, the packet interval changes step by step for a predetermined time, so that the occurrence of fluctuation can be suppressed even when the delay time difference between the short system and the long system is large. .

  The packet non-disruptive transmission system according to claim 4 of the present invention is the packet non-disruptive transmission system according to claim 2 or 3, wherein the packet is switched from the received packet on the second route of the switching destination switched by the switching instruction means. Further provided is a switching cancellation instructing unit that gives an instruction to return to the received packet of the original first route, and the delay control unit performs the switching source first after the time when the cancellation instruction is given from the switching cancellation instructing unit. It is characterized in that the delay amount when reading the received packet of the route from the delay adjustment memory is decreased in inverse proportion to the time.

  As a result, even if the packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval gradually changes in inverse proportion to the time, so the delay time difference between the short system and the long system is large. However, the occurrence of fluctuation can be suppressed.

  The packet non-disruptive transmission system according to claim 5 of the present invention is the packet non-disruptive transmission system according to claim 2 or 3, wherein the packet is switched from the received packet of the second route of the switching destination switched by the switching instruction means. Further provided is a switching cancellation instructing unit that gives an instruction to return to the received packet of the original first route, and the delay control unit performs the switching source first after the time when the cancellation instruction is given from the switching cancellation instructing unit. It is characterized in that the delay amount when reading the received packet of the route from the delay adjustment memory is decreased step by step by a predetermined time.

  As a result, even when the packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval changes step by step, even if the delay time difference between the short system and the long system is large. The occurrence of fluctuation can be suppressed.

  According to a sixth aspect of the present invention, there is provided a packet non-disruptive switching device comprising: a transmission function unit that duplicates a packet to which an order identifier is added and transmits the packet by a plurality of routes; and a packet that has the same identifier received from a plurality of routes. A non-disruptive switching device having a reception function unit that selects an early arrival packet from the inside and transfers the packet downstream; the reception function unit transmits a packet to be transferred downstream from a reception packet of the first route to a second route; Switching instruction means for giving an instruction to switch to the received packet, and a delay until the same packet received from the first path after the time point instructed from the switching instruction means is received from the second path When the time is longer than a preset threshold, the packet delay method for gradually delaying the packet of the switching source first path until the delay time becomes equal to or less than the threshold. Characterized in that it is composed of a.

  As a result, when switching from short-path packet traffic with a short transmission path to long-path packet traffic with a long transmission path and a short delay time, the packet interval at the time of switching changes rapidly. The occurrence of fluctuation due to the difference in delay time between the short system and the long system can be suppressed.

  The packet non-disruptive switching device according to claim 7 of the present invention is the packet non-disruptive switching device according to claim 6, wherein the packet delay means is configured to send the first route after the time point instructed by the switching instruction means. Delay determining means for determining a delay time until the same packet as the packet received from the second path is received from the second path; temporarily holding the received packet in the order of reception; When the delay adjustment memory to be read and the delay time determined by the delay determination means are larger than 0, the received packet of the first route of the switching source is read from the delay adjustment memory until the delay time becomes 0 or less. It is characterized by comprising delay control means for increasing the delay amount in proportion to time.

  As a result, when switching packet traffic from the short system to the long system, the packet interval gradually changes in proportion to the time, so even if the delay time difference between the short system and the long system is large, the occurrence of fluctuations can be suppressed. it can.

  The packet non-disruptive switching device according to an eighth aspect of the present invention is the packet non-disruptive switching device according to the sixth aspect, wherein the packet delay means is configured to send the first route after the time point instructed by the switching instruction means. Delay determining means for determining a delay time until the same packet as the packet received from the second path is received from the second path; temporarily holding the received packet in the order of reception; When the delay adjustment memory to be read and the delay time determined by the delay determination means are larger than 0, the received packet of the first route of the switching source is read from the delay adjustment memory until the delay time becomes 0 or less. It is characterized by comprising delay control means for increasing the delay amount step by step by a predetermined time.

  As a result, when switching the packet traffic from the short system to the long system, the packet interval changes step by step for a predetermined time, so that the occurrence of fluctuation can be suppressed even when the delay time difference between the short system and the long system is large. .

  A packet non-disruptive switching device according to a ninth aspect of the present invention is the packet non-disruptive switching device according to the seventh or eighth aspect, wherein the packet is switched from the received packet of the second route of the switching destination switched by the switching instruction means. Further provided is a switching cancellation instructing unit that gives an instruction to return to the received packet of the original first route, and the delay control unit performs the switching source first after the time when the cancellation instruction is given from the switching cancellation instructing unit. It is characterized in that the delay amount when reading the received packet of the route from the delay adjustment memory is decreased in inverse proportion to the time.

  As a result, even if the packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval gradually changes in inverse proportion to the time, so the delay time difference between the short system and the long system is large. However, the occurrence of fluctuation can be suppressed.

  The packet non-disruptive switching device according to claim 10 of the present invention is the packet non-disruptive switching device according to claim 7 or 8, wherein the packet is switched from the received packet of the second route of the switching destination switched by the switching instruction means. Further provided is a switching cancellation instructing unit that gives an instruction to return to the received packet of the original first route, and the delay control unit performs the switching source first after the time when the cancellation instruction is given from the switching cancellation instructing unit. It is characterized in that a delay amount given every time a received packet of a route is read from the delay adjustment memory is decreased stepwise by a predetermined time.

  As a result, even when the packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval changes step by step, even if the delay time difference between the short system and the long system is large. The occurrence of fluctuation can be suppressed.

  According to an 11th aspect of the present invention, there is provided a packet non-disruptive switching method comprising: a transmitting side that duplicates a packet to which an order identifier is added and transmits the same by a plurality of paths; and a packet that has the same identifier received from a plurality of paths In a packet non-disruptive switching method used in a packet non-interruptible transmission system having a receiving side that selects a first arrival packet from the receiving side and transfers it downstream, the packet that is transferred downstream on the receiving side from the received packet of the first route A switching instruction procedure for giving an instruction to switch to a received packet of the second route, and a packet identical to a packet received from the first route after the time point instructed from the switching instruction procedure is received from the second route. If the delay time until the delay time is longer than a preset threshold value, packets of the first route of the switching source are gradually changed until the delay time becomes equal to or less than the threshold value. Characterized by providing a packet delay procedure delays.

  As a result, when switching from short-path packet traffic with a short transmission path to long-path packet traffic with a long transmission path and a short delay time, the packet interval at the time of switching changes rapidly. The occurrence of fluctuation due to the difference in delay time between the short system and the long system can be suppressed.

  The packet non-disruptive switching device according to claim 12 according to the present invention is the packet non-disruptive switching method according to claim 11, wherein the packet delay procedure is performed after the time point instructed from the switching instruction procedure. A delay determination procedure for determining a delay time until the same packet as the packet received from the second path is received from the second path, and temporarily holding the received packet in the order of reception and receiving the received packet in the order of reception according to the read command When the delay adjustment memory to be read and when the delay time determined by the delay determination procedure is greater than 0, the received packet of the switching source first path is read from the delay adjustment memory until the delay time becomes 0 or less. And a delay control procedure for increasing the delay amount in proportion to time.

  As a result, when switching packet traffic from the short system to the long system, the packet interval gradually changes in proportion to the time, so even if the delay time difference between the short system and the long system is large, the occurrence of fluctuations can be suppressed. it can.

  The packet non-disruptive switching method according to claim 13 of the present invention is the packet non-disruptive switching method according to claim 11, wherein the packet delay procedure is the first route after the time point instructed from the switching instruction procedure. A delay determination procedure for determining a delay time until the same packet as the packet received from the second path is received from the second path, and temporarily holding the received packet in the order of reception and receiving the received packet in the order of reception according to the read command When the delay adjustment memory to be read and when the delay time determined by the delay determination procedure is greater than 0, the received packet of the switching source first path is read from the delay adjustment memory until the delay time becomes 0 or less. And a delay control procedure for increasing the delay amount step by step by a predetermined time.

  As a result, when switching the packet traffic from the short system to the long system, the packet interval changes step by step for a predetermined time, so that the occurrence of fluctuation can be suppressed even when the delay time difference between the short system and the long system is large. .

  A packet non-disruptive switching device according to a fourteenth aspect of the present invention is the packet non-disruptive switching method according to the twelfth or thirteenth aspect, wherein the switching is performed from the received packet of the second route of the switching destination switched by the switching instruction procedure. There is further provided a switching cancellation instruction procedure for giving an instruction to return to the received packet of the original first route, and the delay control procedure is configured such that the switching source first instruction after the time when the cancellation instruction is given from the switching cancellation instruction procedure. It is characterized in that the delay amount when reading the received packet of the route from the delay adjustment memory is decreased in inverse proportion to the time.

  As a result, even if the packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval gradually changes in inverse proportion to the time, so the delay time difference between the short system and the long system is large. However, the occurrence of fluctuation can be suppressed.

  The packet non-disruptive switching method according to claim 15 of the present invention is the packet non-disruptive switching method according to claim 12 or 13, wherein switching is performed from the received packet of the second route of the switching destination switched by the switching instruction procedure. There is further provided a switching cancellation instruction procedure for giving an instruction to return to the received packet of the original first route, and the delay control procedure is configured such that the switching source first instruction after the time when the cancellation instruction is given from the switching cancellation instruction procedure. It is characterized in that a delay amount given every time a received packet of a route is read from the delay adjustment memory is decreased stepwise by a predetermined time.

  As a result, even when the packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval changes step by step, even if the delay time difference between the short system and the long system is large. The occurrence of fluctuation can be suppressed.

  According to the present invention, when the packet transmission path is switched, the time difference between the switching source packet and the switching destination packet is reduced by inserting a delay in the switching source packet traffic, so that fluctuations at the time of switching the delay path occur. Can be minimized.

1 is a configuration diagram of a packet uninterrupted transmission system 1. FIG. It is a flowchart which shows the process of packet uninterrupted transmission. It is explanatory drawing explaining the delay fluctuation at the time of packet non-interruption switching. 1 is a block diagram of a packet non-disruptive switching device 100 according to a first embodiment. It is a block diagram of the packet non-disruptive switching device 200 corresponding to SONET / SDH. It is a block diagram of the packet non-disruptive switching device 250 corresponding to SONET / SDH. It is a block diagram of the packet non-interruptible switching device 260 corresponding to OTN. It is a block diagram of the packet non-disruptive switching device 270 corresponding to OTN. It is a flowchart which shows the delay process in plan switching mode / plan switching cancellation | release mode. It is explanatory drawing which shows a mode in the case of increasing delay amount in proportion to time. It is explanatory drawing which shows a mode in the case of increasing delay amount in steps. It is explanatory drawing which shows a mode in the case of reducing delay amount in proportion to time. It is explanatory drawing which shows a mode in the case of reducing delay amount in steps. It is a flowchart which shows the delay process in plan switching mode. It is a block diagram of the packet non-disruptive switching device 300 according to the second embodiment. It is a flowchart which shows the process of the packet non-disruption switching apparatus 300 which concerns on 2nd Embodiment.

Embodiments of a packet non-disruptive transmission system, a packet non-disruptive switching device, and a packet non-disruptive switching method according to the present invention will be described below with reference to the drawings.
(First embodiment)
[Outline of packet uninterrupted transmission system]
FIG. 1 is an explanatory diagram showing a system configuration example of a packet uninterrupted transmission system 1 according to the first embodiment. In the example of FIG. 1, the system configuration is such that packets transmitted and received between user networks 2 and 3 (users NW2 and 3) depicted at both ends are connected via a wide area network 4 (wide area NW4) depicted in the center. Thus, transmission devices (packet-less interrupt switching devices 5 and 6 according to the present embodiment) for relaying between the two networks are installed at the boundary between the user networks 2 and 3 and the wide area network 4, respectively.

  In FIG. 1, for example, the packet non-disruptive switching device 5 duplicates a packet transmitted from the user NW2, divides it into a plurality of routes A, B, C, and D and transfers them to the wide area NW4 side. One of a plurality of packets received from a plurality of paths is selected and transferred to the user NW2 side.

  Here, the basic operation of the non-interruptible switching devices 5 and 6 will be described with reference to the flowchart of FIG. In the flowchart of FIG. 2, for example, in the case of the packet non-disruptive switching device 5, a transmission side process for transmitting a packet received from the user NW 2 to the wide area NW 4 and a reception side process for transmitting a packet received from the wide area NW 4 to the user NW 2. Composed.

  In the transmission side processing, for example, when a packet is received from the user NW2 in FIG. 1 (step S1), the received packet is duplicated by the number of routes (step S2), and an identifier indicating the order or the like is assigned to each duplicated packet ( In step S3), the plurality of packets are divided into different routes and transmitted to the wide area NW4 (step S4). Note that the identifier may be added after the packet is duplicated, or the identifier may be duplicated after the identifier is added to the packet.

  On the other hand, in the receiving process, for example, when a packet sent from a packet non-disruptive switching device 6 via a wide area NW 4 via a plurality of different routes is received (step S5), the received packet identifier is confirmed and a plurality of routes are confirmed. The normal packet (early arrival packet) received first is selected (step S6), the identifier is deleted from the selected packet and transferred to the user NW2 side (step S7).

  In this way, the packet non-disruptive switching devices 5 and 6 transmit a plurality of duplicated packets through a plurality of routes, so that even if a failure occurs in a part of the routes, packet loss does not occur and route switching is performed. However, a highly reliable network with little delay fluctuation can be realized. In the present embodiment, the names of the packet non-disruptive switching devices 5 and 6 are referred to as “packet non-disruptive switching devices” so that their functions are easily understood. It may be referred to as a relay device. Also, what is called “packet” in this embodiment is not particularly specified, and indicates a group of transmission information units in which header information such as a destination is added to user information, and is used in layer 3 of the OSI layer. Packet, a frame used in layer 2 or the like. Therefore, the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present embodiment are similarly applied to any system or device that transmits digital data by dividing it into predetermined transmission information units. Can do.

  However, when the packet copied as described above is divided into a plurality of transmission paths and transmitted, the lengths of the transmission paths of each system are different. Causes a delay corresponding to the transmission path length difference. Here, for example, in the case of the routes A, B, C and D in FIG. 1, a route having a long transmission path length is referred to as a long system, and conversely, a route having a short transmission path length is referred to as a short system. Since the arrival time of the flowing packet is earlier than the arrival time of the same duplicated packet that flows through the long path, a delay time occurs when switching from the short system to the long system. The state of packet traffic of this conventional method is shown in FIG. FIG. 3A shows a packet sequence received from a short route, a duplicate packet sequence received from a long route, and a packet selected from the short route or long route to a downstream network or device. It is explanatory drawing drawn so that the temporal positional relationship with the packet string of the selection system output might be understood. In the example of FIG. 3 (a), 12 packet sequences before and after switching are shown. The numbers 1 to 12 surrounded by squares each indicate one packet. It shows the same duplicated packet. Hereinafter, these 12 packets are expressed as packet (1), packet (2),..., Packet (12).

  In FIG. 3A, the arrival time of the short packet (1) and the arrival time of the long packet (1) have a transmission path delay of L seconds. For example, when the planned switching from the short system to the long system (switching operation by the operator) is executed after selecting the short system packet (5), the packet (5 in the selected system until the long system packet (6) arrives. ) Is transferred until the packet (6) is transferred, and fluctuations in the transmission path delay occur.

On the other hand, the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present embodiment, as will be described later, the fluctuation due to the transmission path delay as shown in FIG. Occurrence can be reduced.
[Configuration of packet non-interruptible switching devices 5 and 6]
Next, a configuration example of the packet non-disruptive switching devices 5 and 6 will be described in detail with reference to FIG. FIG. 4 is a block diagram of the packet non-disruptive switching device 100 showing one configuration example corresponding to the packet non-disruptive switching devices 5 and 6. In FIG. 4, the packet non-disruptive switching device 100 includes a transmission function unit 101 and a reception function unit 102.

  The transmission function unit 101 shown in FIG. 4 is a block for transmitting data received from the user NW to the wide area NW, and includes a reception unit 111, a copy unit 112, identifier assignment units 113a and 113b, and transmission units 114a and 114b. It consists of. In FIG. 4, for easy understanding, a configuration example in the case of having two routes is drawn. Therefore, the blocks after the copy unit 112 are each of two systems (a system corresponding to the route A of the wide area NW 4 and the wide area NW 4. (B system corresponding to the route B) having the same block, but when transmitting by dividing into more than two routes, the blocks after the copy unit 112 may be provided by the number of routes. Further, in the following description, except when referring to a specific block, for example, the identifier assigning unit 113a and the identifier assigning unit 113b are described by omitting the system name (a, b) like the identifier assigning unit 113. For example, the transmission unit 114 a and the transmission unit 114 b are referred to as the transmission unit 114. In addition, similar blocks appearing in the following drawings are also indicated by omitting the system names (a, b) unless otherwise specified.

  The receiving unit 111 receives a packet from the user NW. As described above, any type of packet may be used. For example, the packet may be an Ethernet (registered trademark) IP packet or a frame such as a MAC frame. Alternatively, it may be a packet corresponding to MPLS (Multi-Protocol Label Switching).

  The copy unit 112 duplicates a plurality of packets having the same information as the packet from the user NW received by the reception unit.

  The identifier assigning unit 113 writes an order identifier (and a route identifier) in each of the duplicated packets. Here, the order identifier is an identifier indicating the transmission order. Each time an integer value of 0, 1, 2, 3... Is transmitted, it is incremented by 1 and the numerical value is written (note that 1, 2, 2, etc.). It may be a natural number of 3 ...). In addition, the path identifier is an identifier indicating the transmission path to be transmitted, and values 1 to n corresponding to the 1st system, the 2nd system, etc. of the transmission path are written, but the path identifier may not be provided.

  The transmission unit 114 transmits a plurality of packets, to which the order identifiers and the like are assigned by the identifier assignment unit 113, to different routes on the wide area NW side. For example, one of the two duplicate packets including the same information to which the same order identifier is assigned is transmitted from the transmission unit 114a to the path A, and the other packet is transmitted from the transmission unit 114b to the path B.

  Next, the reception function unit 102 will be described. In FIG. 4, a reception function unit 102 is a block for transmitting a signal received from a wide area NW to a user NW, and includes a reception unit 121 (reception units 121a and 121b) and a delay determination unit 122 (delay determination units 122a and 122b). ), Delay adjustment memory 123 (delay adjustment memories 123a and 123b), memory 124 (memory 124a and 124b), selection unit 125 (selection units 125a and 125b), counter unit 126, and transmission unit 127 And a delay control unit 128. Here, the block constituted by the delay adjustment block 151 and the delay control unit 128 surrounded by a dotted line frame is a characteristic part of the packet uninterruptible transmission system, the packet uninterruptible switching device, and the packet uninterruptible switching method according to the present invention. is there. In particular, in the case of FIG. 4, the delay adjustment block 151 includes a delay determination block 152 of the delay determination units 122a and 122b and a delay adjustment memory block 153 of the delay adjustment memories 123a and 123b.

  Receiving sections 121a and 121b receive packets sent from route A and route B of the corresponding wide area NW, respectively.

  The delay determination unit 122 determines the delay difference (arrival time difference) of packets having the same order identifier received from each path, and outputs delay difference information to the delay control unit 128. Thereafter, the received packet is held in the delay adjustment memory 123 according to the received path, and then a predetermined delay is inserted according to a command from the delay control unit 128 and stored in the memory 124. That is, it is read from the memory 123 with a delay of a predetermined time (including 0 (no delay)) and stored in the memory 124.

  The delay adjustment memory 123 is composed of a FIFO memory that reads first from a previously written packet. The packets output from the delay determination unit 122 are accumulated in order, and are read out in order from the previously written packet in response to a read command from the delay control unit 128.

  The memory 124 is also composed of a FIFO memory, and the packets read from the delay adjustment memory 123 are accumulated in order, and the selection unit 125 sequentially reads the packets written in advance.

  The selection unit 125 refers to the sequence identifier of the packet read from the memory 124, selects the first-arrival packet among the packets having the same information received from a plurality of paths, and outputs it to the transmission unit 127. A packet having the same information that has not been selected is discarded. That is, if there is an early arrival packet, the later arrival packet is discarded.

  The transmission unit 127 deletes the sequence identifier from the packet output from the selection unit 125, and then transfers the sequence identifier to a downstream network or device.

  When a route switching request (plan switching request) is given by an operator, the delay control unit 128 controls the delay determination unit 122 and the delay adjustment memory 123 to switch the route of packet traffic. For example, in the case of FIG. 4, a switching request from the route A to the route B is made, or conversely, a switching request from the route B to the route A is made. Alternatively, the reverse operation is performed when a plan switch release request for returning the packet traffic once switched is given.

In FIG. 4, the block diagram in the case where the packet non-disruptive switching device 100 is configured as a general packet transfer device has been described using a block diagram, but the present invention is not limited to this.
[Application example to SONET / SDH or OTN]
For example, a transmission device corresponding to Packet over SONET / SDH (SONET (Synchronous Optical Network) / SDH (Synchronous Digital Hierarchy)) or an OTN (Optical Network) The present invention can also be applied to a transmission device that performs such processing. For example, FIG. 5 is a block diagram illustrating a configuration of the packet non-disruptive switching device 200 when the packet non-disruptive switching device 100 described in FIG. 4 is made compatible with Packet over SONET / SDH.

  Here, the difference between FIG. 5 and FIG. 4 will be described. 5 corresponds to SDH, the transmission function unit 201 includes a GFP (Generic Framing Procedure) mapping unit 211, VC (Virtual Container) mapping units 212a and 212b, and a multiplexing unit that multiplexes VC mapped frames. 213 are arranged. In addition, the reception function unit 202 includes a separation unit 221 that separates VC frames received from the wide area NW by the reception unit 121, VC demapping units 222a and 222b, and a GFP demapping unit 223. The other blocks are configured in the same manner as in FIG. For ease of understanding, the same reference numerals as those in FIG. 4 have the same functions as those described with reference to FIG. 4, but the data to be handled is different, such as a GFP frame or a VC frame. For example, a packet received from the user NW by the reception unit 111 of the transmission function unit 201 is mapped to a GFP frame by the GFP mapping unit 211. Then, the copy unit 112 duplicates the GFP frame and outputs it to the identifier assigning unit 113. Further, the GFP frame to which the order identifier and the route identifier are assigned is mapped to the VC frame by the VC mapping unit 212, multiplexed by the multiplexing unit 213, and then transmitted from the transmission unit 114 to the wide area NW through each route. The

  On the other hand, the reception unit 121 of the reception function unit 202 receives the multiplexed frames from each path of the wide area NW, and separates the multiplexed frames by the separation unit 221. Then, the VC demapping unit 222 demaps the GFP frame, and the delay determination unit 122 of the delay adjustment block 151 determines the delay time of the same frame and outputs the demapped GFP frame to the delay control unit 128. At the same time, the frame is temporarily stored in the delay adjustment memory 123. As described with reference to FIG. 4, the delay control unit 128 is temporarily stored in the delay adjustment memory 123 using the delay time determination result input from the delay determination unit 122 in response to the plan switching request / cancellation request. The GFP frame being read is appropriately read and stored in the memory 124. Then, the selection unit 125 selects an early arrival frame with reference to the count value of the counter unit 126 and the sequence identifier of the GFP frame stored in the memory 124, and outputs it to the GFP demapping unit 223. The GFP demapping unit 223 outputs a packet obtained by demapping the GFP frame output from the selection unit 125 to the transmission unit 127, and is transferred from the transmission unit 127 to the user NW.

  As described above, even in the case of the packet non-disruptive switching device 200 corresponding to Packet over SONET / SDH, the same processing as that of the packet non-disruptive switching device 100 described in detail in FIG. 4 can be performed.

  Next, FIG. 6 shows a configuration example of another packet non-interruptible switching device 250 corresponding to Packet over SONET / SDH. In FIG. 6, the same reference numerals as those in FIGS. 4 and 5 indicate blocks having the same function. The difference between FIG. 5 and FIG. 6 is that the position of the VC mapping unit 212 is different in the transmission function unit 251. In FIG. 5, the VC mapping units 212 a and 212 b are arranged after the identifier is assigned by the identifier assigning unit 113, but in FIG. 6, the VC mapping unit 212 is arranged immediately after the GFP mapping unit 211. As a result, the number of blocks of the VC mapping unit 212 can be reduced compared to the case of FIG. 5, so that the circuit scale can be reduced and the cost can be reduced.

  Similarly, in the reception function unit 252, the position of the VC demapping unit 222 is different. In FIG. 4, the VC demapping units 222 a and 222 b are arranged immediately after the separating unit 221, but in FIG. 6, the VC demapping unit 222 is arranged immediately before the GFP demapping unit 223. As a result, the number of blocks of the VC demapping unit 222 can be reduced as compared with the case of FIG. 5, so that the circuit scale can be reduced and the cost can be reduced.

  Next, a configuration example of the packet non-disruptive switching device 260 corresponding to OTN is shown in FIG. In FIG. 7, the same reference numerals as those in FIGS. 4 and 5 indicate blocks having the same function. The difference between FIG. 5 and FIG. 7 is that an OTN mapping unit 214 is arranged instead of the VC mapping unit 212 and the multiplexing unit 213 in the transmission function unit 261 in order to support OTN. In FIG. 7, the GFP frame to which the order identifier and the route identifier are assigned by the identifier assigning unit 113 is mapped to the OTN frame by the OTN mapping unit 214, and is transmitted from the transmission unit 114 to the wide area NW through each route.

  On the other hand, in the reception function unit 262, the OTN frame received from each path of the wide area NW by the reception unit 121 is demapped to the GFP frame by the OTN demapping unit 224, and the demapped GFP frame is the delay adjustment block 151. The delay determination unit 122 determines the delay time of the same frame and outputs it to the delay control unit 128, and temporarily stores the frame in the delay adjustment memory 123. As described with reference to FIG. 4, the delay control unit 128 is temporarily stored in the delay adjustment memory 123 using the delay time determination result input from the delay determination unit 122 in response to the plan switching request / cancellation request. The GFP frame being read is appropriately read and stored in the memory 124. Then, the selection unit 125 selects an early arrival frame with reference to the count value of the counter unit 126 and the sequence identifier of the GFP frame stored in the memory 124, and outputs it to the GFP demapping unit 223. The GFP demapping unit 223 outputs a packet obtained by demapping the GFP frame output from the selection unit 125 to the transmission unit 127, and is transferred from the transmission unit 127 to the user NW.

  As described above, even in the case of the packet non-disruptive switching device 260 corresponding to OTN, the same processing as that of the packet non-disruptive switching device 100 described in detail in FIG. 4 can be performed.

  Next, FIG. 8 shows a configuration example of another packet non-disruptive switching device 270 corresponding to OTN. In FIG. 8, the same reference numerals as those in FIGS. 4 and 7 indicate blocks having the same function. The difference between FIG. 7 and FIG. 8 is that in the transmission function unit 271, the position of the OTN mapping unit 214 is different. In FIG. 7, the OTN mapping unit 214 is arranged after the identifier is assigned by the identifier assigning unit 113, but in FIG. 8, the OTN mapping unit 214 is arranged at the same position instead of the GFP mapping unit 211 of FIG. ing. As a result, the number of blocks of the OTN mapping unit 214 can be reduced as compared with the case of FIG. 7, so that the circuit scale can be reduced and the cost can be reduced.

  Similarly, in the reception function unit 272, the position of the OTN demapping unit 224 is different. In FIG. 7, the OTN demapping units 224 a and 224 b are arranged immediately after the receiving unit 121, but in FIG. 8, the OTN demapping unit 224 is arranged at the same position instead of the GFP demapping unit 223. As a result, the number of blocks of the OTN demapping unit 224 can be reduced as compared with the case of FIG. 7, so that the circuit scale can be reduced and the cost can be reduced.

As described above with reference to FIGS. 5, 6, 7, and 8, the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present invention are applied to SONET / SDH or OTN. It can also be applied to.
[Operations of Delay Control Unit 128 and Delay Adjustment Block 151]
Next, operations of the delay determination block 152, the delay adjustment memory 123, and the delay control unit 128 constituting the delay adjustment block 151 will be described in detail with reference to FIG.

  The delay adjustment memories 123a and 123b are linked to the delay control unit 128, and the delay adjustment memories 123a and 123b are set in response to a plan switching request / cancellation request output from the OpS (operation system) or the counterpart device. Used to adjust the delay amount of the temporarily held packet.

  The delay control unit 128 receives the plan switching request / cancellation request from the OpS, refers to the delay difference information of the packets having the same order identifier received from the respective paths output from the delay determination unit 122, and sends them to the memory 124. Controls the amount of delay for outputting packets.

  FIG. 9 is a flowchart showing a processing example using the delay adjustment memory 123a of the a system in FIG. The same flow chart applies to the b-system delay adjustment memory 123b.

  (Step S11) The received packet is written into the delay adjustment memory 123a.

  (Step S12) It is determined whether or not it is a plan switching mode from the transmission path a to the transmission path b. And when it is plan switching mode, it progresses to S13, and when that is not right, it progresses to S16.

  (Step S13) The delay difference τ (ab) between the packet on the delay adjustment memory 123a side and the packet on the delay adjustment memory 123b side is determined. If the delay difference τ (a−b)> 0 is satisfied, the process proceeds to S14. Otherwise, the process proceeds to S15. If the delay difference τ (a−b) <0, the switching request destination packet originally arrives early, so the packet is transferred to the memory 124a as it is.

(Step S14) Since the packet to the delay adjustment memory 123a is received earlier than the packet having the same information received by the delay adjustment memory 123b, the packet received by the delay adjustment memory 123a A delay amount τ1 is inserted. Note that the delay amount τ1 is controlled to increase linearly in proportion to time, or to increase stepwise by Δτ1 every time a packet is received. The method of inserting the delay amount τ1 will be described in detail later with reference to FIGS.
An example of inserting the delay amount τ1 will be described in detail later with reference to FIGS.

  (Step S15) The packet in which the delay of τ1 is inserted (the packet read with a delay of τ1) is transferred to the memory 124a.

  (Step S16) It is determined whether or not it is a plan switching cancellation mode from the transmission path a to the transmission path b. If it is the plan switching cancellation mode, the process proceeds to S17, and otherwise, the process proceeds to S15. Here, if it is neither the plan switching mode nor the plan switching cancellation mode, the process proceeds to S15, and the packet is transferred to the memory 124a as it is.

(Step S17) Since the packet to the delay adjustment memory 123a is received earlier than the packet having the same information received by the delay adjustment memory 123b, the packet received by the delay adjustment memory 123a A delay amount τ2 is inserted. The delay amount τ2 is controlled to decrease linearly in proportion to time, or to decrease stepwise by Δτ2 every time a packet is received. The method of inserting the delay amount τ2 will be described in detail later with reference to FIGS.
[Method of inserting delay amount τ1]
Next, a method for inserting the delay amount τ1 will be described. First, an example of a method for inserting the delay amount τ1 will be described with reference to the graph of FIG. In FIG. 10, the horizontal axis represents the time axis (t), and the vertical axis represents the delay time difference (delay difference: τ). The method shown in FIG. 10 is a method of increasing the predetermined delay amount τ1 in proportion to the passage of time t. In FIG. 10, the switching request occurrence time from OpS is set to 0, and a delay amount τ1 larger than the delay difference τ (ab) between the a system and the b system is inserted after tz1 seconds in proportion to time. Gradually increase the amount of delay. The delay difference τ (ab) between the a system and the b system is measured by the delay determination unit 122.

  Here, when the delay amount τ1 is close to the delay amount τ (ab), the packet arrival order is changed due to a slight processing delay, delay fluctuation, or the like, so that the switching process may not be performed normally. For this reason, after continuing the process of gradually increasing the delay until the target delay amount τ1 becomes larger than τ (ab) with a margin (tz2), the delay amount of the switching source packet The delay control unit 128 performs processing so as to keep τ1 constant.

  Next, another example of the method of inserting the delay amount τ1 will be described using the graph of FIG. 11, the horizontal axis indicates the time axis (t) and the vertical axis indicates the difference in delay time (delay difference: τ), as in FIG. The method shown in FIG. 10 is a method of increasing linearly in proportion to the time t until the delay amount τ1 is reached, but FIG. 11 shows the switching source until the delay amount τ (ab) or more is reached. In this method, the delay amount Δτ1 is increased step by step every time a packet arrives. Similarly to the case of FIG. 10, the delay amount Δτ1 is increased stepwise until τ1> τ (ab), and the delay amount τ1 having a sufficient margin from the delay amount τ (ab). For the switching source packet after becoming, the delay amount τ1 is maintained. If the arrival frequency of the packet is low, it may be controlled to time out when it does not reach τ1, and the delay amount may be forcibly set to τ1 when timed out.

  Here, a specific state of packet traffic when the plan switching request is given and the delay amount is increased stepwise will be described with reference to FIG. FIG. 3B shows a packet sequence received from the short path and a copy received from the long path, as in FIG. 3A showing the state of the conventional packet traffic described above. It is explanatory drawing drawn so that the temporal positional relationship might be understood with the same packet sequence and the packet sequence of the selection system from which the packet selected from the short system or the long system was output to a downstream network or apparatus. Similarly to FIG. 3A, 12 packet sequences before and after switching are shown by numbers 1 to 12 surrounded by a square. In the following description, these 12 packet sequences are shown. The packet is expressed as packet (1), packet (2),..., Packet (12).

  In FIG. 3B, the arrival time of the short packet (1) and the arrival time of the long packet (1) have a transmission path delay of L seconds. For example, when a request for switching a plan from the short system to the long system is given after selecting the short system packet (5), the packet is output in the selected system until the long system packet (6) arrives in FIG. However, in this embodiment, the control is performed so as to gradually delay each time a short packet arrives until it becomes later than the arrival timing of the long packet. For example, the first short packet (6) to which the plan switching request is given is delayed by a predetermined delay amount Δτ and output to the selection system. Here, the delay amount Δτ is a delay amount that does not cause a problem of fluctuation. The next second short packet (7) is delayed by the delay amount (Δτ × 2) and output to the selection system. Even at this time point, the delayed packet (7) is earlier than the arrival timing of the switching-destination long packet (7), so the next third short packet (8) is further transferred to the delay amount (Δτ × 3) Delay the output and output to the selection system. Similarly, the short packet (9) is delayed by the delay amount (Δτ × 4) and output to the selection system. Then, if the fifth short packet (10) is delayed by the delay amount (Δτ × 5), it will be later than the arrival timing of the long packet (10). Switch to (10) and output to the selection system. Here, Δτ corresponds to Δτ1 in FIG.

As described above, in the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present embodiment, the short-system packet after the time point when the request for switching the plan from the short system to the long system is given. Wait for a predetermined delay amount and output to the selection system, and when the timing of outputting the waiting short system packet to the selection system is later than the arrival timing of the same long system packet, it becomes a long system packet Since the switching is controlled so that the delay difference between the short packet and the long packet is gradually reduced step by step, the occurrence of delay fluctuation can be reduced.
[Method of inserting delay amount τ2]
Next, an example of a method for inserting the delay amount τ2 will be described using the graph of FIG. In FIG. 12, the horizontal axis represents the time axis (t), and the vertical axis represents the delay time difference (delay difference: τ). The method shown in FIG. 12 is a method of decreasing until the predetermined delay amount τ2 becomes 0 in proportion to the passage of time t. In FIG. 12, the delay amount τ2 is gradually increased in proportion to the time so that the delay difference τ (ab) between the a system and the b system becomes 0 after tz 3 seconds, when the switching release request occurrence time from OpS is 0. Reduce to.

  Next, another example of the method of inserting the delay amount τ2 will be described using the graph of FIG. 13, the horizontal axis indicates the time axis (t), and the vertical axis indicates the delay time difference (delay difference: τ). The method shown in FIG. 12 is a method of decreasing the delay amount τ2 linearly in proportion to the time t. However, FIG. 13 shows that every time the packet arrives until the delay amount τ (ab) becomes zero. In this method, the delay amount Δτ2 is decreased step by step. Note that when the arrival frequency of the packet is low and the delay amount has not reached 0, control may be made to time out, and the delay amount may be forcibly set to 0 when the timeout occurs.

  Here, FIG. 13 corresponds to performing the reverse operation of the method of increasing the delay amount Δτ stepwise described with reference to FIG. For example, in FIG. 3B, Δτ corresponds to Δτ2 in FIG. 13, and the delay amount τ2 corresponds to, for example, the delay amount (Δτ × 5). Then, when a plan switch release request for returning from the long system to the short system is given, every time a short system packet arrives, the delay amount τ2 is gradually decreased by Δτ2, and finally the delay amount The plan switching release mode ends when τ2 becomes zero.

  As described above, in the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present embodiment, the short packet after the point when the request for canceling the plan switching from the short system to the long system is given. Since the delay amount is gradually reduced and output to the selection system, the occurrence of delay fluctuation in the plan switching cancellation mode can be reduced.

  As described above, the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present embodiment can be used for the short system and the long system when receiving the plan switching request / cancellation request from the OpS or the opposite device. By gradually reducing the delay difference between the packets, it is possible to minimize delay fluctuations with respect to the transfer destination network or device.

[Modification 1 of the first embodiment]
Next, Modification 1 of the embodiment described above will be described with reference to FIG. In FIG. 11 and FIG. 13, the delay amount adjustment at the time of switching cancellation is controlled to decrease in proportion to the time t or stepwise decrease. However, as shown in FIG. 14, the delay amount is set to zero at a stroke. It doesn't matter. In FIG. 14, steps with the same reference numerals as in FIG. 9 perform the same processing. The difference from FIG. 9 is that in step S12, if the plan switching mode is not set, all the processing proceeds to S15. Therefore, in the plan switching cancel mode, the delay amount is controlled to 0 at once.

[Modification 2 of the first embodiment]
Next, a second modification of the embodiment described above will be described. In the embodiment described above, switching between the a-system and the b-system is performed only by adjusting the delay amount. However, after switching between the a-system and the b-system by inserting a delay, the selection unit 125 performs selection. You may implement | achieve by the method of fixing the system to perform. For example, when switching from the a-system to the b-system, the selection unit 125 performs processing to select the b-system packet traffic after the switching. This eliminates the need for the selection unit 125 to always perform the process of discriminating early arrival packets, thereby reducing the processing load.

  As described above, the packet non-interruptible transmission system, the packet non-interruptible switching device, and the packet non-interruptible switching method according to the present embodiment are configured so that the transmission path has a long delay time from the packet traffic of a short route with a short transmission path and a short delay time. When switching to long-path packet traffic with a long path, the packet interval at the time of switching does not change abruptly, and fluctuations due to the difference in delay time between the short and long systems can be suppressed.

  In particular, when switching the packet traffic from the short system to the long system, the packet interval gradually increases by a predetermined time, so that the occurrence of fluctuation can be suppressed even when the delay time difference between the short system and the long system is large.

Even when packet traffic switched from the short system to the long system is returned from the long system to the short system again, the packet interval gradually decreases by a predetermined time, so even if the delay time difference between the short system and the long system is large, fluctuations Occurrence can be suppressed.
(Second Embodiment)
Next, a second embodiment of the packet non-disruptive transmission system, the packet non-disruptive switching device, and the packet non-disruptive switching method according to the present invention will be described. The packet uninterrupted transmission system according to the second embodiment is the same as the packet uninterrupted transmission system 1 described with reference to FIG. 1 of the first embodiment. Similarly, transmission apparatuses (packet-less interrupt switching apparatuses 5 and 6) that perform relay between two networks arranged at the boundary between the user networks 2 and 3 and the wide area network 4 in FIG. This corresponds to the non-disruptive switching device 300. FIG. 15 shows a block diagram of the packet non-disruptive switching device 300.

  Note that the packet non-disruptive switching device 300 according to the present embodiment gradually reduces the delay difference between the short and long packets by a method different from that of the first embodiment, so that the delay between the short and long packets is reduced. Plan switching is completed to complete switching at the time of reverse rotation. In the first embodiment, as described in FIG. 4, the delay amount of the received packet is adjusted using the delay adjustment memory 123. In the second embodiment, as shown in FIG. The delay amount of the received packet is controlled by a method of processing by the selection unit 125 instead of using the delay adjustment memory 123. As a result, similar to the first embodiment, plan switching without delay fluctuation can be realized.

  In FIG. 15, the packet non-disruptive switching device 300 includes a transmission function unit 301 and a reception function unit 302. The transmission function unit 301 transmits a packet received from the user NW to the wide area NW, and the reception function unit 302 transmits a signal received from the wide area NW to the user NW.

  The transmission function unit 301 illustrated in FIG. 15 includes a reception unit 111, a copy unit 112, identifier assignment units 113a and 113b, and transmission units 114a and 114b. Note that the configuration of the transmission function unit 301 is the same as that of the transmission function unit 101 described with reference to FIG. 4 and performs the same operation as the block having the same reference numeral as in FIG.

  Next, the reception function unit 302 will be described. In FIG. 15, a reception function unit 302 is a block for transmitting a signal received from a wide area NW to a user NW, and includes a reception unit 121 (reception units 121a and 121b) and a delay determination unit 122 (delay determination units 122a and 122b). ), A memory 124 (memory 124a and 124b), a selection unit 325 (selection units 325a and 325b), a counter unit 126, a transmission unit 127, and a delay control unit 328. Here, in addition to the delay determination block 152 and the delay control unit 328 enclosed by the dotted line frame, the part constituted by the delay adjustment selection block 354 is a packet non-disruptive transmission system, a packet non-disruptive switching device, and a packet non-disruptive switch according to the present invention. This is a characteristic part of the switching method. In particular, in the case of FIG. 15, the delay adjustment selection block 354 includes a selection unit 325 and a counter unit 126. The difference from FIG. 4 is that there is no delay adjustment memory 123 (delay adjustment memories 123a and 123b), and the delay control unit 328 controls the selection unit 325 of the delay adjustment selection block 354.

  In the reception function unit 302, the reception units 121a and 121b receive packets transmitted from the route A and the route B of the corresponding wide area NW, respectively. Then, the delay determination unit 122 determines the delay difference (arrival time difference) of the packets having the same order identifier received from each path, and outputs the delay difference information to the delay control unit 128. Thereafter, the received packet is stored in the memory 124 according to the received path.

  The selection unit 325 refers to the sequence identifier of the packet read from the memory 124, selects the first-arrival packet among the packets having the same information received from a plurality of paths, and outputs it to the transmission unit 127. A packet having the same information that has not been selected is discarded. That is, if there is an early arrival packet, the later arrival packet is discarded. In particular, in the present embodiment, the selection unit 325 is linked to the delay control unit 328, and the delay amount given to the packet in accordance with the plan switching request / cancellation request given from the OpS (until the packet is selected in this embodiment). (Equivalent to standby time). The packet selected by adjusting the delay amount in this way is transferred to the downstream network or device after the order identifier is deleted by the transmission unit 127. As described above, when a plan switching request / cancellation request is given, the delay control unit 328 controls the delay determination unit 122 and the selection unit 325 to switch the path of packet traffic. Alternatively, the reverse operation is performed when a plan switch release request for returning the packet traffic once switched is given.

  Here, FIG. 15 is described using a block diagram in the case where the packet non-disruptive switching device 300 is configured as a general packet transfer device, but as described in FIGS. 5 to 8 of the first embodiment, the SONET / SDH, OTN, etc., and the present invention is not limited to these.

  Next, a delay amount control method when a plan switching request / release request is given to the delay control unit 328 of the packet non-disruptive switching apparatus 300 according to the second embodiment will be described with reference to the flowchart of FIG. In the present embodiment, two independent paths are shown, which are referred to as a system and b system, respectively. Therefore, when the number of paths is n (n is a natural number), for example, the same processing is performed for n routes from 1 to n. Hereinafter, a description will be given according to the flowchart of FIG.

  (Step S31) The count value (CS) of the counter unit 126 is reset to zero.

  (Step S32) The presence or absence of a received packet in the a-system memory 124a is confirmed. If there is a received packet in the a-system memory 124a, the process proceeds to S33, and if there is no received packet, the process proceeds to S40.

  (Step S33) The received packet of the memory 124a is read.

  (Step S34) The sequence identifier (Ca) of the read received packet is compared with the count value (CS) of the counter unit. The processing is divided into three conditions of CS = Ca, CS> Ca, and CS> Ca. If CS = Ca, the process proceeds to S35. If CS> Ca, the process proceeds to S39. If CS> Ca, the process proceeds to S40. Proceed to

  (Step S35) It is determined whether or not it is a plan switching mode from the a system to the b system. If Yes, the process proceeds to S36, and if No, the process proceeds to S37.

  (Step S36) It is determined whether or not the waiting time (τ1) has expired. The delay control unit 328 has a timer function. If the preset standby time (τ1) has expired, the process proceeds to S37. If the standby time (τ1) has not expired, the process proceeds to S40. Note that the standby time (τ1) is the same as the method of FIGS. 10 and 11 described in the first embodiment.

  Here, the process of step S35 and step S36 is a part which performs the delay control process which becomes the characteristic of this embodiment.

  (Step S37) The received packet in the memory 124a is output to the transmitting unit 127, and the received packet is deleted from the memory 124a.

  (Step S38) The count value (CS) of the counter is incremented by 1 (CS = CS + 1), and the process proceeds to S40.

  (Step S39) Since the received packet in the memory 124a can be recognized as a packet that has already arrived in another system, the received packet is deleted from the memory 124a.

  (Step S40) The presence or absence of a received packet in the b-system memory 124b is confirmed. If there is a received packet in the b-system memory 124b, the process proceeds to S41. If there is no received packet, the process returns to S32 and waits for the reception of the next packet.

  (Step S41) The received packet of the memory 124b is read.

  (Step S42) The sequence identifier (Cb) of the read received packet is compared with the count value (CS) of the counter unit. The processing is divided into three conditions of CS = Cb, CS> Cb, and CS> Cb. If CS = Cb, the process proceeds to S43. If CS> Cb, the process proceeds to S47. If CS> Cb, the process proceeds to S48. Proceed to

  (Step S43) It is determined whether or not it is a plan switching mode from the b system to the a system. If Yes, the process proceeds to S44. If No, the process returns to S32 to wait for reception of the next packet.

  (Step S44) Similarly to step S36, it is determined whether or not the standby time (τ1) has expired. When the waiting time (τ1) has expired, the process proceeds to S45, and when the waiting time (τ1) has not expired, the process returns to S32 and waits for reception of the next packet.

  Here, similarly to the processing of step S35 and step S36, the processing of step S43 and step S44 is a part for performing the delay control processing which is a feature of this embodiment.

  (Step S45) The received packet in the memory 124b is output to the transmitting unit 127, and the received packet is deleted from the memory 124b.

  (Step S46) The count value (CS) of the counter is incremented by 1 (CS = CS + 1), and the process returns to S32 to wait for reception of the next packet.

  (Step S47) Since the received packet in the memory 124b can be recognized as a packet that has already arrived in another system, the received packet is deleted from the memory 124b, and the process returns to S32 to wait for the next packet to be received.

  (Step S48) It is determined whether CS <Ca. If CS <Ca, the process proceeds to S49. If CS> Ca, the process returns to S32 to wait for reception of the next packet.

  (Step S49) The count value (CS) of the counter is incremented by 1 (CS = CS + 1), and the process returns to S32 to wait for reception of the next packet.

  As described above, in the packet non-disruptive switching device 300 according to the present embodiment, the received packet taken into the memory 124a or the memory 124b is output to the transmission unit 127 with a delay of a predetermined waiting time in the plan switching mode. Therefore, the occurrence of fluctuation can be suppressed.

  In particular, as described with reference to FIGS. 10 and 11 of the first embodiment, the timing at which the selection unit 325 reads the received packet from the memory 124 so that the delay control unit 328 gradually reduces the delay difference between the two paths. By controlling, delay fluctuation can be minimized.

  In the flowchart of FIG. 16, the processing only when the plan switching is requested is shown, but the processing when the plan switching is canceled can be performed as in the first embodiment. In this case, the control may be performed so that the delay amount is gradually reduced to zero.

  Furthermore, in the second embodiment, since the packet delay adjustment is performed in the selection process in the selection unit 325, the delay adjustment memory 123 used in the first embodiment is not necessary. As a result, the circuit scale of the packet non-disruptive switching device 300 can be reduced, and the cost can be reduced.

  Note that FIG. 16 shows a flowchart for performing the b-system processing after the a-system processing, but it is also possible to perform the processing of all routes in parallel and determine the packets to be sent downstream.

  Further, the processing procedure of the transfer packet in the packet non-interruptible transmission system, the packet non-interruptible switching device, and the packet transmission method according to the present embodiment is not limited to the present embodiment.

[Modification 1 of Second Embodiment]
Next, Modification 1 of the embodiment described above will be described. In the embodiment described above, the switching between the a system and the b system is performed only by adjusting the delay amount. However, after switching between the a system and the b system by inserting a delay, the selection unit 325 performs the selection. You may implement | achieve by the method of fixing the system to perform. For example, when switching from the a-system to the b-system, the selection unit 325 performs processing so that the b-system packet traffic is fixedly selected after the switching. This eliminates the need for the selection unit 325 to always perform the process of determining an early arrival packet, thereby reducing the processing load.

  As described above, the packet non-interruptible transmission system, the packet non-interruptible switching device, and the packet non-interruptible switching method according to the present embodiment, when receiving a plan switching request from the OpS or the opposite device, By gradually reducing the delay difference, it is possible to minimize delay fluctuations with respect to the transfer destination network or device.

  The packet uninterrupted transmission system, the packet uninterruptible switching device, and the packet uninterruptible switching method according to the present invention have been described by way of example in the respective embodiments. However, without departing from the spirit or the main features thereof Can be implemented in various ways. Therefore, the above-described embodiment is merely an example in all respects and should not be interpreted in a limited manner. The present invention is shown by the scope of claims, and the present invention is not limited to the text of the specification. Further, all modifications and changes belonging to the equivalent scope of the claims are within the scope of the present invention.

1 ... Packet uninterrupted transmission system 2, 3 ... User NW
4 ... Wide area network
5, 6, 100, 200, 250, 260, 270, 300... Non-interruptible switching device 101, 201, 251, 261, 271, 301... Transmission function unit 102, 202, 252, 262, 272 302... Reception function unit 111... Reception unit 112... Copy unit 113 a, 113 b... Identifier assignment unit 114 a, 114 b... Transmission unit 121 a, 121 b. Delay determination units 123a, 123b ... delay adjustment memories 124a, 124b ... memories 125a, 125b, 325a, 325b ... selection unit 126 ... counter unit 127 ... transmission units 128, 328 ... Delay control unit 151 ... delay adjustment block 152 ... delay determination block 153 ... delay adjustment memory block 211 GFP mapping unit 212, 212a, 212b ... VC mapping unit 213 ... multiplexing unit 214, 214a, 214b ... OTN mapping unit 221 ... separation unit 222, 222a, 222b ... VC demapping unit 223 ... GFP demapping unit 224, 224a, 224b ... OTN demapping unit 354 ... delay adjustment selection block

Claims (15)

A transmission side that duplicates a packet with an order identifier added and transmits it via a plurality of routes; and a reception side that selects an early arrival packet from among packets with the same identifier received from a plurality of routes and forwards it downstream In a packet uninterrupted transmission system having
Switching instruction means for giving an instruction to switch a packet transferred downstream on the receiving side from a received packet of the first path to a received packet of the second path;
When the delay time until the same packet received from the first path after the time point instructed by the switching instruction means is received from the second path is longer than a preset threshold, the delay A packet non-interruptible transmission system, comprising: a packet delay unit that gradually delays the packet of the first path of the switching source until the time becomes equal to or less than the threshold value. The packet uninterrupted transmission system according to claim 1,
The packet delay means includes
A delay determination unit that determines a delay time until the same packet received from the first route after the time point instructed by the switching instruction unit is received from the second route;
A delay adjustment memory that temporarily holds the received packets in the order of reception and reads out the received packets in the order of reception according to the read command;
When the delay time determined by the delay determination means is greater than 0, the delay amount when reading the received packet of the first path of the switching source from the delay adjustment memory is proportional to time until the delay time becomes 0 or less. And a non-interruptible transmission system characterized by comprising delay control means for increasing the number of delays. The packet uninterrupted transmission system according to claim 1,
The packet delay means includes
A delay determination unit that determines a delay time until the same packet received from the first route after the time point instructed by the switching instruction unit is received from the second route;
A delay adjustment memory that temporarily holds the received packets in the order of reception and reads out the received packets in the order of reception according to the read command;
When the delay time determined by the delay determination means is greater than 0, the amount of delay when reading the received packet of the switching source first path from the delay adjustment memory until the delay time becomes 0 or less by a predetermined time. A packet non-interruptible transmission system comprising: delay control means for increasing in stages. The packet uninterrupted transmission system according to claim 2 or 3,
A switch release instruction means for giving an instruction to return from the received packet of the second route of the switching destination switched by the switching instruction means to the received packet of the first route of the switching source,
The delay control unit decreases the delay amount when reading the received packet of the first route of the switching source from the delay adjustment memory after the time when the cancellation instruction is given from the switching cancellation instruction unit in inverse proportion to time. A non-interruptible transmission system for packets. The packet uninterrupted transmission system according to claim 2 or 3,
A switch release instruction means for giving an instruction to return from the received packet of the second route of the switching destination switched by the switching instruction means to the received packet of the first route of the switching source,
The delay control unit decreases the delay amount when reading the received packet of the first route of the switching source from the delay adjustment memory after the time when the cancellation instruction is given from the switching cancellation instruction unit in inverse proportion to time. A non-interruptible transmission system for packets. A transmission function unit that duplicates a packet to which an order identifier is added and transmits it via a plurality of routes, and a reception function that selects an early arrival packet from among packets to which the same identifier is received from a plurality of routes and forwards it downstream A packet non-disruptive switching device having a
The reception function unit includes:
Switching instruction means for giving an instruction to switch a packet to be transferred downstream from a received packet on the first path to a received packet on the second path;
When the delay time until the same packet received from the first path after the time point instructed by the switching instruction means is received from the second path is longer than a preset threshold, the delay A packet non-interruptible switching device comprising: packet delay means for gradually delaying the packet of the switching source first path until the time becomes equal to or less than the threshold value. The packet non-disruptive switching device according to claim 6,
The packet delay means includes
A delay determination unit that determines a delay time until the same packet received from the first route after the time point instructed by the switching instruction unit is received from the second route;
A delay adjustment memory that temporarily holds the received packets in the order of reception and reads out the received packets in the order of reception according to the read command;
When the delay time determined by the delay determination means is greater than 0, the delay amount when reading the received packet of the first path of the switching source from the delay adjustment memory is proportional to time until the delay time becomes 0 or less. And a non-interruptible packet switching device characterized by comprising delay control means for increasing the number of packets. The packet non-disruptive switching device according to claim 6,
The packet delay means includes
A delay determination unit that determines a delay time until the same packet received from the first route after the time point instructed by the switching instruction unit is received from the second route;
A delay adjustment memory that temporarily holds the received packets in the order of reception and reads out the received packets in the order of reception according to the read command;
When the delay time determined by the delay determination means is greater than 0, the amount of delay when reading the received packet of the switching source first path from the delay adjustment memory until the delay time becomes 0 or less by a predetermined time. A packet non-interruptible switching device comprising: delay control means for increasing in stages. The packet non-disruptive switching device according to claim 7 or 8,
A switch release instruction means for giving an instruction to return from the received packet of the second route of the switching destination switched by the switching instruction means to the received packet of the first route of the switching source,
The delay control unit decreases the delay amount when reading the received packet of the first route of the switching source from the delay adjustment memory after the time when the cancellation instruction is given from the switching cancellation instruction unit in inverse proportion to time. A non-interruptible packet switching device characterized by The packet non-disruptive switching device according to claim 7 or 8,
A switch release instruction means for giving an instruction to return from the received packet of the second route of the switching destination switched by the switching instruction means to the received packet of the first route of the switching source,
The delay control means increments the delay amount given every time a received packet of the first route of the switching source is read from the delay adjustment memory after a time point when the cancellation instruction is given from the switching cancellation instruction means by a predetermined time. Packet non-disruptive switching device, characterized in that the number is reduced. A transmission side that duplicates a packet with an order identifier added and transmits it via a plurality of routes; and a reception side that selects an early arrival packet from among packets with the same identifier received from a plurality of routes and forwards it downstream In a packet uninterruptible switching method used in a packet uninterrupted transmission system having
A switching instruction procedure for giving an instruction to switch a packet to be transferred downstream on the receiving side from a received packet on the first path to a received packet on the second path;
When the delay time until the same packet received from the first path after the time point instructed from the switching instruction procedure is received from the second path is longer than a preset threshold, the delay A packet delay procedure for gradually delaying the packet of the first path of the switching source until the time becomes equal to or less than the threshold value. The packet non-disruptive switching method according to claim 11,
The packet delay procedure is:
A delay determination procedure for determining a delay time until a packet identical to a packet received from the first path after the time point instructed from the switching instruction procedure is received from the second path;
A delay adjustment memory that temporarily holds the received packets in the order of reception and reads out the received packets in the order of reception according to the read command;
When the delay time determined by the delay determination procedure is larger than 0, the delay amount when the received packet of the switching source first path is read from the delay adjustment memory until the delay time becomes 0 or less is proportional to the time. And a non-interruptible packet switching method, characterized by comprising: The packet non-disruptive switching method according to claim 11,
The packet delay procedure is:
A delay determination procedure for determining a delay time until a packet identical to a packet received from the first path after the time point instructed from the switching instruction procedure is received from the second path;
A delay adjustment memory that temporarily holds the received packets in the order of reception and reads out the received packets in the order of reception according to the read command;
When the delay time determined by the delay determination procedure is larger than 0, the delay amount when reading the received packet of the first route of the switching source from the delay adjustment memory until the delay time becomes 0 or less is set for each predetermined time. A packet non-interruptible switching method, characterized by comprising a delay control procedure that increases in stages. The packet non-disruptive switching method according to claim 12 or 13,
A switching cancellation instruction procedure for giving an instruction to return from the received packet of the second route of the switching destination switched by the switching instruction procedure to the received packet of the first route of the switching source;
In the delay control procedure, the delay amount when the received packet of the first route of the switching source is read from the delay adjustment memory after the time when the release instruction is given from the switch release instruction procedure is decreased in inverse proportion to the time. A packet non-disruptive switching method characterized by The packet non-disruptive switching method according to claim 12 or 13,
A switching cancellation instruction procedure for giving an instruction to return from the received packet of the second route of the switching destination switched by the switching instruction procedure to the received packet of the first route of the switching source;
In the delay control procedure, a delay amount given every time a received packet of the first route of the switching source is read from the delay adjustment memory after a time point when the release instruction is given from the switch release instruction procedure is incremented by a predetermined time. Packet non-disruptive switching method, characterized in that the number is reduced.

Images