JP2015046682A - Band control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a band control device which has made it possible to perform dynamic band control efficiently at high speed.SOLUTION: Provided is a band control device for controlling the band of a path which is set in a relay route of a network, the band control device having: a band movement determination unit for determining, by route information, band movement between edge nodes of the relay route, and outputting band movement information indicating the determination result; a band allocation calculation unit for calculating band allocation in accordance with the band movement information; and a storage unit for holding a band allocation history which is a history of the band allocations calculated by the band allocation calculation unit. The band movement determination unit correlates, on the basis of the route information received from the edge nodes, a pre-switching path, which is a path whose connection is to be switched, and a post-switching path with each other, and creates the band movement information including information about correlation between the pre-switching path and the post-switching path. The band allocation calculation unit estimates, from the band allocation history, the amount of band movement to the post-switching path and allocates an amount of band to the post-switching path by referring to the band allocation history of the pre-switching path.

Description

  The present invention relates to a bandwidth control technique for a relay route of a network.

  The wide-area metro network (NW) is provided between the core network and the access network, and corresponds to a network that concentrates traffic from home gateways at homes on the access network side to the edge router that is the first hop of L3. FIG. 25 is a block diagram showing an example of a conventional wide area metro NW configured with an optical L2 switch network (optical L2NW).

  FIG. 25 illustrates a case where an optical L2 management apparatus having a flow management unit that manages flow allocation at an edge node of the optical L2NW and a DBA (Dynamic Bandwidth Allocation) / scheduler is provided in the optical L2NW.

  In the wide area metro NW, there is an increasing demand for improved reliability by changing the edge connection flexibly and changing accommodation according to the load. For this reason, in the optical L2NW shown in FIG. 25, it has been studied to dynamically change the bandwidth of the relay path of the optical L2NW between the edge and the home gateway according to the amount of traffic to be used efficiently.

  In Non-Patent Document 1, a wide area metro NW is transferred via a path (MPLS (Multi-Protocol Label Switching) path, optical path, etc.), and flexible connection changes are made on the path by flow control (flow distribution / flow management). A technique for realizing Ethernet transfer capable of performing the above is disclosed. Non-Patent Document 2 discloses an optical L2NW that can efficiently use the bandwidth of an optical path by dynamic change by centralized control.

  FIG. 26 is a diagram in which the techniques disclosed in Non-Patent Documents 1 and 2 are applied to the optical L2NW shown in FIG. 25, and FIG. 27 is a block diagram showing a configuration example of an apparatus that performs band control.

  The address control unit 301 shown in FIG. 26 and FIG. 27 is related to the technology disclosed in Non-Patent Document 1, and the path bandwidth adjustment unit 303 is related to the technology disclosed in Non-Patent Document 2.

  When detecting an edge failure by a route change trigger (address update trigger) that is a trigger for notifying a route change, the address control unit 301 causes the path endpoint node 50 to switch the path connection. In addition, since the path bandwidth adjustment unit 303 periodically performs bandwidth allocation calculation, it collects traffic amount information, which is traffic amount information, from the path end point node 50 in an observation period of a fixed period, and based on the collected traffic amount The next bandwidth allocation is calculated.

  Regarding flow control, Non-Patent Documents 3 and 4 disclose techniques for dynamically controlling a path bandwidth by transferring a path without limiting the relay route to the optical L2NW. Non-Patent Documents 3 and 4 disclose a bandwidth control technique in MPLS.

Takuki Date, 3 others, “Proposal of distributed processing configuration of wide area virtual L2SW architecture”, Proceedings of 2013 General Conference, IEICE, March 5, 2013, B-6-87 Kimishima Naoki, 4 others, “Characteristic evaluation of dynamic bandwidth control method against traffic fluctuation in optical L2SW-NW”, IEICE Technical Report, IEICE, January 24, 2013, vol. 112, no. 398, p. 289-294 Rie Hayashi, 3 others, “Traffic flow measurement based bandwidth control method”, Proceedings of 2013 General Conference, IEICE, March 5, 2013, B-6-103 Aki Fukuda, 3 others, “Bandwidth Control Method for User Throughput Optimization”, Proceedings of 2013 General Conference, IEICE, March 5, 2013, B-6-104

  With respect to the path switching for the edge failure, there is a problem that the method of checking the traffic volume and changing the bandwidth delays the follow-up and delays the recovery until the necessary bandwidth is secured. There are two possible causes for this problem.

  The first is that the band tracking is delayed only during the period of observing traffic. Hereinafter, this factor is referred to as “factor 1”. FIG. 28 is a sequence diagram for explaining Factor 1 of the problem of the conventional technique.

  Conventionally, since the bandwidth allocation calculation is periodically performed, the following calculation is performed based on the traffic amount in the same observation period as the bandwidth allocation calculation cycle. Therefore, as shown in FIG. 28, the necessary bandwidth is not reflected in the calculation unless the increased traffic amount is collected for the collection period. It takes two calculation cycles to be reflected in the bandwidth allocation calculation after the switched traffic starts to flow.

  Second, the band changing operation is not started until traffic flows in. Hereinafter, this factor is referred to as “factor 2”. FIG. 29 is a sequence diagram for explaining the factor 2 of the problem of the prior art.

  Conventionally, as an operation on the edge side, a route change trigger is output to the core NW side and the metro NW side after the edge information transition, but bandwidth reservation is not started until the route change of the core NW is completed. That is, the route change is performed at the time of trigger reception on the metro NW side, but the bandwidth change is not started until traffic flows after the IP route change on the core NW side.

  The present invention has been made to solve the above-described problems of the technology, and an object of the present invention is to provide a bandwidth control device that can perform dynamic bandwidth control efficiently at high speed. .

In order to achieve the above object, a bandwidth control device of the present invention is a bandwidth control device that controls a bandwidth of a path set in a relay route of a network,
A band movement determination unit that determines a movement of a band between edge nodes of the relay path based on path information received from the edge node, and outputs band movement information indicating a determination result;
A bandwidth allocation calculator for calculating bandwidth allocation according to the bandwidth shift information received from the bandwidth shift determination unit;
A storage unit that holds a bandwidth allocation history that is a history of bandwidth allocation calculated by the bandwidth allocation calculation unit;
The band movement determination unit associates a pre-switching path and a post-switching path, which are connection switching targets, based on the route information, and includes association information between the pre-switching path and the post-switching path. Creating the band movement information,
The bandwidth allocation calculation unit estimates a bandwidth movement amount to the post-switching path from the bandwidth allocation history recorded in the storage unit, and refers to the bandwidth allocation history of the path before switching in the post-switching path. This is a configuration for performing allocation calculation for allocating bandwidth.

  ADVANTAGE OF THE INVENTION According to this invention, the bandwidth reservation with respect to the path switching at the time of edge switching can be accelerated, and the occurrence of data loss due to the delay in bandwidth reservation can be prevented with respect to the edge switching in which significant bandwidth fluctuations occur.

It is a figure for demonstrating the outline | summary of the band control apparatus of one Embodiment of this invention. It is a block diagram which shows the example of 1 structure of the zone | band control apparatus of 1st Embodiment. It is a flowchart which shows the procedure of the band movement determination method by the band movement determination part shown in FIG. It is a figure which shows an example of trigger information, transfer setting information, and path information. It is a figure for demonstrating the operation | movement procedure shown in FIG. It is a flowchart which shows the procedure of the band allocation calculation method by the band allocation calculation part shown in FIG. It is a figure which shows an example of a bandwidth allocation history. It is a figure for demonstrating the specific example of the band movement determination method shown in FIG. It is a figure for demonstrating the specific example of the band movement determination method shown in FIG. It is a figure for demonstrating the specific example of the band allocation calculation method shown in FIG. It is a sequence diagram which shows operation | movement of the band control apparatus of 1st Embodiment. 1 is a block diagram illustrating a configuration example of a bandwidth control device according to a first embodiment. FIG. 3 is a flowchart illustrating a procedure of a traffic amount collection method according to the first embodiment. It is a flowchart which shows the procedure of the presence determination method of the failure edge function path information in Example 1, and the path | pass determination method after switching. It is a figure for demonstrating the operation | movement procedure shown to FIG. 13 and FIG. FIG. 10 is a sequence diagram illustrating an operation of the bandwidth control apparatus according to the second embodiment. FIG. 10 is a diagram for explaining the operation of the bandwidth control apparatus according to the third embodiment. FIG. 10 is a diagram for explaining the operation of the bandwidth control apparatus according to the third embodiment. It is a figure for demonstrating the case where several edges are accommodated in 1 path | pass. It is a figure for demonstrating the operation | movement of the band estimation by the number of addresses in 2nd Embodiment. FIG. 10 is a block diagram illustrating a configuration example of a bandwidth control device according to a fourth embodiment. In Example 4, it is a figure for demonstrating the method to exclude the address which is not utilized. FIG. 10 is a diagram for explaining the operation of the bandwidth control apparatus according to the fifth embodiment. FIG. 10 is a block diagram illustrating a configuration example of a bandwidth control device according to a sixth embodiment. It is a block diagram which shows an example at the time of comprising the conventional wide area metro NW by the optical L2NW. FIG. 26 is a block diagram illustrating a case where the techniques disclosed in Non-Patent Documents 1 and 2 are applied to the optical L2NW illustrated in FIG. FIG. 27 is a block diagram illustrating a configuration example of a device that performs bandwidth control in the optical L2NW illustrated in FIG. 26. It is a sequence diagram for demonstrating the factor 1 of the problem of a prior art. It is a sequence diagram for demonstrating the factor 2 of the problem of a prior art.

  The bandwidth control apparatus according to an embodiment of the present invention is characterized in that bandwidth control is dynamically performed using bandwidth movement path determination and path bandwidth history before switching.

  FIG. 1 is a diagram for explaining an outline of a bandwidth control apparatus according to an embodiment of the present invention.

  In the bandwidth control apparatus according to the embodiment of the present invention, in addition to the address control associated with the connection switching of the edge node of the relay route, the bandwidth control is performed to determine the bandwidth shift between the edge nodes of the relay route based on the route information. The band movement information indicating the determination result is reflected in the path band adjustment process.

  Specifically, the bandwidth allocation history, which is the bandwidth allocation history calculated by the bandwidth allocation calculation unit, is stored in the storage unit, and the bandwidth shift determination unit performs connection switching based on the route information including the route change information. Associating the pre-switching path and the post-switching path, which are the target paths, and notifying the band allocation calculating unit including the information on the association between the pre-switching path and the post-switching path, and the band allocation calculating unit Bandwidth allocation is calculated based on movement information and bandwidth allocation history.

  As described above, in the bandwidth control device according to the embodiment of the present invention, the pre-switching path and the post-switching path are determined based on the address movement information accompanying the route change, and the bandwidth shift amount is estimated from the bandwidth history of the pre-switching path. By doing so, the relay band is changed simultaneously with the transfer control update.

In contrast to recovery by edge failure switching, the method of checking the traffic volume and changing the bandwidth delays the follow-up and delays the recovery until the necessary bandwidth is secured, as shown in FIG. The band movement position is determined from the change, and the band change is also started at the determination timing.
Hereinafter, embodiments of the bandwidth control device of the present invention will be described in detail. In the embodiment, the case where the relay path is the optical L2NW will be described. However, the relay path is not limited to the optical L2NW.

(First embodiment)
The configuration of the bandwidth control device of this embodiment will be described. FIG. 2 is a block diagram illustrating a configuration example of the bandwidth control apparatus according to the present embodiment.

  As shown in FIG. 2, the bandwidth control device 10 is communicably connected to a path endpoint node 50 that is an edge node of the optical L2NW. In FIG. 1, for the sake of simplicity, a case where one path end point node 50 is connected is shown, but the bandwidth control apparatus 10 is connected to a plurality of path end point nodes 50.

  The path endpoint node 50 includes a learning function unit 51, a connection information notification unit 52, a transfer control unit 53, a path selection transfer unit 54, a traffic amount information notification unit 55, an output control unit 56, and an output unit 57. Have

  The learning function unit 51 learns the source (Source: SRC) and destination (Destination: DST) addresses of the input packet and notifies the connection information notification unit 52 of address information including these addresses. An input packet for notifying a route change corresponds to a route change trigger. The connection information notification unit 52 transmits the address information notified from the learning function unit 51 to the bandwidth control device 10. The transfer control unit 53 instructs the path change by controlling the path selection transfer unit 54 according to the transfer setting information received from the bandwidth control device 10.

  The path selection transfer unit 54 passes the packet to the output unit 57 under the control of the transfer control unit 53. The traffic amount information notification unit 55 transmits traffic amount information, which is information on the traffic amount for each path, to the bandwidth control device 10. The output control unit 56 controls the output unit 57 according to the output information received from the band control device 10. The output unit 57 transmits the traffic amount information to the traffic amount information notification unit 55 and outputs a packet according to the control of the output control unit 56.

  In the present embodiment, the learning function unit 51, the connection information notification unit 52, the transfer control unit 53, the path selection transfer unit 54, the traffic amount information notification unit 55, the output control unit 56, and the output unit 57 execute each function. However, a CPU (Central Processing Unit) (not shown) and a memory (not shown) for storing a program are provided in advance in the path endpoint node 50, and the CPU executes processing according to the program. Thus, a part or the whole may be configured virtually.

  The bandwidth control device 10 includes an address control unit 11 and a path bandwidth adjustment unit 13. The address control unit 11 includes a storage unit 12, a connection information management unit 21, a transfer setting control unit 22, a setting command creation unit 23, and a band movement determination unit 31. The storage unit 12 stores external connection information, transfer setting information, and route / path information. In the present embodiment, a band shift determination unit 31 having a band shift determination function is added.

  The path bandwidth adjustment unit 13 includes a storage unit 14, a traffic collection unit 24, a bandwidth allocation calculation unit 32, and an output command creation unit 25. In the present embodiment, the bandwidth allocation history is stored in the storage unit 14, and the bandwidth allocation calculation function in the bandwidth allocation calculation unit shown in FIG. 27 is changed.

  The configuration of the address control unit 11 shown in FIG. 2 will be described in detail.

  The external connection information is information for managing an address (for example, a MAC (Media Access Control) address) for each position in the optical L2 NW for the path endpoint node 50.

  The transfer setting information is managed in a table in which a start path end point node ID, a transfer path ID, an end point path end node ID, and an external output port ID are recorded corresponding to the destination address described in the packet header. .

  The path / path information is managed in a table in which path identifiers, start path end point node IDs, and end path end point node IDs are recorded corresponding to path IDs that differ for each path. This “path identifier” is an identifier described in the packet, and differs depending on the technology for realizing the relay route. For example, it may be an additional MAC address for MAC encapsulation or an MPLS header. Hereinafter, a case where the path identifier is a bit string having the same value as the path ID will be described.

  The connection information management unit 21 receives the address information from the path endpoint node 50. When the address information includes route change information, the connection information management unit 21 updates the external connection information according to the address information and controls the transfer of the address information. Notification to the unit 22. The route change information corresponds to route information including moving address information that is information of address movement accompanying the route change.

  When the transfer setting control unit 22 receives the moving address information from the connection information management unit 21, the transfer setting control unit 22 updates the transfer setting information and the route / path information corresponding to the moving address information, and sets the updated transfer setting information to the setting command generation unit. 23. The setting command creation unit 23 transmits the transfer setting information received from the transfer setting control unit 22 to the path endpoint node 50.

  The band movement determination unit 31 determines the correspondence between the path before switching and the path after switching from the transfer setting information, the route / path information, and the movement address information of the reception trigger, and includes the information on the correspondence between the path before switching and the path after switching. The movement information is transmitted to the band allocation calculation unit 32, and the band allocation calculation unit 32 is instructed to perform band allocation using the band movement information.

  The configuration of the path bandwidth adjustment unit 13 shown in FIG. 2 will be described in detail.

  In the bandwidth allocation history, information on the allocated bandwidth amount and the movement candidate flag indicating whether or not it is a movement candidate at the time of the next allocation calculation corresponds to the path, and the bandwidth allocation calculation unit 32 performs the allocation calculation. Recorded each time it is done.

  The traffic collection unit 24 passes the traffic amount information periodically received from the path endpoint node 50 to the bandwidth allocation calculation unit 32.

  When the band allocation information is received from the band shift determining unit 31, the band allocation calculating unit 32 switches the band allocation to the post-switching path from the band allocation history without using the traffic amount information collected by the traffic collecting unit 24. The bandwidth transfer amount to the subsequent path is estimated, and the allocation calculation for allocating the bandwidth amount to the post-switching path is performed with reference to the bandwidth allocation history of the path before switching. Here, it is assumed that the bandwidth allocation calculation unit 32 performs allocation calculation at regular intervals.

  When receiving the allocation calculation result from the bandwidth allocation calculation unit 32, the output command generation unit 25 generates output information according to the allocation calculation and transmits it to the path endpoint node 50.

  In the present embodiment, the connection information management unit 21, the transfer setting control unit 22, the setting command creation unit 23, the band movement determination unit 31, the traffic collection unit 24, the output command creation unit 25, and the bandwidth allocation calculation unit 32 have various functions. The bandwidth control device 10 is provided in advance with a CPU (not shown) and a memory (not shown) for storing a program, and the CPU executes processing according to the program. A part or all of them may be configured virtually.

  Moreover, in FIG. 2, although the memory | storage part 12 and the memory | storage part 14 are shown by the separate structure, you may comprise by the same memory | storage device.

  Next, the operation of the bandwidth control device 10 of this embodiment will be described.

  FIG. 3 is a flowchart showing the procedure of the band movement determination method by the band movement determination unit shown in FIG. FIG. 4 is a diagram illustrating an example of trigger information, transfer setting information, and path information. FIG. 5 is a diagram for explaining the operation procedure shown in FIG.

  In the present embodiment, as illustrated in FIG. 4, the trigger information will be described in the case of including a new learning address, an input path endpoint node ID, and an input port ID.

  As shown in FIG. 4, the transfer setting information is a table in which a destination address described in a packet, a start path end node ID, a transfer path ID, an end path end node ID, and an external output port ID are provided as records. Managed. As shown in FIG. 4, the path information is managed in a table in which a path ID, a path identifier, a start point end point node ID, and an end point end point node ID are provided as records.

  As illustrated in FIG. 3, when the band movement determination unit 31 receives a trigger, the band movement determination unit 31 determines an edge-side end point of the post-switching path (step 101). FIG. 5A shows the trigger input path end point node at that time.

  Subsequently, the band shift determination unit 31 searches whether the change address is set as the DST MAC in the transfer setting information (step 102). If it is set, the band movement determination unit 31 determines the access-side endpoint of the post-switching / pre-switching path (step 103). FIG. 5B shows the path end point node of the path start point in which the changed address retrieved and extracted is set as the DST MAC. For example, the band movement determination unit 31 extracts a path set with the trigger address (GARP SRC MAC) as the DST MAC. GARP is an example of a route change trigger.

  Thereafter, the band shift determination unit 31 determines the end point on the edge side of the post-switching path (step 104). FIG. 5C shows the end point path end point node of the path extracted by the search.

  Furthermore, the band movement determination unit 31 associates the pre-switching path with the post-switching path (step 105). The band movement determination unit 31 associates the pre-switching path and the post-switching path that have the same start point path end point node (FIG. 5D).

  After step 105, the band movement determination unit 31 notifies the band allocation calculation unit 32 of the correspondence information between the path before switching and the path after switching (step 106), and then returns to the trigger standby state. In step 102, if the change address is not set as DST MAC in the transfer setting information, the band shift determination unit 31 determines that band shift does not occur (no new learning or no accommodating user at the switching edge). (Step 107), it returns to the trigger standby state.

  FIG. 6 is a flowchart showing the procedure of the bandwidth allocation calculation method by the bandwidth allocation calculator shown in FIG. FIG. 7 is a diagram showing an example of the bandwidth allocation history.

  As shown in FIG. 7, in the bandwidth allocation history, a bandwidth amount and a movement candidate flag are recorded every time bandwidth allocation is performed, corresponding to the path ID. FIG. 7 shows a state in which the allocated bandwidths before and once are recorded.

  When receiving the calculation start instruction, the bandwidth allocation calculation unit 32 determines whether or not the bandwidth movement information is received (step 151). When the band movement information is not received, the band allocation calculation unit 32 allocates a band to each path using the collected traffic amount information as in the conventional technique (step 153).

  On the other hand, if it is determined in step 151 that the band movement information is received, the band allocation calculation unit 32 does not use the traffic information for band allocation of the band movement information to the post-switching path, and stores the history of the path before switching. The bandwidth amount is assigned with reference to (step 152). Thereafter, the bandwidth allocation calculation unit 32 notifies each path endpoint node of the bandwidth allocation result, as in the prior art (step 154).

  The bandwidth allocation calculation unit 32 records the bandwidth allocation result of each path in the bandwidth allocation history, and sets a movement candidate flag for the previous allocation result for a path whose bandwidth has been significantly reduced (step 155). Thereafter, the bandwidth allocation calculation unit 32 enters a state of waiting for calculation start.

  Next, a specific example of the band movement determination method described with reference to FIG. 3 will be described. 8 and 9 are diagrams for explaining a specific example of the band movement determination method.

  Here, as shown in band shift determination (1) in FIG. 8, it is assumed that the IDs of the path endpoint nodes are 1 to 4, and the respective MACs and VLANs are as shown in the figure.

  In the band shift determination (1) shown in FIG. 8, the path endpoint node 1 switches to the path endpoint node 2, and two packets of GARP (SRC MAC A, VLAN a) and GARP (SRC MAC A, VLAN b) are triggered. It shall be. The trigger information at this time is shown in FIG.

  In the band movement determination (2), the band movement determination unit 31 extracts a path set with the trigger address (GARP SRC MAC) as the DST MAC. In (2) shown in the transfer setting information table of FIG. 9, the band shift determination unit 31 pays attention to the start path end point node IDs “3” and “4”.

  Subsequently, in the band movement determinations (3) and (4), the band movement determination unit 31 refers to the path information, and associates the path before switching and the path after switching having the same start point endpoint node (ID “1”). The band movement determination unit 31 updates the band movement information to “path 1 → path 3” and “path 2 → path 4”.

  Next, a specific example of the bandwidth allocation calculation method described with reference to FIG. 6 will be described. FIG. 10 is a diagram for explaining a specific example of the bandwidth allocation calculation method.

  Upon receiving the band movement information from the band movement determining unit 31, the band allocation calculating unit 32 sets a movement candidate flag on the path where the traffic volume has been allocated and the traffic amount has become 0, and stores it in the history. The example of FIG. 10 shows a case where the bandwidth amount 60 Gbps allocated to the path 1 is allocated to the path 3 and 40 Gbps allocated to the path 2 is allocated to the path 4 at the next bandwidth allocation calculation.

  The effect of the bandwidth control apparatus of this embodiment will be described with reference to the drawings. FIG. 11 is a sequence diagram showing the operation of the bandwidth control apparatus of this embodiment.

  Comparing FIG. 29 showing the conventional case and FIG. 11 of the present embodiment, it can be seen that the traffic inflow waiting time and the traffic observation time are shortened in the present embodiment.

  According to the present embodiment, by estimating the moving band from the path change trigger and starting the band change, the traffic observation time and the waiting time until the traffic inflow are shortened, and the restoration time that has been completed until the band is secured is shortened. be able to. As a result, it is possible to prevent the occurrence of data loss due to the delay in securing the band with respect to edge switching in which a significant band fluctuation occurs.

  An example that can be derived from the bandwidth control apparatus according to the above-described embodiment will be described below. In the embodiment, a configuration different from the configuration shown in FIG. 2 will be described in detail, and a detailed description of the same configuration will be omitted.

  In this embodiment, a path with a traffic volume of 0 is regarded as a fault edge connection path, and a path before switching and a path after switching are extracted by combining with a trigger inflow end point.

  The configuration of the bandwidth control apparatus according to this embodiment will be described.

  FIG. 12 is a block diagram illustrating a configuration example of the bandwidth control apparatus according to the present embodiment.

  What is different from the configuration shown in FIG. 2 is the function of the traffic collection unit. The traffic collection unit 33 according to the present embodiment determines that a path to which a bandwidth is allocated but whose bandwidth usage has changed to 0 is a connection path with a fault edge, and includes a fault edge connection path including information on the path ID. Information is notified to the band movement determination unit 31.

  The band movement determination unit 31 determines the presence / absence of failure edge connection information, and if there is failure edge connection information, associates the pre-switching path and the post-switching path based on the failure edge connection information and the route information of the route change trigger. .

  Next, the operation of the bandwidth control apparatus 10 of this embodiment will be described.

  FIG. 13 is a flowchart showing the procedure of the traffic volume collection method in this embodiment, and FIG. 14 is a flowchart showing the procedure of the failure edge function path information presence / absence determination method and the post-switching path determination method in this embodiment. is there. FIG. 15 is a diagram for explaining the operation procedure shown in FIGS. 13 and 14.

  As shown in FIG. 13, the traffic collection unit 33 determines whether or not there is a path with a traffic volume of 0 (step 201). It is determined whether or not a bandwidth is assigned to the path of the quantity “0” (step 202).

  When the bandwidth is allocated, the traffic collection unit 33 extracts the path where the traffic amount is 0 as shown in FIG. 15A, and connects the path ID with the traffic amount of 0 and the allocated bandwidth as a fault edge connection. The band movement determination unit 31 is notified as path information (step 203). Subsequently, the traffic collection unit 33 notifies the bandwidth allocation calculation unit 32 of the traffic amount of each path (step 204).

  If there is no path with a traffic volume of 0 as determined in step 201 and no bandwidth is allocated to a path with a traffic volume of “0” as determined in step 202, the traffic collection unit 33 performs the process in step 204. move on.

  As shown in FIG. 14, when the path change trigger is received, the band movement determination unit 31 determines the edge side end point (trigger input path end point node) of the path after switching (step 251). Subsequently, the band movement determination unit 31 determines whether or not fault edge connection path information is received from the traffic collection unit 33 (step 252). When the failure edge connection path information is received, the band shift determination unit 31 uses the failure edge connection path as the pre-switch path information, and determines the post-switch path together with the trigger input path end point node (step 253). Since the pre-switching path has been extracted, the band movement determination unit 31 can determine the pre-switching path and the post-switching path based on the edge side end points of the post-switching path (FIG. 15B).

  Thereafter, the band movement determination unit 31 associates the pre-switching path with the post-switching path (step 254). As shown in FIG. 15C, the paths of the same start point path end point node correspond to each other.

  After step 254, the band movement determining unit 31 notifies the band allocation calculating unit 32 of the correspondence information between the path before switching and the path after switching (step 255), and then returns to the trigger standby state. If there is no failure edge connection path information as a result of the determination in step 252, the band shift determination unit 31 performs the processing of steps 103 to 105 in the flow illustrated in FIG. 3, and then proceeds to step 255.

  In the present embodiment, it is possible to make the band movement determination earlier by detecting the failure edge by the traffic collecting unit.

  In the above-described embodiment, the case where the bandwidth allocation calculation unit periodically performs allocation calculation at regular time intervals has been described. However, the present embodiment is not limited to periodically performing the allocation calculation. Control is performed.

  In the bandwidth control apparatus according to the present embodiment, the bandwidth allocation calculation unit 32 realizes the interrupt calculation by performing the allocation calculation at the reception timing of the path change trigger.

  FIG. 16 is a sequence diagram showing the operation of the bandwidth control apparatus of this embodiment.

  Compared with the sequence diagram shown in FIG. 11, it can be seen that the bandwidth change is reflected before the next bandwidth allocation calculation as shown in FIG.

  According to the present embodiment, when the band movement information is received, it is possible to further reduce the waiting time until the start of calculation by starting the calculation by interrupting without waiting for the next calculation start timing. Become.

  In this embodiment, when the address is realized by the MAC and the VLAN, the pre-switching path and the post-switching path are determined by the first trigger MAC without paying attention to the VLAN.

  In the bandwidth control apparatus 10 according to the present embodiment, when the path movement determination unit 31 receives route information from a plurality of triggers with the same destination from the path endpoint node 50, the MAC address of the first trigger received first is used. The path before switching and the path after switching are associated.

  17 and 18 are diagrams for explaining the operation of the bandwidth control apparatus of this embodiment.

  Here, it is assumed that the IDs of the path endpoint nodes are 1 to 4 and the respective MACs and VLANs are as shown in the figure, as shown in band shift determination (1) in FIG.

  In the band shift determination (1) shown in FIG. 17, the path end point node 1 is switched to the path end point node 2, and two packets of GARP (SRC MAC A, VLAN a) and GARP (SRC MAC A, VLAN b) are triggered. It shall be. The trigger information at this time is shown in FIG. In this embodiment, only the first trigger is used.

  In the band shift determination (2), the band shift determination unit 31 extracts a path set with the first trigger address (GARP SRC MAC) as the DST MAC. In (2) shown in the transfer setting information table of FIG. 18, the band shift determination unit 31 pays attention to the start path end point node IDs “3” and “4”.

  Subsequently, in the band movement determinations (3) and (4), the band movement determination unit 31 refers to the path information, and associates the path before switching and the path after switching having the same start point endpoint node (ID “1”). The band movement determination unit 31 updates the band movement information to “path 1 → path 3” and “path 2 → path 4”.

  According to the present embodiment, since the band movement determination is performed by the first trigger, it becomes possible to notify the band allocation information to the band allocation calculation unit earlier.

(Second Embodiment)
In the first embodiment, the case where one edge is accommodated in one path has been described. However, in the present embodiment, the bandwidth control method of the present invention is extended when a plurality of edges are accommodated in one path.

Since switching of one edge does not always move all traffic in a path, it is necessary to determine the amount of traffic that moves along with edge switching in addition to the paths before and after switching. It is a figure for demonstrating the case where an edge is accommodated.

  In the present embodiment, in the bandwidth control apparatus 10 shown in FIG. 2, the bandwidth movement determination unit 31 estimates the moving bandwidth ratio from the number of addresses in the path, and the band movement information including the estimated moving bandwidth ratio information is obtained. The bandwidth allocation calculation unit 32 is notified.

  FIG. 20 is a diagram for explaining the operation of bandwidth estimation based on the number of addresses.

The estimated movement band in the case of forward change is calculated by the following equation.
Estimated moving band = number of moving addresses / total number of addresses in forward path × path band before switching The estimated number of moving addresses and the estimated moving band in the case of reverse direction change are respectively calculated by the following equations.
Estimated number of moving addresses = number of moving addresses / total number of addresses in forward path × total number of addresses in reverse path Estimated moving band = reverse path band before switching × number of estimated moving addresses / total number of backward addresses before switching Here, it is assumed that the estimation is as follows.
(Uplink / Downlink) The traffic volume per address in the same path is uniform.
(Down) The accommodation to the edge / server is uniform within the same ground.

  Hereinafter, examples that can be derived from the bandwidth control apparatus of this embodiment will be described. In the examples, configurations different from those of the first embodiment will be described in detail, and detailed descriptions of similar configurations will be omitted.

  In this embodiment, the moving band is estimated using the pre-switching path band and the number of moving addresses, and the estimated result is used for band movement information.

  The configuration of the bandwidth control apparatus according to this embodiment will be described. FIG. 21 is a block diagram showing an example of the configuration of the bandwidth control apparatus of this embodiment. The band movement allocation unit 31 estimates the movement band allocation from the total number of addresses in the path and the number of address movements, and associates the path before switching and the path after switching based on the estimation result.

  In this embodiment, since the number of addresses is used for bandwidth estimation, a difference in bandwidth amount becomes large unless addresses that are not actually used are excluded. Therefore, unused addresses are excluded as follows.

  Next, a method for excluding unused addresses from the band estimation address will be described. FIG. 22 is a diagram for explaining a method of excluding addresses that are not used in this embodiment.

  The path end point node 50 measures the time from the last hit for each transfer entry of the destination address, and notifies the band movement determination unit 31 that the address has been expired after a predetermined time has elapsed. When the information on the address that has been subjected is received from the path endpoint node 50, the band movement determination unit 31 removes the address that has been exposed from the parameter x of the band estimation, as shown in FIG. In this way, by expiring addresses that have not been used for a long time, the addresses that are not used are excluded from the total number of addresses in the path at the time of bandwidth estimation. However, this Expire is an operation different from Expire performed in the transfer setting.

  According to the present embodiment, it is possible to estimate the allocation of the movement band based on the number of movements of the address without depending on the traffic amount fluctuation or the trigger.

  In the present embodiment, when the edge address is different for each service, weighted band estimation is performed by service identification, and the band estimation result is used for band movement information.

  In the bandwidth control apparatus 10 of the present embodiment, when the communication service is different for each address, the bandwidth movement determination unit 31 estimates the traffic amount by weighting the service characteristics for each address.

  FIG. 23 is a diagram for explaining the operation of the bandwidth control apparatus according to this embodiment.

The estimated movement band in the case of forward change is calculated by the following equation.
Estimated mobile bandwidth = Product sum of weights by service of mobile addresses / Product sum of weights by service of all addresses in the forward path x Pre-switching path bandwidth The number of estimated mobile addresses and the estimated mobile bandwidth in the case of reverse direction change are as follows: Calculated by the formula.
Estimated number of moving addresses = Product sum of weights by service of moving addresses / Product sum of weights by service of all addresses in the forward path × Total number of addresses in reverse path Estimated moving band = Reverse path band before switching × The above estimated movement The number of addresses / the total number of reverse addresses before switching According to the present embodiment, it is possible to reflect the estimation result of the band movement weighted for each service in the band allocation calculation.

  In this embodiment, traffic in a path is measured for each address and used for moving band notification.

  FIG. 24 is a block diagram showing an example of the configuration of the bandwidth control apparatus of this embodiment.

  In addition to the configuration shown in FIG. 2, the path setting node 70 of the present exemplary embodiment includes an address unit traffic amount information notifying unit 61 that collects traffic amounts corresponding to addresses and notifies the bandwidth control device 10 of the information. Have. Further, the path selection / transfer unit 62 of this embodiment has a function of a packet counter that measures the number of packets corresponding to an address, in addition to the function equivalent to the packet selection / transfer unit 54 shown in FIG.

  In the bandwidth control apparatus 10 of the present embodiment, address unit traffic volume information, which is traffic volume information corresponding to addresses, is stored in the storage unit 12 shown in FIG. The address control unit 11 further includes an address unit traffic information collection unit 34 that stores address unit traffic volume information received from the path contact node 70 in the storage unit 12.

  The band movement determination unit 31 uses the address unit traffic amount information in the path for the band movement determination.

  According to the present embodiment, it becomes possible to reflect the traffic amount for each address in the bandwidth allocation calculation.

  In addition, when the bandwidth allocation calculation unit described in the first embodiment performs allocation calculation at regular time intervals, each of Example 5 and Example 6 to which Example 4 and Example 4 are applied is applied. Also good.

  In addition, the sixth embodiment may be applied to each of the second embodiment and the third embodiment to which the first embodiment and the first embodiment are applied. The second embodiment may be applied to each of the fifth embodiment and the sixth embodiment in which the fourth embodiment and the fourth embodiment are applied. Furthermore, the third embodiment to which the second embodiment is applied may be applied to each of the fourth embodiment, the fifth embodiment to which the fourth embodiment is applied, and the sixth embodiment.

DESCRIPTION OF SYMBOLS 10 Band control apparatus 11 Address control part 13 Path band adjustment part 31 Band movement determination part 32 Band allocation calculation part 33 Traffic collection part

Claims (8)

A bandwidth control device for controlling the bandwidth of a path set in a network relay route,
A band movement determination unit that determines a movement of a band between edge nodes of the relay path based on path information received from the edge node, and outputs band movement information indicating a determination result;
A bandwidth allocation calculator for calculating bandwidth allocation according to the bandwidth shift information received from the bandwidth shift determination unit;
A storage unit that holds a bandwidth allocation history that is a history of bandwidth allocation calculated by the bandwidth allocation calculation unit;
The band movement determination unit associates a pre-switching path and a post-switching path, which are connection switching targets, based on the route information, and includes association information between the pre-switching path and the post-switching path. Creating the band movement information,
The bandwidth allocation calculation unit estimates a bandwidth movement amount to the post-switching path from the bandwidth allocation history recorded in the storage unit, and refers to the bandwidth allocation history of the path before switching in the post-switching path. A bandwidth control device that performs allocation calculation for allocating bandwidth. The bandwidth control device according to claim 1, wherein
The bandwidth control apparatus, wherein the bandwidth allocation calculation unit performs the allocation calculation at regular intervals. The bandwidth control device according to claim 2, wherein
A traffic collecting unit that determines a path whose bandwidth usage has changed to 0 as a connection path to a fault edge and notifies the band movement determination unit of fault edge connection information including information on the fault edge and the path; ,
The band movement determination unit determines the presence / absence of failure edge connection information, and determines that there is failure edge connection information, and corresponds to the pre-switching path and the post-switching path based on the failure edge connection information and the route information. A band control device. The bandwidth control device according to claim 2, wherein
The band allocation calculation unit performs the allocation calculation at a reception timing of a path change trigger including the path information from the edge node. The bandwidth control device according to claim 2, wherein
The band movement determining unit associates the pre-switching path and the post-switching path using information on a destination MAC address by a trigger in which a MAC address with the same destination is set as the route information. The bandwidth control device according to claim 2, wherein
It further includes an address unit traffic amount information collection unit that collects traffic amount information corresponding to addresses, and stores address unit traffic amount information that is traffic amount information for each address in the storage unit,
The band movement determination unit is a band control device that uses the address unit traffic amount information in a path for band movement determination. The bandwidth control device according to claim 2, wherein
When the band movement determination unit associates the path before switching and the path after switching based on the route information, the band movement determination unit estimates movement band allocation from the total number of addresses in the path and the number of moving addresses, A bandwidth control device that associates the pre-switching path with the post-switching path. The bandwidth control device according to claim 7, wherein
The band movement determination unit is a band control device that performs weighted band estimation based on service identification.