JP6234916B2 - Network system and control method thereof - Google Patents

Description

The present invention relates to a network system using a TDM (Time Division Multiplexing) method and a control method thereof.

FIG. 8 shows an example of scheduling in a network system using the TDM scheme.
In FIG. 8, nodes A to E are connected in a ring shape via a link, and are operated by determining the timing (hereinafter referred to as TS (time slot)) for transmitting and receiving data in advance. Here, a table indicating the TS allocation state is a schedule table, and an operation for determining the schedule table is a scheduling. In scheduling, it is assumed that a common TS is assigned to a link through which a path between nodes passes, and the same TS cannot be used on the same link.

In the schedule table shown in FIG. 8, the TDM frame length t in repetition units is set to 500 μs.
TS1 to TS5 of 100 μs are set as 1 TS (2G on the path A-C between the nodes A and C.
bps) is assigned, and 2TS (4 Gbps) is assigned to the path BD between the nodes B and D. In this case, TS1 is assigned to the path A-C, and TS2 and TS3 are assigned to the path BD.

Various heuristic solutions have been proposed for such scheduling. Conventionally, T is assigned sequentially for each given topology and traffic, one path at a time.
A method of searching for S (Non-Patent Document 1) is common. However, for each path, it is necessary to check the status of TS vacancy on all links through which the path passes, and it is difficult to increase the scale of the network system. Hierarchical scheduling (Non-Patent Document 2) has been proposed for this problem.

In hierarchical scheduling, multiple paths are logically bundled into multiple groups,
As shown in FIG. 9, scheduling is performed for each group to create a schedule table, and the schedule tables are connected to perform overall scheduling. At this time, the scheduling is a hierarchical structure that performs scheduling focusing on each group and arbitration processing of the scheduling results of each group. It is possible to cope with the large scale of the system.

X. Zhang and C. Qiao, “Pipelined transmission scheduling in all-optical TDM / WDM rings,” in Proc. Int. Conf. Computer Communication and Networks, pp. 144-149, Sept. 1997. Nakagawa, Hattori, Katayama, Ogawa, “Performance evaluation of optical TDM resource allocation using path grouping,” Society Conference of IEICE, B-12-15, September 2014. X. Cao et al., “Waveband Switching in Optical Networks,” IEEE Commun. Mag., Vol.41, no.4, pp.105-112, Apr. 2003.

When grouping multiple paths, First-Fit and First-Fit path described in Non-Patent Document 2
In the pair method and the grouping method in the wavelength group path network described in Non-Patent Document 3, the allocation efficiency decreases, the traffic volume that can be accommodated by the same resource decreases, or the required resource amount for accommodating the same traffic volume increases. There is a problem to do.

FIG. 10 shows an example of grouping by the first-fit path pair method.
In FIG. 10, the path information set for the nodes A to E based on the network information includes a path ID, a start point node, and an end point node. Traffic information is allocated T for each path.
S number is shown. Traffic information is updated regularly. The path pair table indicates combinations of paths that go around the ring when two paths are connected. The number of hops is the number of links constituting the path, is acquired from the network information, and is fixed unless there is a node addition or link addition. Fi
In the rst-Fit path pair method, a path group is formed by combining two path pairs in the path pair table in ascending order. The schedule table includes path pair 1 (which is path group 1)
The scheduling results of path 1, 5) and path pair 2 (path 2, 9) are shown.

Here, the path pair table is updated as needed by updating traffic information.
Since the path pair combination by the Fit path pair method is fixed, the calculation time is suppressed. However, since the path pairs are grouped in ascending order, as shown in the schedule table of FIG. 10, TSs that could be used by other paths when not grouped cannot be used, and allocation efficiency may be reduced.

FIG. 11 shows an example of grouping in the wavelength group path network.
Here, four paths having the same route or the same link passing through are grouped. For example, group 1 is a grouping of paths passing between nodes B-C. In this case, since the same TS is not shared by a plurality of paths, an unused TS is generated, and allocation efficiency may be reduced.

The present invention creates a path group by a simple comparison operation considering the traffic distribution of the path,
It is an object of the present invention to provide a network system and a control method thereof that enable scheduling that effectively uses resources of the entire network in a short calculation time.

According to a first aspect of the present invention, there is provided a network system in which a plurality of nodes are connected in a ring shape via a link, and scheduling is performed to determine a schedule table indicating a path and time slot allocation state set between the nodes. For paths set in between
When logically connected, a set of multiple paths that go around the ring is listed as a loop,
Further, for each loop, the standard deviation of the traffic of the constituent path is calculated as the traffic deviation, and a corresponding loop table is created. With reference to the loop table, the traffic deviation is equal to or less than a predetermined threshold and the number of constituent paths is the predetermined number. A path group creating means for creating a path group by combining one or a plurality of loops under a condition that falls within an allowable range, and scheduling includes:
The schedule table is determined based on the combination of path group constituent paths.

In the network system of the first invention, the path group creating means searches in order from a loop having a large number of constituent paths with reference to the loop table, and when there is no loop whose traffic deviation is equal to or less than a predetermined threshold, the threshold value Continue searching for loops by changing to a larger value, excluding loops that contain paths that overlap the loop's configuration path determined as a path group, and continuing to search for loops that become the next path group .

In the network system of the first invention, the path group creation means searches the loop table in order from the loop having the smallest traffic deviation, or starts from the loop having the largest number of constituent paths and starts from the loop having the smallest traffic deviation. If there is no loop in which traffic deviation is less than or equal to the predetermined threshold, the threshold is changed to a large value and the loop search is continued, and a path that overlaps with the loop configuration path determined as a path group is detected. A process of continuing to search for a loop that becomes the next path group is performed by excluding the included loop.

In the network system according to the first aspect, the path group creation means is configured so that when a loop corresponding to the condition becomes a path group and the number of loops corresponding to the condition becomes 1 or 0, the path is not assigned to the path group. Is processed as a path group by a predetermined number.

According to a second aspect of the present invention, there is provided a network system control method in which a plurality of nodes are connected in a ring shape via a link, and scheduling is performed to determine a schedule table indicating a path and time slot allocation state set between the nodes. For paths set between nodes, a set of multiple paths that circulate around the ring when logically connected is listed as a loop, and the standard deviation of the traffic of the constituent paths is calculated as a traffic deviation for each loop. Step 1 for creating an associated loop table and a combination of one or a plurality of loops under the condition that the traffic deviation is equal to or less than a predetermined threshold and the number of constituent paths is within a predetermined allowable range with reference to the loop table Creating a group, and the scheduling comprises: To determine the schedule table by a combination of growth path.

In the network system control method according to the second aspect of the present invention, step 2 refers to the loop table in order from the loop having the largest number of constituent paths, and when there is no loop whose traffic deviation is equal to or less than a predetermined threshold, Continue the loop search by changing the threshold value to a larger value, and exclude the loop that contains a path that overlaps the loop configuration path that has been established as a path group, and continue the search for the loop that will be the next path group. Do.

In the network system control method according to the second aspect of the invention, step 2 refers to the loop table to search in order starting from the loop having the smallest traffic deviation, or from the loop having the largest number of constituent paths to the loop having the smallest traffic deviation. If there is no loop whose traffic deviation is less than or equal to the predetermined threshold, the path is searched for in order and the loop search is continued and the loop search is continued, and the path overlaps with the loop path that is determined as a path group A process that continues searching for the loop that becomes the next path group is performed.

In the network system control method according to the second aspect of the present invention, step 2 is not assigned to a path group when the loop corresponding to the condition becomes a path group and the number of loops corresponding to the condition becomes 1 or 0. A process of distributing a predetermined number of paths as path groups is performed.

The present invention lists loops, further calculates a traffic deviation for each loop, creates a correlated loop table, refers to the loop table, the traffic deviation is less than a predetermined threshold, and the number of constituent paths is predetermined. By creating a path group by combining one or a plurality of loops under a condition that falls within the allowable range, scheduling is performed on the schedule table, and TS resources can be effectively used.

In addition, the search is performed in order from the loop with the largest number of constituent paths, and the loop including the path that overlaps the constituent path of the loop determined as the path group is excluded, and the loop that becomes the next path group is searched, thereby shortening. You can create a pass group in time.

It is a figure which shows the structural example of the path group preparation means in the network system of this invention. It is a figure which shows the process example of the loop preparation part 11 and the traffic deviation calculating part 12. FIG. It is a figure which shows the relationship between the structure path | route of a loop, and a traffic deviation. It is a figure which shows the process example (1) of the loop selection / combination calculating part 14. FIG. It is a figure which shows the process example (2) of the loop selection / combination calculating part 14. FIG. It is a figure which shows the process example (3) of the loop selection / combination calculating part 14. FIG. 6 is a diagram illustrating a processing example of a path distribution calculation unit 15. FIG. It is a figure which shows the example of a scheduling of the network system using a TDM system. It is a figure which shows the example of hierarchical scheduling. It is a figure which shows the example of grouping by a First-Fit path pair system. It is a figure which shows the example of grouping in a wavelength group path | route network.

FIG. 1 shows a configuration example of path group creation means in the network system of the present invention.
In FIG. 1, the path group creation means includes a loop creation unit 11, a traffic deviation calculation unit 1.
2, a loop table 13, a loop selection / combination calculation unit 14, and a path distribution calculation unit 15. The path group creation means is provided in a network controller (scheduler) that manages a schedule table between nodes. Each node provides traffic information (the number of requested TSs) to the network controller, and transfers the TDM signal using a TS corresponding to the schedule table notified from the network controller.

FIG. 2 shows a processing example of the loop creation unit 11 and the traffic deviation calculation unit 12.
In FIG. 2, in a full mesh network in which nodes A to E are connected in a ring shape, an ID is assigned to a path set between the nodes. Here, 1 to 20 are assigned in ascending order of the start point (transmission source) node and the end point (destination) node. Further, the requested TS number (assigned TS number) at the control timing is associated as traffic information of each path. Traffic information is updated from time to time.

The loop creation unit 11 inputs path information (path ID, start point node, end point node) and lists a set of a plurality of paths that make a round of the ring as a “loop” when logically connected. In general, a full mesh network of N nodes ring configuration path number of loops 2, 3, ..., becomes N paths, patterns the number of each configuration pass number is _N C _i relative configuration path number i, 5 Node The number of loops that can be realized in the ring is 26.

The traffic deviation calculation unit 12 inputs the traffic information of each path, and calculates a standard deviation (traffic deviation) corresponding to the traffic (number of assigned TSs) of the constituent paths for each loop listed in the loop creation unit 11. . FIG. 3 shows the relationship between the constituent paths in the loops 19 and 20 and the traffic deviation. The loop 19 includes a path 7 (BD), a path 16 (DE), and a path 18 (E-
B), and the number of TSs is 7, 2, and 5, so the traffic deviation is 2.52. The loop 20 is composed of paths 11, 16, and 19, and the number of TSs is 3, 2, and 3, so the traffic deviation is 0.58.

In this way, the loop table 13 in which the loop ID, the number of paths, the constituent paths, and the traffic deviation are associated is generated. The loop table 13 shown in FIG. 2 sorts the loops 1 to 26 in ascending order of the number of constituent paths 2 to 5 and in ascending order of the constituent path IDs 1 to 20.

The feature of the present invention is that when the loop selection / combination operation unit 14 creates a path group by combining one or more loops of the loop table 13, the loop selection / combination selecting unit 14 selects from the loops having a small traffic deviation, and the constituent paths do not overlap. Combine loops. At this time, if the number G of path groups to be created (for example, 5) is set with respect to the number M of paths obtained from the network information (20 in the above example), the standard number of paths constituting one path group is M / G ( For example, 4
) And the allowable range of the number of constituent paths of one path group is M / G ± α (for example, when α = 1, 3 to 5). The allowable range of the number of constituent paths of one path group may be asymmetric with respect to M / G (for example, 3 to 6).

Here, in order to select from loops having a small traffic deviation, the loop selection / combination calculation unit 14 searches the loop table 13 in descending order of the number of constituent paths, and searches in ascending order of loop IDs for the same constituent paths. There is a method of finding a loop in which the traffic deviation is less than or equal to a threshold value. If there is no loop below the threshold, the threshold is increased and the same search is continued. In this case, since a loop to be a path group is selected from the larger number of constituent paths, FIGS.
As described below, the search efficiency can be improved.

Alternatively, the loop selection / combination calculation unit 14 may sort the loop table 13 in ascending order of traffic deviation and search for the order in which the traffic deviation increases in order to find a loop whose traffic deviation is equal to or less than a threshold value. In this case, if the traffic deviation of the first loop to be searched is less than or equal to the threshold value, the loop is found immediately. If it is not less than or equal to the threshold value, there is no need to search for subsequent loops, and a similar search is performed by increasing the threshold value. Continue.

Further, in the loop selection / combination calculation unit 14, the loop table 13 is sorted in ascending order of traffic deviation for each number of constituent paths, and searched in descending order of the number of constituent paths and in ascending order of traffic deviation, and the traffic deviation is below a threshold value. It is also possible to find a loop that becomes. In this case, the search efficiency can be improved from both the number of constituent paths and the traffic deviation.

The loop selection / combination calculation unit 14 selects one or a plurality of loops and sequentially creates a path group. However, each time a path group is created in the loop table 13, it is not assigned to a path group. A path (unallocated path) is recorded (details will be described later).

FIG. 4 shows a processing example (1) of the loop selection / combination calculation unit 14.
In FIG. 4, the loop selection / combination computing unit 14 searches for loops in descending order of the number of constituent paths in the loop table 13 and in ascending order of loop IDs, and when a loop satisfying the following conditions is found, the path group table Add to 16. Here, a loop with 5 configuration paths
Next, the search proceeds in the order of the loops 21, 22, 23, 24, and 25 having the number of configuration paths of 4, and the loops of the number of configuration paths of 3 and 2 in the following.

The conditions for confirming a pass group are:
(a) The traffic deviation of the loop is below a predetermined threshold (for example, the threshold is 1)
(b) The number of constituent paths is within a predetermined allowable range (for example, 3 to 5)
And In this case, the loop 26 with 5 constituent paths is not selected because the traffic deviation is 1.30 and does not satisfy the condition (a) (× in the figure), and the next loop 21 has the traffic deviation of 0.96 and satisfies the condition (a). Satisfied and the number of configuration paths is 4 and the condition (b) is satisfied, so it is determined as path group 1 (◯ in the figure), and configuration paths 1, 6, 11, and 13 of loop 21 are written to the path group table 16 .

If there is no loop that satisfies the condition (a), the same search is performed while changing the condition (a ′) or the like with an increased threshold. If the condition (b) is not satisfied, a search is made for the next loop to be combined with the path group, and the condition (a) is satisfied until the total number of constituent paths of the combined loops satisfies the condition (b). , (a ′),...

Next, the loop selection / combination calculation unit 14 that has determined the path group 1 excludes a loop that includes a path that overlaps the path constituting the loop 21 of the path group 1 in preparation for creating the next path group. Update the loop table 13 to narrow down the target loop. In this case, the loop 21 of the path group 1 and the loops 1, 3,..., 20, 22 to 26 including the paths overlapping with the constituent paths 1, 6, 11, 13 are flagged as “excluded”. Exclude from Table 13. The updated loop table 13 is shown in FIGS. At this time, the unallocated paths are 16 paths obtained by excluding the configuration paths 1, 6, 11, and 13 of the path group 1 from all 20 paths.

In FIG. 5, the loop selection / combination computing unit 14 searches for loops in descending order of the number of constituent paths in the loop table 13 and in ascending order of loop IDs, and finds a loop that satisfies the above conditions (a) and (b). Added to the path group table 16. Here, the search proceeds in the order of loops 15, 16, 19 with the number of configuration paths 3, and then loops 2, 4, 6, 7, 9, 10 with the number of configuration paths 2. However, since any traffic deviation exceeds 1 and does not satisfy the condition (a), a similar search is performed by changing the threshold value to the condition (a ′) increased to 2, for example. In this case, the loop 15 satisfies the new condition (a ′) with the traffic deviation of 1.15, and satisfies the condition (b) with the number of configuration paths of 3 and satisfies the condition (b). , 12 and 17 are written to the path group table 16.

Note that when the number of loops constituting the loop table 13 is equal to or less than the predetermined number, the loop table 13 is not a method of searching for loops in descending order of the number of constituent paths of the loop table 13 and in ascending order of loop IDs. Sorting may be performed in ascending order of deviation, and the method may be changed so that all loops are searched in ascending order of traffic deviation.

Next, the loop selection / combination calculation unit 14 that has determined the path group 2 prepares a path 15 for the path group 2 and a path that overlaps the constituent paths 2, 12, and 17 in preparation for creating the next path group. The included loops 2, 4, 9, 16 are excluded from the loop table 13. The updated loop table 13 is shown in FIGS. At this time, the unassigned paths exclude the configuration paths 1, 6, 11, and 13 of the path group 1 and the configuration paths 2, 12, and 17 of the path group 2 from all 20 paths.
13 passes.

In FIG. 6, the loop selection / combination computing unit 14 searches for a loop in descending order of the number of constituent paths in the loop table 13 and in ascending order of the loop ID, and a loop satisfying the above conditions (a ′) and (b) When found, it is added to the path group table 16. Here, the search proceeds in the order of the loop 19 with the number of configuration paths 3, and then the loops 6, 7, and 10 with the number of configuration paths 2. However, the traffic deviation of the loop 19 exceeds 2 and does not satisfy the condition (a ′), the next loop 6 also exceeds 2 and the traffic deviation does not satisfy the condition (a ′), and the next loop 7 does not satisfy the condition (a ′). The deviation (1.41) satisfies the condition (a ′), but the number of constituent paths is 2 and the condition (b) is not satisfied. Therefore, the search for the loop to be combined is continued. However, since there is no loop that satisfies the condition (a ′) where the traffic deviation is 2 or less, a similar search is performed by changing the threshold to the condition (a ″) where the threshold is increased to 3, for example.

In this case, the loop 19 with the number of configuration paths 3 satisfies the condition (a ") with a traffic deviation of 2.52, but becomes impossible because the configuration path 18 overlaps with the configuration path of the loop 7, and the next loop 6 is the target. Loop 6 satisfies the condition (a ") with a traffic deviation of 2.12, and there is no duplication of configuration paths, and the number of configuration paths is 2 and 4 together with the loop 7 to satisfy the condition (b). . Therefore, the loops 7 and 6 are determined as the path group 3, and the constituent paths 7, 8, 14, and 18 are written in the path group table 16.

Next, the loop selection / combination calculating unit 14 that has determined the path group 3 prepares to create the next path group, and the loops 6 and 7 of the path group 3 and the constituent paths 7, 8, 14,
A loop 19 including a path overlapping with 18 is excluded from the loop table 13 and the loop 10 remains.
The updated loop table 13 is shown in FIGS. At this time, there are 20 unassigned paths.
9 paths (3, 4, 5, 9, 10, 15,
16, 19, 20). The loop selection / combination calculation unit 14 ends the process when the loop that can be searched in the loop table 13 becomes 1 or 0, and notifies the path distribution calculation unit 15 of the processing.

In FIG. 7, the path distribution calculation unit 15 includes unallocated paths 3, 4, 5, 9, 10, 15, 16, 19,
Assign 20 to pass groups 4 and 5. Here, the constituent paths 16 and 20 of the loop 10 remaining in the path group table 16 are distributed to the path group 4, and the remaining unassigned paths 3, 4, 5
, 9, 10, 15, and 19 are assigned 2 paths as path group 4 and 5 paths as path group 5 in ascending order of path ID. Alternatively, the four paths may be simply assigned as the path group 4 and the remaining five paths may be assigned as the path group 5 in ascending order of the path ID.

The path group table 16 composed of the path groups 1 to 5 completed by the above processing is as follows.
The result is shown in FIG. The schedule table is such a path group table 16
However, it can be seen that, for example, the path groups 1 to 4 including the loop can suppress scheduling of the TS on the schedule table and can perform scheduling using the TS resources effectively.

11 Loop creation unit 12 Traffic deviation calculation unit 13 Loop table 14 Loop selection / combination calculation unit 15 Path distribution calculation unit 16 Path group table

Claims (8)

In a network system in which a plurality of nodes are connected in a ring shape via a link, and scheduling is performed to determine a schedule table indicating an assignment state of paths and time slots set between the nodes,
For the paths set between the nodes, a set of a plurality of paths that circulate around the ring when logically connected is listed as a loop, and the standard deviation of the traffic of the constituent paths is calculated as a traffic deviation for each loop. Create an associated loop table, refer to the loop table, and combine the path group by combining one or more loops under the condition that the traffic deviation is below a predetermined threshold and the number of constituent paths is within a predetermined allowable range. It has a path group creation means to create,
In the scheduling, the schedule table is determined based on a combination of constituent paths of the path group. The network system according to claim 1,
The path group creating means searches the loop table in order from the loop having the largest number of constituent paths, and changes the threshold value to a larger value when there is no loop in which the traffic deviation is equal to or less than the predetermined threshold value. The loop search is continued, the loop including the path that overlaps with the loop configuration path determined as the path group is excluded, and the search for the loop that becomes the next path group is continued. Network system. The network system according to claim 1,
The path group creating means searches the loop table in order from a loop with a small traffic deviation, or searches in order from a loop with a large number of constituent paths and in order from a loop with a small traffic deviation. If there is no loop whose deviation is less than or equal to the predetermined threshold value, the threshold value is changed to a larger value and the loop search is continued, and loops that include a path that overlaps the constituent path of the loop determined as the path group are excluded. Then, the network system is characterized in that the process of continuing to search for a loop that becomes the next path group is performed. The network system according to claim 1,
In the path group creation means, a loop corresponding to the condition becomes the path group,
When the number of loops satisfying the condition becomes 1 or 0, a process of distributing a predetermined number of paths not assigned to the path group as a path group is performed. In a control method of a network system in which a plurality of nodes are connected in a ring shape via a link, and scheduling is performed to determine a schedule table indicating an allocation state of paths and time slots set between the nodes.
For the paths set between the nodes, a set of a plurality of paths that circulate around the ring when logically connected is listed as a loop, and the standard deviation of the traffic of the constituent paths is calculated as a traffic deviation for each loop. Creating a correlated loop table 1;
Creating a path group by combining one or a plurality of loops under the condition that the traffic deviation is equal to or less than a predetermined threshold with reference to the loop table and the number of constituent paths is within a predetermined allowable range;
In the scheduling, the schedule table is determined by a combination of constituent paths of the path group. In the network system control method according to claim 5,
In step 2, the loop table is searched in order from the loop with the largest number of constituent paths, and when there is no loop in which the traffic deviation is less than or equal to the predetermined threshold, the threshold is changed to a larger value. A network characterized by continuing a search for a loop, excluding a loop including a path that overlaps a path constituting a loop determined as the path group, and continuing a search for a loop that becomes a next path group How to control the system. In the network system control method according to claim 5,
The step 2 refers to the loop table to search in order from a loop with a small traffic deviation, or sequentially from a loop with a large number of constituent paths and from a loop with a small traffic deviation, and the traffic deviation is If there is no loop that is less than or equal to the predetermined threshold value, the threshold value is changed to a larger value and loop search is continued, and loops that include paths that overlap the loop configuration path determined as the path group are excluded. A method for controlling a network system, characterized in that processing for continuing to search for a loop that becomes the next path group is performed. In the network system control method according to claim 5,
In step 2, when a loop corresponding to the condition becomes the path group and the number of loops corresponding to the condition becomes 1 or 0, a path that is not assigned to the path group is determined by a predetermined number of path groups. A method for controlling a network system, characterized by performing a process of distributing as a network.

Images