CN103312628B - The dispatching method and device of aggregated links in a kind of packet network - Google Patents

Abstract

The invention discloses a method and device for scheduling aggregation links in a packet-switching network. LSPs currently waiting to preempt resources judge one by one whether each member link satisfies the preemption condition, and select from one or more member links that meet the preemption condition. Select the LSP group with the lowest preemption cost as the preempted LSP group. By adopting the present invention, the preemption cost during scheduling can be minimized, and resources can be rescheduled for preempted LSPs, thereby effectively realizing high efficiency and reliability of link aggregation and ensuring the quality requirements of each LSP.

Description

A method and device for scheduling aggregated links in a packet switching network

technical field

The invention relates to the technical field of packet switching networks, in particular to a scheduling method and device for aggregation links in a packet switching network.

Background technique

With the growth of data traffic and the improvement of service quality requirements, high bandwidth provision has increasingly become the most important feature of packet switching networks. Link aggregation is an important technology to meet the high bandwidth requirements of the network.

Link aggregation technology (described in IEEE 802.3ad) is to combine several physical links between two devices into a logical link (also called an aggregated link), which is logically As a whole, the internal composition and details of the transmitted data are transparent to the upper layer services.

In the prior art, in order to allocate resources to newly built or failed LSPs (Label Switched Paths) when they recover, the resources are preempted according to the priority, so that resources can be obtained when the newly built or failed LSPs are restored. There are two defects in doing so: first, the resources with low priority are preempted when they are found, and the cost of being preempted is not weighed; second, the preempted LSP(s) are not properly processed.

In order to effectively utilize the aggregated link bandwidth and avoid data loss due to service interruption as much as possible, the Chinese patent CN200810025657.8 "A Method for Realizing Link Preemption and Link Ports" proposes the following scheme: When the link with high priority is about to preempt the link with low priority, use the link with high priority and high priority to be in the delayed preemption state; The link with higher priority is placed on the preempted link to preempt the link with lower priority. The advantage of this scheme is that it can avoid the loss of service data flow when the high-priority link on the aggregated link is preempted, but it does not consider the cost of preempting the low-priority link and whether It can re-preempt or obtain the resource situation, but only consider how the high priority preempts the low priority link without data loss, so it will still cause a large cost and loss of user data on other preempted links.

In the packet switching network, technologies such as MPLS, MPLS-TP, PBB, and VLAN switching have been used to realize TE traffic engineering. According to the current TE traffic engineering technology, the control plane and the data plane are separated. Before transmitting user data, the service provider first formulates an SLA (Service Level Agreement) according to the user's requirements. The SLA includes different services that meet the user's QoS (Quality of Service) requirements. The bandwidth parameters of the flow, and the control plane configures the forwarding path of the service flow to the data plane according to the bandwidth parameters of different service flows determined in advance and the current network resource usage status, and when the resource usage status of the network changes (such as LSP deletion or increase or decrease of network links, etc.), further modify the forwarding path of the existing service flow and notify the data plane, when the data plane receives the service flow data packet, it will forward the data according to the forwarding path configured by the control plane, thus ensuring The QoS requirements of user service flows can also optimize the use of network resources, so TE traffic engineering technology is currently more and more widely used in operator networks.

Since service providers need to charge different service levels and different bandwidths for different users, how to ensure high-priority service quality is also one of the important characteristics of TE traffic engineering technology. Furthermore, if the control plane can make full use of this feature, it can ensure that high-priority services can get high-quality services at a relatively low cost, realize the link aggregation function, and solve the problems of low-priority services that cannot obtain resources and The shortcoming that the bandwidth is not fully utilized is also very necessary in the packet switching network.

Contents of the invention

The technical problem solved by the present invention is to provide a method and device for scheduling aggregated links in a packet-switching network, which can minimize the preemption cost during scheduling, and try to reschedule resources for preempted LSPs, thereby effectively realizing link aggregation and high efficiency. Reliability, ensuring the quality requirements of each LSP.

In order to solve the above-mentioned technical problems, the present invention provides a method for scheduling aggregated links in a packet-switched network,

The label switched path (LSP) of the resource currently to be preempted judges whether each member link satisfies the preemption condition one by one, and selects the least preemption cost from one or more member links satisfying the preemption condition as the preempted LSP group.

Further, the method also includes:

After selecting the preempted LSP group, adding the LSPs in the preempted LSP group to the LSP queue of resources to be preempted;

Wherein, the preempted LSP group includes one or more LSPs.

Further, the preemption condition includes: the available bandwidth of the member link is greater than or equal to the committed bandwidth of the LSP of the current resource to be preempted.

Further, the available bandwidth of the member link includes: a sum of idle bandwidth of the member link and unreserved bandwidth of available LSPs in the member link.

Further, when judging whether each member link meets the preemption condition, it specifically includes:

If the lowest priority LSP of the member link is equal to or higher than the establishment priority of the LSP of the current resource to be preempted, then determining that the member link does not meet the preemption condition;

If the LSP with the lowest priority of the member link is lower than the establishment priority of the LSP of the current resource to be preempted, then sequentially transfer the LSP with the lowest priority in the member link to the LSP with a priority lower than the current resource to be preempted. The unreserved bandwidth of the LSP that establishes the priority of the LSP is made available until the sum of the total unreserved bandwidth made available and the idle bandwidth of the member link is greater than or equal to the commitment of the LSP currently to be preempted bandwidth, it is determined that the member link meets the preemption condition; if the sum of the total unreserved bandwidth set as available in the member link and the idle bandwidth of the member link is still less than the current resource to be preempted The committed bandwidth of the LSP, it is determined that the member link does not meet the preemption condition.

Further, the LSP group with the smallest preemption cost refers to: the number of preempted LSPs is the smallest; or, the priority of the preempted LSPs is low; or, the preemption time is the shortest.

The present invention also provides a scheduling device for aggregated links in a packet switching network, the device includes: a resource query module, and a scheduling processing module, wherein

The resource query module is used to query and judge whether each member link satisfies the preemption condition of the LSP of the resource to be preempted, and return one or more member links that meet the preemption condition to the scheduling processing module;

The scheduling processing module is used to select the one with the smallest preemption cost from one or more member links satisfying the preemption condition as the preempted LSP group.

Further, the scheduling processing module is further configured to, after selecting the preempted LSP group, add the LSPs in the preempted LSP group to the LSP queue of resources to be preempted;

Wherein, the preempted LSP group includes one or more LSPs.

Further, the resource query module is configured to, when judging whether each member link satisfies the preemption condition, specifically include:

If the lowest priority LSP of the member link is equal to or higher than the establishment priority of the LSP of the current resource to be preempted, then determining that the member link does not meet the preemption condition;

If the LSP with the lowest priority of the member link is lower than the establishment priority of the LSP of the current resource to be preempted, then sequentially transfer the LSP with the lowest priority in the member link to the LSP with a priority lower than the current resource to be preempted. The unreserved bandwidth of the LSP that establishes the priority of the LSP is made available until the sum of the total unreserved bandwidth made available and the idle bandwidth of the member link is greater than or equal to the commitment of the LSP currently to be preempted bandwidth, then it is determined that the member link satisfies the preemption condition, and returns the member link and/or the unreserved bandwidth in the member link to the available LSP group to the scheduling processing module; if the member link If the sum of the total unreserved bandwidth set as available and the idle bandwidth of the member link is still less than the committed bandwidth of the LSP currently to be preempted, it is determined that the member link does not meet the preemption condition.

Further, the group of preempted LSPs with the smallest preemption cost selected by the scheduling processing module includes: the number of preempted LSPs is the smallest; the priority of the preempted LSPs is low; or the preemption time is the shortest.

In view of the fact that the existing link aggregation FF (Fixed Filtering) resource reservation scheduling method cannot effectively solve the disadvantages of less cost and preempted LSPs not being processed well, the present invention proposes a new scheduling scheme, according to each According to the LSP situation of each member link, find a scheduling method with the least preemption cost, and try to reschedule resources for the preempted LSP, so as to effectively realize the high efficiency and reliability of link aggregation, and ensure the quality requirements of each LSP.

Description of drawings

The accompanying drawings described here are used to provide a further understanding of the present invention and constitute a part of the application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations to the present invention. In the attached picture:

Fig. 1 is a network schematic diagram of member links and LSPs in an aggregated link;

FIG. 2 is a flowchart of a link aggregation scheduling method in the process of creating an LSP according to an embodiment of the present invention.

Detailed ways

This implementation mode provides a method for scheduling aggregated links in a packet-switched network, and adopts the following scheme:

The LSP currently to preempt resources judges whether each member link meets the preemption condition one by one, and selects the least preemption cost from one or more member links that meet the preemption condition as the preempted LSP group.

Wherein, the preempted LSP group includes one or more LSP(s), further, the method further includes: after selecting the preempted one or more LSP(s), adding the preempted The preempted LSP(s) are all added to the LSP list of resources to be preempted.

Further, the preemption condition includes: the available bandwidth of the member link is greater than or equal to the committed bandwidth of the LSP of the current resource to be preempted.

Further, the available bandwidth is specifically: a sum of idle bandwidth of the member link and unreserved bandwidth of available LSPs in the member link.

Further, when judging whether each member link meets the preemption condition, it specifically includes:

If the lowest priority LSP of the member link is equal to or higher than the establishment priority of the LSP of the current resource to be preempted, then determining that the member link does not meet the preemption condition;

If the lowest priority LSP of the member link is lower than the establishment priority of the LSP of the current to-be-preempted resource, sequentially transfer the lowest-priority LSP in the member link to a priority lower than the current to-be-preempted resource. Set the unreserved bandwidth of the LSP with the priority of the LSP of the resource to be available until the sum of the total unreserved bandwidth and the idle bandwidth of the member link is greater than or equal to the LSP of the current resource to be preempted The committed bandwidth of the member link is determined to meet the preemption condition; if all the total unreserved bandwidth set as available in the member link (that is, the establishment of the LSP whose priority is lower than the current resource to be preempted) takes precedence If the sum of the total unreserved bandwidth of LSPs at the same level) and the idle bandwidth of the member link is still less than the committed bandwidth of the LSP currently to be preempted, it is determined that the member link does not meet the preemption condition.

Further, the preemption cost is the smallest, including but not limited to: the number of preempted LSPs is the smallest; or, when the number of preempted LSPs is the smallest, the preempted bandwidth is the smallest; the preempted LSP has the lowest priority; or , the preemptive resource utilization rate is the highest, as far as possible to ensure that the preempted LSPs can obtain resources; or, the preemption takes the least time and finds the resource fastest, and so on.

As shown in Figure 2, the scheduling method of the aggregated links in the packet switching network provided by this embodiment, the specific steps are as follows:

Step 1. Query idle resources according to the load balancing strategy. If the query returns success, go to step N; if the query returns failure, go to step 2.

Step 2: Set the bandwidth occupied by the LSP with the lowest priority as "available", and then judge whether the available bandwidth of each member link satisfies the condition.

If there is a member link that meets the requirement, go to step N.

If multiple member links meet the requirements, select a member link that satisfies the least number of preempted LSP(s), and enter step N.

If no member link meets the requirements, go to step 3.

Step 3, set the bandwidth occupied by the LSP with the next lowest priority as "available", and the calculation process is the same as step 2.

Step N-1, temporarily set the bandwidth occupied by the LSP with the second highest priority (that is, equal to or higher than the LSP priority previously set as available, but lower than the established priority) as "available", and the calculation process is the same as Step 2, if there is still no member link meeting the condition at this time, the entire preemptive scheduling process fails.

In step N, the preemptive scheduling is successful, and the member links and the preempted LSP(s) are returned. If there is an LSP(s) to be preempted, it is added to an LSP list of resources to be preempted (also called a list of LSPs to be processed) and waits for processing.

In order to make the purpose, technical solution and advantages of the present invention more clear, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined arbitrarily with each other.

In all the implementation details below, there are the following agreed description methods:

1) Linka-b represents a member link of an aggregated link (also called a logical link) Linka, which is also a physical link;

2) BW1 in Linka-b (BW1, BW2) indicates the available bandwidth of the physical link, and BW2 indicates the unreserved available bandwidth of the link, also called unreserved bandwidth.

3) MaxLspBW in Linka(MaxLspBW) indicates the unreserved maximum LSP bandwidth of the logical link Linka. In order to avoid misordering in the transmission of user data packets, it is stipulated in the present invention that one LSP is not allowed to be transmitted across multiple member links at the same time, so MaxLspBW should be the maximum value among the unreserved bandwidths of all member links. If the committed bandwidth of an LSP is BW0, the LSP can only pass through the logical link when BW0<=MaxLspBW. The logical link Linka also has other bandwidth parameters that can be described, such as "maximum link bandwidth (bandwidth that can be used), maximum reservable bandwidth, unreserved bandwidth (sum of unreserved bandwidth on each member link)", etc. .

4) LSP (Prior, CIR, link ID) represents some key characteristic value information of an LSP, wherein: Prior represents the establishment/maintenance priority level (establishment priority during establishment, and maintenance priority after establishment); CIR represents committed bandwidth ; Link ID indicates the ID of the member link of the LSP passing through the aggregated link.

Embodiment one

This implementation example describes the implementation process of the link aggregation scheduling method "when a unidirectional LSP is created passing through an aggregated link".

Assume that in this embodiment, the current main status information of the aggregated link consists of the following three sets of values:

1) Link1-1 (100M, 0M), Link1-2 (100M, 0M);

2) Link1(0M);

3) LSP1 (1, 20M, Link1-1), LSP2 (3, 20M, Link1-1), LSP3 (5, 30M, Link1-1), LSP4 (5, 30M, Link1-1), LSP5 (1, 30M, Link1-2), LSP6(3, 30M, Link1-2), LSP7(5, 10M, Link1-2), LSP8(5, 10M, Link1-2), LSP9(5, 20M, Link1-2) .

Now LSP10 is established in the network through distributed RSVP-TE signaling (its establishment priority level is 4, and its CIR is 30M). After routing calculation, LSP10 needs to go through Link1. When the order reaches the aggregation link endpoint, as shown in Figure 2, the flow chart of the steps of implementing the link aggregation control technology function in the LSP creation process is as follows:

Step 1, LSP10 (4, 30M, Link1) establishes signaling to reach node A, and according to the signaling request information, it is determined that the LSP needs to pass through the link Link1 (also the local egress link of the LSP).

Link1 (0M) is found in the query, and LSP10 requires a CIR of 30M, and there are no idle resources, so it needs to preempt and go to step 2.

Step 2, for the two LSPs with the lowest priority (keeping priority 5) in Link1-1: LSP3 (5, 30M, Link1-1), LSP4 (5, 30M, Link1-1), first set the LSP occupied by LSP4 The bandwidth is temporarily set to "Available", then there is 30M available bandwidth in the Link1-1 member link, which meets the conditions that the LSP10 establishment priority is 4 and the CIR is 30M; then check whether other member links meet the preemption conditions;

For the three LSPs with the lowest priority (keeping priority 5) in Link1-2, LSP7 (5, 10M, Link1-2), LSP8 (5, 10M, Link1-2), LSP9 (5, 20M, Link1-2 ) First temporarily set the bandwidth of LSP9 to "available", then there is 20M available bandwidth in the Link1-1 member link, which does not meet the conditions that the LSP10 establishment priority is 4 and the CIR is 30M; continue to temporarily set the bandwidth of LSP8 to "Available", then there is 30M available bandwidth in the Link1-2 member link, which meets the preemption condition of LSP10;

Comparing the costs spent by the member links Link1-1 and Link1-2, the number of LSPs seized by Link1-1 is the least, so go to step N.

Step N, the preemptive scheduling is successful, and the member links and the preempted LSP(s) are returned:

Link1-1(LSP1(1, 20M, Link1-1), LSP2(3, 20M, Link1-1), LSP10(4, 30M, Link1-1), LSP3(5, 30M, Link1-1));

The preempted LSP4 (5, 30M, --) is added to the LSP list to be processed and waits for processing.

Embodiment two

This implementation example describes the implementation process of LSP rescheduling when the link aggregation is preempted when a unidirectional LSP is created and passes through an aggregated link.

Assume that the current main status information of the aggregated link in this embodiment consists of the following three sets of values:

1) Link1-1 (50M, 0M), Link1-2 (50M, 10M);

2) Link1 (10M);

3) LSP1 (3, 10M, Link1-1), LSP2 (4, 20M, Link1-1), LSP3 (4, 10M, Link1-1), LSP4 (5, 10M, Link1-1), LSP5 (2, 20M, Link1-2), LSP6 (3, 10M, Link1-2), LSP7 (6, 10M, Link1-2).

Now LSP8 is established in the network through distributed RSVP-TE signaling (its establishment priority level is 3, and its CIR is 30M). After routing calculation, LSP8 needs to go through Link1. When the order reaches the aggregation link endpoint, as shown in Figure 2, the flow chart of the steps of implementing the link aggregation control technology function in the LSP creation process is as follows:

Step 1, LSP8 (3, 30M, Link1) establishes signaling to reach node A, and according to the signaling request information, it is determined that the LSP needs to pass through the link Link1 (which is also the local egress link of the LSP).

Link1 (10M) is queried, but LSP8 requires a CIR of 30M, and there are no idle resources, so it needs to preempt and go to step 2.

Step 2, for Link1-1, an LSP with the lowest priority (keep priority 5): LSP4(5, 10M, Link1-1), first temporarily set the bandwidth occupied by LSP4 to "available", then Link1-1 There is 10M available bandwidth in the member link, which does not meet the requirements of LSP8;

For an LSP with an idle bandwidth of 10M in Link1-2, which does not meet the establishment conditions of LSP8, and has the lowest priority (maintaining the priority as 6): LSP7 (6, 10M, Link1-2), temporarily set the bandwidth occupied by LSP7 to "Available", the available bandwidth at this time is: 10M idle bandwidth plus 10M unreserved bandwidth of LSP7, a total of 20M, which still does not meet the conditions of LSP8, and the lowest maintenance priority of the remaining LSPs is 3, which is the same as the establishment priority of LSP8 , therefore, the member link Link1-2 does not meet the preemption condition of LSP8.

Step 3, for Link1-1 continue an LSP with the lowest priority (keep the priority as 4): temporarily set the bandwidth of LSP2 occupied by LSP2 (4, 20M, Link1-1) to "available", then Link1-1 There is a total of 30M available bandwidth in the member links, which meets the conditions of LSP8; in addition, there is another preemption method, that is, to preempt LSP3 (4, 10M, Link1-1), LSP2 (4, 20M, Link1-1), select The one with the least cost is the first one, enter step N.

Step N, the preemptive scheduling is successful, and the member links and the preempted LSP(s) are returned:

Link1-1(LSP1(3, 10M, Link1-1), LSP3(4, 10M, Link1-1), LSP8(3, 30M, Link1-1))

The preempted LSP2 (4, 20M, --) and LSP4 (5, 10M, --) are added to the pending list and wait for processing.

Thereafter, the LSPs in the pending list enter step 1 again. The specific treatment is as follows:

Step 1: The LSP2 waiting to be processed has an establishment priority of 4 and a CIR of 20M. If Link1 (10M) is queried, there is not enough idle resources, and it needs to be preempted to proceed to Step 2.

Step 2, for an LSP with an available bandwidth of 10M in Link1-2, which does not meet the establishment conditions of LSP2, and has the lowest priority (maintaining the priority as 6): LSP7 (6, 10M, Link1-2), use the bandwidth occupied by LSP7 Temporarily set it to "Available", which meets the conditions of LSP2, so the member link Link1-2 meets the conditions of LSP8.

Step N, the preemptive scheduling is successful, and the member links and the preempted LSP(s) are returned:

Link1-2(LSP7(2, 20M, Link1-2), LSP6(3, 10M, Link1-2), LSP2(4, 20M, Link1-2));

The preempted LSP7 (6, 20M, --) and LSP4 (5, 10M, --) are added to the pending list and wait for processing.

It should be noted that the above are feasible specific examples described in the present invention for preempting the minimum number of LSPs and the rescheduling range of the preempted LSPs. The examples do not limit the protection scope of the patent of the present invention, and the technology of the present invention can be applied to minimum cost (including Preempting the minimum bandwidth under the premise of preempting the minimum number of LSPs, preempting priority and preempting the weight adjustment of the number of LSPs) and any specific technical scope such as rescheduling.

In addition, the above-mentioned embodiment is for the preemption processing of the aggregated link FF mode at the control plane level, wherein the trigger condition for controlling the control plane of link resource scheduling may be a signaling request, a network management configuration request, and the like. As for the underlying implementation process of link aggregation link bandwidth link adjustment, and the SE (shared filtering) preemption method, etc., no detailed description will be given. And the above description is only a detailed description of the method of least changing the number of LSPs based on the priority. However, other cost calculations should also be within the protection of the present invention.

It can be seen from the above scheme that the present invention can find a member link with the least number of preempted LSPs according to the actual situation of different member links, and try to reschedule resources for the preempted LSP, thereby effectively realizing efficient and reliable link aggregation It can guarantee the quality requirements of each LSP, which is very useful in packet switching networks with traffic engineering requirements.

In addition, an embodiment of the present invention also provides a scheduling device for aggregated links in a packet switching network, the device mainly includes: a resource query module, and a scheduling processing module, wherein

The resource query module is used to query and judge whether each member link satisfies the preemption condition of the LSP of the resource to be preempted, and return one or more member links that meet the preemption condition to the scheduling processing module;

The scheduling processing module is used to select the one with the smallest preemption cost from one or more member links satisfying the preemption condition as the preempted LSP group.

Further, the scheduling processing module is further configured to, after selecting the preempted LSP group, add the LSPs in the preempted LSP group to the LSP queue of resources to be preempted;

Wherein, the preempted LSP group includes one or more LSPs.

Further, the resource query module is configured to, when judging whether each member link satisfies the preemption condition, specifically include:

If the lowest priority LSP of the member link is equal to or higher than the establishment priority of the LSP of the current resource to be preempted, then determining that the member link does not meet the preemption condition;

If the LSP with the lowest priority of the member link is lower than the establishment priority of the LSP of the current resource to be preempted, then sequentially transfer the LSP with the lowest priority in the member link to the LSP with a priority lower than the current resource to be preempted. The unreserved bandwidth of the LSP that establishes the priority of the LSP is made available until the sum of the total unreserved bandwidth made available and the idle bandwidth of the member link is greater than or equal to the commitment of the LSP currently to be preempted bandwidth, then it is determined that the member link satisfies the preemption condition, and returns the member link and/or the unreserved bandwidth in the member link to the available LSP group to the scheduling processing module; if the member link If the sum of the total unreserved bandwidth set as available and the idle bandwidth of the member link is still less than the committed bandwidth of the LSP currently to be preempted, it is determined that the member link does not meet the preemption condition.

Further, the group of preempted LSPs with the smallest preemption cost selected by the scheduling processing module includes: the number of preempted LSPs is the smallest; the priority of the preempted LSPs is low; or the preemption time is the shortest.

The above are only preferred implementation examples of the present invention, and are not intended to limit the present invention. The present invention can also have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can Various corresponding changes and modifications are made in the present invention, but these corresponding changes and modifications should all belong to the protection scope of the appended claims of the present invention.

Obviously, those skilled in the art should understand that each module or each step of the above-mentioned present invention can be realized by a general-purpose computing device, and they can be concentrated on a single computing device, or distributed in a network formed by multiple computing devices Alternatively, they may be implemented in program code executable by a computing device so that they may be stored in a storage device to be executed by a computing device, and in some cases in an order different from that shown here The steps shown or described are carried out, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps among them are fabricated into a single integrated circuit module for implementation. As such, the present invention is not limited to any specific combination of hardware and software.

Claims (10)

1. the dispatching method of aggregated links in a kind of packet network, which is characterized in that

Currently wait for that the label switched path (LSP) of preempting resources judges whether each Member Link meets preemptive condition one by one, and Selected from the one or more Member Links for meeting preemptive condition preemption cost it is minimum as the LSP groups being preempted；

It when there is the LSP being preempted, is added into the LSP lists for waiting for preempting resources, waits pending；

Wherein, judge whether each Member Link meets preemptive condition and include：If the lowest priority of LSP is low in the Member Link Priority is established in the current LSP for waiting for preempting resources, then priority in the Member Link is less than successively and is currently waited for The not reserved bandwidth of the LSP for establishing priority of the LSP of preempting resources is set to available, and available total band is not reserved until being set to The wide summation with the idle bandwidth of the Member Link is more than or equal to the guaranteed bandwidth of the current LSP for waiting for preempting resources, then Judge that the Member Link meets preemptive condition.

2. the method as described in claim 1, which is characterized in that the method further includes：

After selecting the LSP groups being preempted, the LSP in the LSP groups being preempted is added to the LSP teams for waiting for preempting resources In row；

Wherein, the LSP groups being preempted include one or more LSP.

3. the method as described in claim 1, which is characterized in that

The preemptive condition includes：The available bandwidth of the Member Link is more than or equal to holding for the current LSP for waiting for preempting resources Promise bandwidth.

4. method as claimed in claim 3, which is characterized in that

The available bandwidth of the Member Link, including：The idle bandwidth of the Member Link with it is available in the Member Link The summation of the not reserved bandwidth of LSP.

5. method as described in claim 3 or 4, which is characterized in that judging whether each Member Link meets preemptive condition When, it specifically includes：

If the LSP of the lowest priority of the Member Link is excellent at or above establishing for the current LSP for waiting for preempting resources First grade then judges that the Member Link is unsatisfactory for preemptive condition；

If all in the Member Link be set to available total not reserved bandwidth and the idle bandwidth of the Member Link Summation then judges that the Member Link is unsatisfactory for preemptive condition still less than the guaranteed bandwidth of the current LSP for waiting for preempting resources.

6. such as any one of them method of claim 1,2,3,4, which is characterized in that

The LSP groups of the preemption cost minimum refer to：The number for the LSP being preempted is minimum；Alternatively, the LSP's being preempted is preferential Grade is low；Alternatively, it is most short to seize the time.

7. the dispatching device of aggregated links in a kind of packet network, which is characterized in that described device includes：Resource query mould Block and schedule process module, wherein

The resource query module is used for, and is inquired and is judged whether each Member Link meets seizing for the LSP for waiting for preempting resources Condition, and the one or more Member Links for meeting preemptive condition are returned into the schedule process module；

The schedule process module is used for, and selects preemption cost minimum from the one or more Member Links for meet preemptive condition As the LSP groups being preempted；

The dispatching device is additionally operable to when there is the LSP being preempted, and is added into the LSP lists for waiting for preempting resources, at waiting Reason；

Wherein, the resource query module judges whether each Member Link meets preemptive condition and include：If in the Member Link The lowest priority of LSP establishes priority less than the current LSP's for waiting for preempting resources, then successively will be in the Member Link Priority be less than the LSP for establishing priority for the LSP for currently waiting for preempting resources not reserved bandwidth be set to it is available, until being set to can Total not reserved bandwidth and the summation of the idle bandwidth of the Member Link currently waits for preempting resources more than or equal to described The guaranteed bandwidth of LSP then judges that the Member Link meets preemptive condition.

8. device as claimed in claim 7, which is characterized in that

The schedule process module is additionally operable to, will be in the LSP groups that be preempted after selecting the LSP groups being preempted LSP is added in the LSP queues for waiting for preempting resources；

Wherein, the LSP groups being preempted include one or more LSP.

9. device as claimed in claim 7, which is characterized in that

The resource query module is used for, and when judging whether each Member Link meets preemptive condition, is specifically included：

If the LSP of the lowest priority of the Member Link is excellent at or above establishing for the current LSP for waiting for preempting resources First grade then judges that the Member Link is unsatisfactory for preemptive condition；

If all in the Member Link be set to available total not reserved bandwidth and the idle bandwidth of the Member Link Summation then judges that the Member Link is unsatisfactory for preemptive condition still less than the guaranteed bandwidth of the current LSP for waiting for preempting resources.

10. device as claimed in claim 9, which is characterized in that

The LSP groups of the preemption cost minimum of the schedule process module selection being preempted, including：The number for the LSP being preempted At least；The priority for the LSP being preempted is low；Alternatively, it is most short to seize the time.