CN108337189B - Bandwidth scheduling method and device - Google Patents

Bandwidth scheduling method and device Download PDF

Info

Publication number
CN108337189B
CN108337189B CN201810082837.3A CN201810082837A CN108337189B CN 108337189 B CN108337189 B CN 108337189B CN 201810082837 A CN201810082837 A CN 201810082837A CN 108337189 B CN108337189 B CN 108337189B
Authority
CN
China
Prior art keywords
path
bandwidth
link
occupied
standby
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810082837.3A
Other languages
Chinese (zh)
Other versions
CN108337189A (en
Inventor
李亮
王攀
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
New H3C Technologies Co Ltd
Original Assignee
New H3C Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by New H3C Technologies Co Ltd filed Critical New H3C Technologies Co Ltd
Priority to CN201810082837.3A priority Critical patent/CN108337189B/en
Publication of CN108337189A publication Critical patent/CN108337189A/en
Application granted granted Critical
Publication of CN108337189B publication Critical patent/CN108337189B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/52—Queue scheduling by attributing bandwidth to queues
    • H04L47/525—Queue scheduling by attributing bandwidth to queues by redistribution of residual bandwidth
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/52—Queue scheduling by attributing bandwidth to queues
    • H04L47/522—Dynamic queue service slot or variable bandwidth allocation

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention relates to a bandwidth scheduling method and device. The method comprises the following steps: determining at least one link constituting a first path when the first path is loaded in the network; acquiring the actual available bandwidth of each link in at least one link occupied by the first path; identifying whether the bandwidth actually occupied by the path meets the bandwidth occupied by the first path; and if the actual available bandwidth of the path does not meet the requirement occupied bandwidth, allocating the bandwidth for the first path from the residual available bandwidth of each link in at least one link according to the path attribute of the first path. According to the bandwidth scheduling method and device, bandwidth can be allocated to the path according to the path attribute of the loaded path, the bandwidth occupied by the standby path can be reduced dynamically, and the utilization rate of the bandwidth is improved under the condition that the disaster recovery effect is guaranteed as much as possible, so that a network can bear more data stream services.

Description

Bandwidth scheduling method and device
Technical Field
The present invention relates to the field of network communication technologies, and in particular, to a bandwidth scheduling method and apparatus.
Background
The service flow refers to the flow in the network including identification attributes such as a source node, a source interface, a destination node, a destination interface and the like and some custom attributes. For example: traffic carrying voice conference traffic and traffic carrying video communication traffic can be considered as one traffic stream. Data flow refers to abstracting a service flow in a Network into a data flow inside a Software Defined Network (SDN) controller.
In general, in order to ensure that a service is not interrupted when a path of a data stream fails (which may also be referred to as traffic non-interruption), a plurality of paths need to be simultaneously selected for the data stream. For example, to achieve the disaster recovery effect, two paths are found for the data stream in the path calculation process: one path is a main path, and the other path is a standby path. Under normal conditions, the main path carries the traffic of the data stream, the standby path only occupies the bandwidth (or pre-occupies the bandwidth) but does not pass the traffic, and when the main path fails, the traffic of the data stream is switched to the standby path quickly. Therefore, the service is not interrupted when the path fails.
However, the bandwidth occupied by the backup path may reduce the link bandwidth utilization, so that the service traffic that the network can carry is reduced. For example, the SDN controller obtains a data flow attribute and an occupied bandwidth value, and calculates a primary path and a secondary path for the data flow, where the primary path and the secondary path all occupy the same bandwidth, that is, the secondary path occupies 100% of the bandwidth. The method is simple to operate, when the paths are calculated, the processing modes of the main path and the standby path are the same, the disaster recovery effect is good, however, in practical application, the probability of the failure of the main path is extremely low, if the standby path occupies 100% of the bandwidth, the network bandwidth is wasted, and the data stream which can be carried by the network is reduced.
If the backup path does not occupy the bandwidth, the service disaster recovery effect cannot be guaranteed when the main path fails. For example, the bandwidth occupied by the primary path is a bandwidth value acquired by the SDN controller, and the bandwidth occupied by the secondary path is 0% to 100%, typically 20%, of the bandwidth value acquired by the SDN controller. That is, the primary path fully occupies the bandwidth, and the backup path partially occupies the bandwidth. The method is simple to operate, and reduces the bandwidth waste. However, when the bandwidth occupied by the backup path is small, it is difficult to ensure the disaster recovery effect.
Therefore, it is necessary to implement a method for improving the utilization rate of the bandwidth under the condition of ensuring the disaster recovery effect as much as possible, so that the network can carry more data stream services.
Disclosure of Invention
In view of this, the present invention provides a bandwidth scheduling method and apparatus, which allocate bandwidth to a path according to a path attribute of the loaded path, so as to dynamically reduce bandwidth occupied by a backup path, and improve the utilization rate of the bandwidth under the condition of ensuring the disaster recovery effect as much as possible, so that a network can bear more data stream services.
According to an aspect of the present invention, there is provided a bandwidth scheduling method, the method including:
determining at least one link constituting a first path when the first path is loaded in a network;
acquiring the actual path occupied bandwidth of each link in the at least one link occupied by the first path, wherein the actual path occupied bandwidth is the minimum value of the residual allocable bandwidth of each link in the at least one link;
identifying whether the actual occupiable bandwidth of the path meets the required occupied bandwidth of the first path;
if the path actual available bandwidth meets the required available bandwidth, allocating the first path actual available bandwidth from the remaining allocable bandwidth of each link in the at least one link;
and if the actual available bandwidth of the path does not meet the bandwidth occupied by the demand, allocating bandwidth for the first path from the residual available bandwidth of each link in the at least one link according to the path attribute of the first path.
According to another aspect of the present invention, there is provided a bandwidth scheduling apparatus, the apparatus comprising:
a first determining module for determining at least one link constituting a first path when the first path is loaded in a network;
an obtaining module, configured to obtain an actual bandwidth that the first path occupies for each link in the at least one link, where the actual bandwidth that the path occupies is a minimum value of remaining allocable bandwidths of each link in the at least one link;
the identification module is used for identifying whether the actual occupied bandwidth of the path meets the bandwidth occupied by the requirement of the first path;
a first allocating module, configured to allocate, if the path actual available bandwidth meets the required available bandwidth, the actual available bandwidth of the first path for the first path from the remaining available bandwidth of each link in the at least one link;
and a second allocating module, configured to allocate bandwidth to the first path from the remaining allocable bandwidth of each link in the at least one link according to the path attribute of the first path if the bandwidth actually available for the path does not meet the bandwidth occupied by the demand.
According to the bandwidth scheduling method and device in the embodiment, when the path of the path can actually occupy the bandwidth and cannot occupy the bandwidth according to the requirement of the path, the bandwidth is allocated to the path according to the loaded path attribute of the path, the bandwidth occupied by the standby path can be dynamically reduced, and the utilization rate of the bandwidth is improved under the condition that the disaster recovery effect is guaranteed as much as possible, so that the network can bear more data stream services.
Drawings
Fig. 1 shows a flow chart of a bandwidth scheduling method according to an embodiment of the invention.
Fig. 2 shows a flow chart of a bandwidth scheduling method according to an embodiment of the invention.
Fig. 3 shows a flow chart of a bandwidth scheduling method according to an embodiment of the invention.
Fig. 4 shows a schematic diagram of step S1052 according to an example of the present invention.
Fig. 5 shows a flowchart of a bandwidth scheduling method according to an embodiment of the present invention.
Fig. 6 shows a schematic diagram of step S1052 and step S1053 according to an example of the present invention.
Fig. 7 shows a flowchart of a bandwidth scheduling method according to an embodiment of the present invention.
Fig. 8 shows a flowchart of a bandwidth scheduling method according to an embodiment of the present invention.
Fig. 9 shows a block diagram of a bandwidth scheduling apparatus according to an embodiment of the present invention.
Fig. 10 shows a block diagram of a bandwidth scheduling apparatus according to an embodiment of the present invention.
Fig. 11 shows a block diagram of a bandwidth scheduling apparatus according to an embodiment of the present invention.
Detailed Description
Various exemplary embodiments, features and aspects of the present invention will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers can indicate functionally identical or similar elements. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present invention. It will be understood by those skilled in the art that the present invention may be practiced without some of these specific details. In some instances, methods, procedures, components, and circuits that are well known to those skilled in the art have not been described in detail so as not to obscure the present invention.
In order to solve the foregoing technical problem, an embodiment of the present invention provides a bandwidth scheduling method, which allocates bandwidth to a path according to a path attribute of a loaded path, and dynamically adjusts bandwidth occupied by a backup path, so as to improve a utilization rate of the bandwidth and enable a network to carry more data stream services while ensuring a disaster recovery effect as much as possible.
Fig. 1 shows a flow chart of a bandwidth scheduling method according to an embodiment of the invention. As shown in fig. 1, the method is applied to an SDN controller, and includes the following steps:
step S101, when the first path is loaded in the network, at least one link constituting the first path is determined.
When a new service is established in the SDN network, the SDN controller computes a first path for a data flow of the service. In the embodiment of the present invention, the calculating the first path refers to a process of determining nodes for transmitting a data stream, a node order, and configuring links between the nodes into the first path.
And the SDN controller loads the calculated first path into the SDN network and allocates bandwidth for the first path. At this point, the SDN controller determines at least one link that makes up the first path.
The first path may refer to a path corresponding to a data flow of any new service supported by the current network, which is not limited in the embodiment of the present invention.
In the embodiment of the present invention, the bandwidth of a path refers to the link bandwidth of a link constituting the path, and the bandwidth occupied by the path on each link is equal inside the SDN controller.
Step S102, acquiring the actual occupied bandwidth of the first path occupying each link in the at least one link, wherein the actual occupied bandwidth of the path is the minimum value of the residual allocable bandwidth of each link in the at least one link.
The actual bandwidth occupied by the path refers to the minimum value of the remaining allocable bandwidths of each link in the links formed by the path, wherein the remaining allocable bandwidth of the link is the difference between the total bandwidth of the link and the bandwidth occupied by all the paths formed by the link.
Therefore, when a new path is loaded in the network, in order to determine the bandwidth actually occupied by the path, at least one link constituting the path and the remaining allocable bandwidth of each link of the at least one link are determined.
In step S101, the SDN controller determines at least one link constituting the first path, and then obtains a remaining allocable bandwidth of each link in the at least one link, and then selects a minimum value of the remaining allocable bandwidth of each link as an actual available bandwidth of the path of the first path.
Step S103, identifying whether the bandwidth actually occupied by the path meets the bandwidth occupied by the requirement of the first path.
The required occupied bandwidth of the path is the occupied bandwidth of the path on each link forming the path, which is specified by a user.
After acquiring the actual occupied bandwidth of the path of the first path, the SDN controller judges the relationship between the actual occupied bandwidth of the path of the first path and the required occupied bandwidth: if the actual available bandwidth of the path of the first path can meet the bandwidth occupied by the demand of the first path, the remaining allocable bandwidth of each link in at least one link forming the first path can meet the bandwidth occupied by the demand of the first path; if the actual occupied bandwidth of the path of the first path cannot meet the occupied bandwidth required by the first path, the remaining allocable bandwidth of at least one link composing the first path cannot meet the occupied bandwidth required by the first path.
Step S104, if the path actual available bandwidth meets the requirement occupied bandwidth, allocating the actual available bandwidth of the first path for the first path from the residual allocable bandwidth of each link in at least one link.
As above, if the actual occupiable bandwidth of the path of the first path can satisfy the required occupiable bandwidth of the first path, which indicates that the remaining allocable bandwidth of each of at least one link constituting the first path can satisfy the required occupiable bandwidth of the first path, the SDN controller allocates the actual occupiable bandwidth of the first path for the first path from the remaining allocatable bandwidth of each of at least one link.
Optionally, the SDN controller further marks the bandwidth state of the first path as a first state, that is, as a strict state, that is, the bandwidth allocated to the first path can meet the requirement of the first path to occupy the bandwidth.
Further, the SDN controller updates the remaining allocable bandwidth of each of the at least one link. For example, the SDN controller takes a difference value between a remaining allocable bandwidth before each link in the at least one link is allocated and an actually-occupied bandwidth of a path of the first path as the remaining allocable bandwidth of the updated link.
According to the path attribute of the first path, the SDN controller updates the occupied total bandwidth of the path formed by each link in at least one link. In the embodiment of the present invention, each link in at least one link may be used for a primary path or a backup path constituting a certain path. That is, each link in the at least one link may be a partial link in a primary path or a backup path of a path.
In an example, if the first path is a main path, the SDN controller allocates a sum of a total occupied bandwidth of the main path composed of each link in at least one previous link and an actual occupiable bandwidth of the path of the first path as an updated total occupied bandwidth of the main path composed of each link; and if the first path is the standby path, taking the sum of the total occupied bandwidth of the standby path formed by each link in at least one link before distribution and the actual available bandwidth of the path of the first path as the updated total occupied bandwidth of the standby path formed by each link.
Step S105, if the actual available bandwidth of the path does not meet the required available bandwidth, allocating bandwidth for the first path from the residual available bandwidth of each link in the at least one link according to the path attribute of the first path.
As above, if the actual bandwidth occupied by the path of the first path cannot satisfy the bandwidth occupied by the demand of the first path, it indicates that the remaining allocable bandwidth of at least one of the at least one link constituting the first path cannot satisfy the bandwidth occupied by the demand of the first path, that is, the remaining allocable bandwidth of a part of the at least one link constituting the first path is smaller than the bandwidth occupied by the demand of the first path, and if the bandwidth occupied by the path of the current first path is the bandwidth actually occupied by the path of the first path, the demand of the first path cannot be satisfied.
In this case, the SDN controller determines a path attribute of the first path, allocates bandwidth to the first path according to the path attribute, and the path attribute represents an active/standby attribute of the first path.
If the first path is a standby path, the SDN controller allocates an actual available bandwidth of the first path for the first path from the remaining allocable bandwidth of each link in the at least one link. The SDN controller marks the bandwidth state of the first path as a second state, namely, a marginal state, namely, the bandwidth allocated to the first path cannot meet the requirement of the first path to occupy the bandwidth, and the occupied bandwidth of the standby path is dynamically reduced.
Optionally, the SDN controller further updates the remaining allocable bandwidth of each link in the at least one link, and the total occupied bandwidth of the backup path formed by each link in the at least one link. For example, the SDN controller takes a difference value between a remaining allocable bandwidth before each link in the at least one link is allocated and an actual available bandwidth of a path of the first path as the updated remaining allocable bandwidth. And the SDN controller takes the sum of the total occupied bandwidth of the standby path formed by each link in at least one link before allocation and the actual available bandwidth of the path of the first path as the updated total occupied bandwidth of the standby path formed by each link.
If the first path is a main path, the occupied bandwidth of a part or all of standby paths composed of links, of which the remaining allocable bandwidth cannot meet the bandwidth occupied by the first path, in at least one link composing the first path may be released (the bandwidth of the standby path is dynamically reduced) so that the remaining allocable bandwidth of the link meets the bandwidth occupied by the first path, and then the SDN controller allocates the bandwidth to the first path. The present embodiment is not limited to the specific embodiment. Here, releasing the occupied bandwidth of a path means setting the occupied bandwidth of the path to 0 on a link through which the path passes.
According to the bandwidth scheduling method of the embodiment, when the path of the path can actually occupy the bandwidth which cannot meet the requirement of the path to occupy the bandwidth, the bandwidth is allocated to the path according to the loaded path attribute of the path, so that the bandwidth occupied by the backup path can be dynamically reduced, and the utilization rate of the bandwidth is improved under the condition of ensuring the disaster recovery effect as much as possible, so that the network can bear more data stream services.
Fig. 2 shows a flow chart of a bandwidth scheduling method according to an embodiment of the invention. As shown in fig. 2, step S105 in fig. 1 specifically includes:
step S1051, if the first path is the main path, determining at least one first link, of the at least one link, where the remaining allocable bandwidth does not satisfy the required occupied bandwidth.
In order to meet the bandwidth requirement of the primary path, the SDN controller releases the occupied bandwidth of a standby path composed of first links, wherein the first links are at least one link, and the residual allocable bandwidth does not meet the bandwidth occupied by the first links.
Specifically, the SDN controller may compare a relationship between a remaining allocable bandwidth of each of the at least one link and a required occupied bandwidth of the first path one by one, so as to determine at least one first link, of the at least one link, in which the remaining allocable bandwidth does not satisfy the required occupied bandwidth of the first path.
Step S1052, releasing the standby path formed by each first link in the at least one first link to occupy the bandwidth of each first link, and updating the remaining allocable bandwidth of each first link.
In one example, in particular, for each of the at least one first link: if the first link is used for forming the standby path, the SDN controller releases one standby path in all standby paths formed by the first link to occupy the occupied bandwidth of the first link. And the SDN controller updates the remaining allocable bandwidth of the first link according to the released occupied bandwidth. For example, the SDN controller sets the sum of the remaining allocable bandwidth of the first link, from which the standby path occupied bandwidth is not released, and the released occupied bandwidth as the updated remaining allocable bandwidth of the first link.
If the updated remaining allocable bandwidth of the first link meets the requirement of occupying the bandwidth, or the first link does not have a standby path capable of releasing the bandwidth, the SDN controller stops releasing the standby path composed of the first link from occupying the occupied bandwidth of the first link; if the updated remaining allocable bandwidth of the first link cannot meet the bandwidth occupied by the demand and the first link still has a standby path capable of releasing the bandwidth, the SDN controller continues to release the standby path composed of the first link to occupy the bandwidth occupied by the first link.
Optionally, the SDN controller may also release, at each time, multiple standby paths in all standby paths formed by the first link to occupy the occupied bandwidth of the first link, and a manner of releasing the standby paths to occupy the occupied bandwidth of the first link is not limited in the embodiment of the present invention, and the foregoing manner is only an example of the embodiment of the present invention.
Thus, for each first link in at least one first link, the maximum remaining allocable bandwidth of each first link (i.e., the updated remaining allocable bandwidth of the first link after the standby path composed of the first links is stopped to release the occupied bandwidth of the first link) can be obtained by dynamically reducing the occupied bandwidth of the standby path, so as to meet the bandwidth requirement of the main path.
Step S1053, updating the actual bandwidth that the first path occupies the path of each link in the at least one link according to the updated remaining allocable bandwidth of each first link in the at least one first link.
As described above, for each of the at least one first link, after the SDN controller completes the process of releasing the occupied bandwidth of the standby path in step S1052, the updated remaining allocable bandwidth of each first link may also be obtained. The SDN controller re-determines the minimum value according to the updated remaining allocable bandwidth of each first link, and updates the actual available bandwidth of the path where the first path occupies each link in the at least one link according to the updated remaining allocable bandwidth of each first link, that is, determines the minimum value of the updated remaining allocable bandwidth of each first link as the actual available bandwidth of the path where the first path occupies each link in the at least one link.
Step S1054, allocating an updated path actual occupied bandwidth of the first path to the first path from the updated remaining allocable bandwidth of each link in the at least one link.
After determining the updated path actually-occupiable bandwidth of the first path, the SDN controller allocates the updated path actually-occupiable bandwidth of the first path to the first path from the updated remaining allocable bandwidth of each link of at least one link constituting the first path.
That is, for at least one link that makes up the first path: the SDN controller allocates an updated path actual available bandwidth of the first path for the first path from the updated remaining allocable bandwidth of each first link. The SDN controller allocates an updated path actual available bandwidth of the first path for the first path from the remaining allocable bandwidths of the links other than the first link in the at least one link.
The SDN controller releases the occupied bandwidth of the standby path of the first link of which the residual allocable bandwidth does not meet the requirement of the first path, and updates the residual allocable bandwidth of each first link. And the SDN controller updates the actual available bandwidth of the path of the first path according to the updated residual allocable bandwidth of each first link, and then allocates the updated actual available bandwidth of the path of the first path to the first path. According to the bandwidth scheduling method provided by the embodiment of the invention, the SDN controller can improve the actual occupied bandwidth of the loaded path by dynamically adjusting the occupied bandwidth of the standby path, ensure the transmission of data streams on the main path as much as possible, and improve the utilization rate of the bandwidth under the condition of ensuring the disaster recovery effect as much as possible, so that the network can bear more data stream services.
Fig. 3 shows a flow chart of a bandwidth scheduling method according to an embodiment of the invention. As shown in fig. 3, step S1052 in fig. 2 includes:
step S10521, if the first link has the backup path that can release the occupied bandwidth, sequentially releasing the backup path with the lowest priority from the backup paths formed by the first link to occupy the first occupied bandwidth of the first link.
The existence of the standby path capable of releasing the occupied bandwidth in the first link means that the standby path in the standby path composed of the first link is not released to occupy the bandwidth.
In the embodiment of the present invention, the service quality of the data stream is divided into 8 priorities, and one priority may be assigned to the data stream according to the attribute of the data stream or the designation of the user. In the SDN controller, priorities are allocated to paths of the data flows according to the priorities of the data flows. Wherein, the priority allocated to the main path is 1-8, the priority allocated to the standby path is 9-16, and the smaller the numerical value, the higher the priority.
For example, if the priority of the data stream specified by the user is p, the priority of the main path calculated for the data stream is p, and the priority of the backup path is p + 8.
In the embodiment of the invention, the occupied bandwidth of the standby path is released according to the priority of the standby path composed of the first link, and the SDN controller firstly releases the occupied bandwidth of the standby path with low priority and then releases the occupied bandwidth of the standby path with high priority so as to meet the bandwidth requirement of the data stream with high priority as much as possible.
In one example, the SDN controller orders the standby paths according to priorities of the standby paths composed of the first links, so as to release occupied bandwidth of the standby paths according to the priorities. The sorting order may be from low to high, or from high to low, which is not limited in the embodiments of the present invention.
Step S10522, updating the remaining allocable bandwidth of the first link according to the released first occupied bandwidth each time.
That is to say, after releasing a standby path with the lowest priority from among standby paths composed of the first link to occupy the first occupied bandwidth of the first link each time, the SDN controller updates the remaining allocable bandwidth of the first link according to the first occupied bandwidth. For example, the sum of the remaining allocable bandwidth of the first link without releasing the standby path occupied bandwidth and the first occupied bandwidth is used as the updated remaining allocable bandwidth of the first link.
Step S10523, if the updated remaining allocable bandwidth of the first link meets the requirement occupied bandwidth, or the first link does not have a standby path capable of releasing the bandwidth, stopping releasing the standby path composed of the first link from occupying the first occupied bandwidth of the first link.
As described above, after the remaining allocable bandwidth of the first link is updated each time, the SDN controller determines whether the updated remaining allocable bandwidth of the first link satisfies the bandwidth occupied by the first path. If yes, the SDN controller stops releasing a standby path formed by the first link to occupy the first occupied bandwidth of the first link; if the standby path does not exist, the SDN controller continuously judges whether the first link still has the standby path capable of releasing the bandwidth, if so, the standby path with the lowest priority in the standby paths composed of the first link is continuously released to occupy the first occupied bandwidth of the first link, and if not, the standby path composed of the first link is stopped to release to occupy the first occupied bandwidth of the first link.
Through the examples shown in fig. 3 and fig. 4, the actual available bandwidth of the first path can be updated, the actual available bandwidth of the updated path of the first path is increased, the SDN controller allocates bandwidth to the first path according to the actual available bandwidth of the updated path of the first path, the bandwidth occupied by the backup path is dynamically reduced, and the bandwidth utilization rate can be increased under the condition that the disaster recovery effect is guaranteed as much as possible, so that the network can carry more data flow services.
Fig. 4 shows a schematic diagram of step S1052 according to an example of the present invention, in this embodiment, it is assumed that m first links, L1 and L2 … Lm respectively, of the at least one link, where the remaining allocable bandwidth does not meet the demand occupied bandwidth are determined. For each of the m first links, taking first link L1 as an example, as shown in fig. 4:
the SDN controller judges whether a standby path capable of releasing occupied bandwidth exists in the standby paths formed by the L1. If yes, the SDN controller selects the standby path with the lowest priority from the standby paths formed by the L1, and releases the standby path with the lowest priority to occupy the first occupied bandwidth of the L1. The SDN controller updates the remaining allocable bandwidth of the L1 according to the released first occupied bandwidth; the SDN controller judges whether the updated residual allocable bandwidth of the L1 meets the bandwidth occupied by the requirement of the first path, and if yes, the SDN controller stops releasing the first occupied bandwidth of the L1 occupied by the standby path formed by the L1; if the standby path is not satisfied, continuously judging whether a standby path capable of releasing the bandwidth still exists in the standby paths composed of the L1, if so, continuously releasing the first occupied bandwidth of the L1 occupied by the standby path with the lowest priority in the standby paths composed of the L1 by the SDN controller, and if not, stopping releasing the first occupied bandwidth of the L1 occupied by the standby path composed of the L1.
After the first link L1 completes the above process, that is, after the standby path composed of L1 stops releasing the first occupied bandwidth of L1, the process shown in fig. 4 is performed on the first link L2 … Lm.
Then, the SDN controller determines the minimum value of the updated remaining allocable bandwidths of L1 and L2 … Lm as the actual available bandwidth of the path where the first path occupies each link in the at least one link. For example, if the minimum value of the updated remaining allocable bandwidths of L1 and L2 … Lm is the updated remaining allocable bandwidth of L2, the updated remaining allocable bandwidth of L2 is used as the updated path actually occupied bandwidth of the first path.
Fig. 5 shows a flowchart of a bandwidth scheduling method according to an embodiment of the present invention. As shown in fig. 5, steps S1052, S1053 in fig. 2 include:
step S10531, when there is a link that does not release the occupied bandwidth of the standby path in the at least one first link, selecting one link that does not release the occupied bandwidth of the standby path in the at least one first link as the current first link.
Step S10532, if the current first link has a backup path that can release bandwidth, sequentially releasing the backup path with the lowest priority in the backup paths composed of the current first link to occupy the second occupied bandwidth of the current first link, and updating the remaining allocable bandwidth of the current first link each time according to the released second occupied bandwidth until the updated remaining allocable bandwidth of the current first link meets the required occupied bandwidth, or the current first link does not have a backup path that can release bandwidth.
Step S10533, when the updated remaining allocable bandwidth of the current first link does not satisfy the required occupied bandwidth and the current first link does not have a standby path capable of releasing the bandwidth, updating the actual occupied bandwidth of the path of the first path according to the updated remaining allocable bandwidth of the current first link.
In the foregoing example shown in fig. 4, after performing the process shown in fig. 4 for each of the at least one first link, the SDN controller determines a minimum value of the updated remaining allocable bandwidth of each of the at least one first link as a path actually occupiable bandwidth of each of the at least one link occupied by the first path.
Unlike the examples shown in fig. 3 and 4. In the example shown in fig. 5, for each first link in at least one first link, one first link may be arbitrarily selected as a current first link, and a standby path formed by the current first link may be released to occupy the occupied bandwidth of the current first link and the remaining allocable bandwidth of the current first link may be updated first until the updated remaining allocable bandwidth of the current first link meets the required occupied bandwidth, or the current first link does not have a standby path capable of releasing the bandwidth.
Then, judging whether the updated residual allocable bandwidth of the current first link meets the requirement for occupying the bandwidth and whether the current first link has a standby path capable of releasing the bandwidth, and if the updated residual allocable bandwidth of the current first link does not meet the requirement for occupying the bandwidth and the current first link does not have the standby path capable of releasing the bandwidth, updating the actual path capable of occupying the bandwidth of the first path according to the updated residual allocable bandwidth of the current first link; if the updated remaining allocable bandwidth of the current first link meets the requirement of occupying the bandwidth, or if the current first link still has a standby path capable of releasing the bandwidth, the remaining allocable bandwidth of the current first link is certainly not less than the minimum value of the remaining allocable bandwidth of each link in the at least one first link, and the actual path-occupied bandwidth of the first path does not need to be updated according to the updated remaining allocable bandwidth of the current first link.
And then judging whether a link which does not release the occupied bandwidth of the standby path exists in the at least one first link, if so, randomly selecting one link which does not release the occupied bandwidth of the standby path from the at least one first link as the current first link, and continuing the process until the link which does not release the occupied bandwidth of the standby path does not exist in the at least one first link. The method shown in fig. 5 is described in more detail and clearly below with reference to fig. 6.
Fig. 6 shows a schematic diagram of step S1052 and step S1053 according to an example of the present invention. In this embodiment, it is assumed that m first links, L1 and L2 … Lm respectively, of the at least one link, in which the remaining allocable bandwidth does not satisfy the demand occupied bandwidth, are determined. As shown in fig. 6, the SDN controller first determines whether there is a link that does not release the occupied bandwidth of the standby path in L1 and L2 … Lm, where L1 and L2 … Lm do not release the occupied bandwidth of the standby path at this time. If so, the SDN controller selects one of the links L1, L2 … Lm, which does not release the occupied bandwidth of the standby path, as the current first link.
Taking the current first link as L1 as an example, the SDN controller determines whether a standby path capable of releasing occupied bandwidth exists in the standby paths formed by L1. If yes, the SDN controller selects the standby path with the lowest priority from the standby paths formed by the L1, and releases the standby path with the lowest priority to occupy the first occupied bandwidth of the L1. The SDN controller updates the remaining allocable bandwidth of the L1 according to the released first occupied bandwidth; the SDN controller determines whether the updated remaining allocable bandwidth of L1 meets the bandwidth occupied by the first path. If yes, the SDN controller stops releasing a standby path formed by L1 to occupy the first occupied bandwidth of L1; if the standby path is not satisfied, continuously judging whether a standby path capable of releasing the bandwidth still exists in the standby paths composed of the L1, if so, continuously releasing the first occupied bandwidth of the L1 occupied by the standby path with the lowest priority in the standby paths composed of the L1 by the SDN controller, and if not, stopping releasing the first occupied bandwidth of the L1 occupied by the standby path composed of the L1.
After the first link L1 completes the above process, that is, after the first occupied bandwidth of L1 is stopped to be released from the standby path composed of L1, if the updated remaining allocable bandwidth of L1 does not satisfy the bandwidth occupied by the first path and there is no standby path of the releasable bandwidth in L1, the SDN controller updates the actual available bandwidth of the path of the first path according to the updated remaining allocable bandwidth of L1. That is, the SDN controller takes the updated remaining allocable bandwidth of L1 as the updated path actually occupiable bandwidth of the first path. It should be noted that after the first link L1 completes the above process, if the updated remaining allocable bandwidth of L1 meets the required occupied bandwidth of the first path, the SDN controller does not update the actual occupied bandwidth of the first path.
After the actual available bandwidth of the first path is updated, it is determined whether a link that does not release the available bandwidth of the standby path exists in L1 and L2 … Lm, at this time, L1 in L1 and L2 … Lm releases the available bandwidth of the standby path and L2 … Lm does not release the available bandwidth of the standby path, if so, one link that does not release the available bandwidth of the standby path in L2 … Lm is selected as the current first link, and the process shown in fig. 6 is continuously executed by taking the current first link as L2 as an example, until no link that does not release the available bandwidth of the standby path exists in L1 and L2 … Lm, that is, all L1 and L2 … Lm release the available bandwidth of the standby path.
Through the examples shown in fig. 5 and 6, the actual available bandwidth of the first path can be updated, the actual available bandwidth of the updated path of the first path is increased, the SDN controller allocates bandwidth to the first path according to the actual available bandwidth of the updated path of the first path, the bandwidth occupied by the backup path is dynamically reduced, and the utilization rate of the bandwidth can be increased under the condition that the disaster recovery effect is guaranteed as much as possible, so that the network can carry more data stream services.
Fig. 7 shows a flowchart of a bandwidth scheduling method according to an embodiment of the present invention. As shown in fig. 7, the step S105 further includes:
step S1055, identifying whether the updated path actually occupied bandwidth of the first path satisfies the bandwidth occupied by the first path.
As can be seen from fig. 4 and fig. 6, after the occupied bandwidth of the standby path of each first link in at least one first link is released, the minimum value of the updated remaining allocable bandwidth of each first link may meet the bandwidth occupied by the first path as required, or may not meet the bandwidth occupied by the first path as required, that is, the bandwidth actually occupied by the updated path of the first path may meet the bandwidth occupied by the first path as required, or may not meet the bandwidth occupied by the first path as required. Therefore, the SDN controller determines a relationship between an actual available bandwidth of the updated path of the first path and a required available bandwidth of the first path, so as to mark a state of the first path.
Step S1056, if the updated path of the first path actually can occupy the bandwidth to meet the requirement of the first path to occupy the bandwidth, marking the bandwidth status of the first path as the first status.
That is to say, the bandwidth allocated by the SDN controller for the first path can meet the requirement of the first path to occupy the bandwidth, and the bandwidth state of the first path is marked as the first state, that is, as the strict state.
Step S1057, if the updated path of the first path actually can occupy the bandwidth and does not meet the requirement of the first path to occupy the bandwidth, marking the bandwidth state of the first path as the second state.
That is to say, the bandwidth allocated by the SDN controller for the first path cannot meet the requirement of the first path to occupy the bandwidth, and the bandwidth state of the first path is marked as the second state, that is, as the marginal state.
Step S1058, updating the total occupied bandwidth of the main path formed by each link in the at least one link, and updating the updated remaining allocable bandwidth of each link in the at least one link again.
The SDN controller allocates part or all of the updated remaining allocable bandwidth of each of the at least one link to the first path. That is, the primary path formed by each link in the at least one link is added with the first path, the SDN controller may update the total occupied bandwidth of the primary path formed by each link in the at least one link. For example, the SDN controller may use a sum of the occupied total bandwidth of the primary path composed of each link in the at least one link before allocation and an actual occupiable bandwidth of the first path as the updated occupied total bandwidth of the primary path composed of each link in the at least one link.
The SDN controller allocates part or all of the updated remaining allocable bandwidth of each of the at least one link to the first path. The SDN controller may also update the updated remaining allocable bandwidth of each of the at least one link again. For example, the SDN controller takes the difference between the updated remaining allocable bandwidth of each link and the actual occupiable bandwidth of the first path as the updated remaining allocable bandwidth of each link again.
By marking the bandwidth state of the first path according to the condition of the bandwidth allocated to the first path, when redundant residual allocable bandwidth exists in a link in the network, the path with the bandwidth state meeting a certain condition is preferentially selected to allocate the bandwidth. According to the bandwidth scheduling method provided by the embodiment of the invention, the utilization rate of the bandwidth can be improved, so that the network can bear more data stream services.
Fig. 8 shows a flowchart of a bandwidth scheduling method according to an embodiment of the present invention. As shown in fig. 8, the method further comprises:
step S201, when the second path is removed from the network, determining a path attribute of the second path, and releasing a third occupied bandwidth of the second path.
When the data stream corresponding to the service is completely transmitted, the SDN controller may remove a path corresponding to the data stream in the network, or as described above, release the occupied bandwidth of the path due to the need of bandwidth scheduling, the need of removing the path from the network due to a path failure, and the like.
As an example, the second path may refer to any path where transmission of the corresponding data stream is completed, which is not limited in this embodiment of the present invention. Before releasing the occupied bandwidth of the second path, the SDN controller determines the path attribute of the second path, so as to determine a path to which bandwidth is allocated when bandwidth is subsequently allocated.
And for each link forming the second path, the SDN controller releases a third occupied bandwidth of the second path. As mentioned above, the occupied bandwidth of the second path on each link that it passes through is equal, and therefore, the third occupied bandwidth may also refer to the bandwidth occupied by the second path on the current link.
Step S202, according to the released third occupied bandwidth, the residual allocable bandwidth of the current link forming the second path is updated.
For example, for each link constituting the second path, the SDN controller uses the sum of the third occupied bandwidth and the remaining allocable bandwidth of the current link as the updated remaining allocable bandwidth of the current link.
Step S203, according to the path attribute of the second path, a third path having the second state is selected from all paths formed by the current link.
In one example, when the path attribute characterizes that the second path is the primary path, the SDN controller selects a third path having the second state and a lower priority than the second path from all paths composed of the current links.
In step S201, the SDN controller determines a path attribute of the second path. If the determined path attribute characterizes that the second path is the primary path, the SDN controller may allocate bandwidth for a path (third path) whose bandwidth state is marginal and whose priority is lower than that of the second path in all paths composed of the current link.
A path that is marginal in bandwidth status does not meet the bandwidth it needs to occupy in the previous allocation process. Thus, the SDN controller may prioritize these paths when there is remaining allocable bandwidth; in addition, in order to ensure normal transmission of data streams, the bandwidth of the main path should be ensured as much as possible.
Therefore, in the embodiment of the present invention, the SDN controller selects the third path with a bandwidth state being marginal and a priority lower than that of the second path from all paths formed by the current link, and allocates bandwidth to the third path from the updated remaining allocable bandwidth of the current link, so as to ensure data stream transmission of the main path and dynamically adjust the bandwidth of the standby path at the same time.
In another example, when the path attribute characterizes the second path as a backup path, the SDN controller selects a third path having the second state from all paths composed of current links.
In step S201, the SDN controller determines a path attribute of the second path. If the determined path attribute characterizes that the second path is a backup path, the SDN controller may allocate bandwidth for the marginal path for the bandwidth state in all paths composed of the current link.
Step S204, when the updated remaining allocable bandwidth of the current link is greater than 0, sequentially allocating bandwidth to the path with the highest priority in the third paths from the remaining allocable bandwidth, and updating the updated remaining allocable bandwidth again.
The SDN controller releases the third occupied bandwidth of the second path and updates the remaining allocable bandwidth of the current link in steps S201 and S202. If the updated remaining allocable bandwidth of the current link is greater than 0, it indicates that bandwidth can be allocated to other paths composed by the current link. In this embodiment of the present invention, the SDN controller preferentially allocates a bandwidth to the third path.
Specifically, the SDN controller may rank the third paths according to their priorities. According to the sorting result, the SDN controller allocates bandwidth to the third path (one of the third path having the highest priority) from the remaining allocable bandwidths in sequence from high to low according to the priority, and the specific allocated bandwidth amount may be determined according to the bandwidth required by the third path, the bandwidth currently occupied by the third path on the current link, and the remaining allocable bandwidth on the current link.
Further, after the SDN controller allocates the bandwidth to the third path, the occupied bandwidth of the third path is min (the occupied bandwidth of the third path + the remaining allocable bandwidth, min (the path of the third path on the other links may actually occupy the bandwidth)), that is, the minimum value of the remaining allocable bandwidth of each link through which the third path passes.
After the bandwidth is allocated to the path with the highest priority in the third paths, the SDN controller updates the updated remaining allocable bandwidth of the current link again, and the specific updated value may be determined according to the bandwidth allocated to the third paths.
Step S205, when the re-updated remaining allocable bandwidth is greater than 0, repeatedly executing a process of sequentially allocating bandwidth to the path with the highest priority in the third path from the remaining allocable bandwidth, and updating the updated remaining allocable bandwidth again until the re-updated remaining allocable bandwidth of the current link is 0, or the third path does not exist in all paths formed by the current link.
If the remaining allocable bandwidth of the current link is still greater than 0 after being allocated once, that is, when the remaining allocable bandwidth after being updated again is greater than 0 as described above, the SDN controller repeats the allocation process until the remaining allocable bandwidth of the current link is all allocated to the third path formed by the current link, that is, until the remaining allocable bandwidth after being updated again is 0 or no path with a bandwidth state that is marginal and a priority lower than that of the third path exists in all paths formed by the current link, and the SDN controller stops the bandwidth allocation process.
According to the bandwidth scheduling method provided by the embodiment of the invention, the transmission of the data stream on the main path can be ensured as much as possible, the occupied bandwidth of the standby path is dynamically adjusted, and the utilization rate of the bandwidth is improved under the condition that the disaster recovery effect is ensured as much as possible, so that the network can bear more data stream services.
Optionally, in a possible implementation manner, the method may further include a step of allocating occupied bandwidth to the path after the bandwidth occupied by the path is expanded, and the bandwidth utilization rate may be improved through this step. The method specifically comprises the following steps:
when the fact that the required occupied bandwidth of the first path is expanded is detected, and the expansion value of the required occupied bandwidth does not exceed the minimum value of the link residual allocable bandwidths of each link in the at least one link, the SDN controller allocates the bandwidth corresponding to the expansion value of the required occupied bandwidth for the first path from the residual allocable bandwidths of each link in the at least one link.
When the expansion of the required occupied bandwidth of the first path is detected, and the expansion value of the required occupied bandwidth exceeds the minimum value of the link residual allocable bandwidths of each link in the at least one link, the SDN controller allocates bandwidth for the first path from the residual allocable bandwidths of each link in the at least one link according to the path attribute of the first path.
The first path may refer to the first path after the bandwidth has been allocated in the method illustrated in fig. 1, or may refer to any path in the network, which is not limited in this embodiment of the present invention, and in this embodiment, the first path after the bandwidth has been allocated in the method illustrated in fig. 1 is taken as an example for description.
The change of the required occupied bandwidth of the path may be detected by related software, taking the first path as an example, if the required occupied bandwidth of the first path is detected to be enlarged, the SDN controller may determine an enlarged value of the required occupied bandwidth, and then determine a size relationship between the enlarged value of the required occupied bandwidth and a minimum value of link remaining allocable bandwidth of each link in the at least one link. After the bandwidth occupied by the first path is increased, the specific process of allocating the bandwidth to the first path by the SDN controller may refer to a process of loading the first path in the network, which is specifically described as follows:
if the expanded value of the required occupied bandwidth does not exceed (is less than or equal to) the minimum value of the remaining allocable bandwidths of each link in the at least one link, the SDN controller may allocate the bandwidth for the first path according to the manner described in step S104, which is not described again.
If the expansion value of the required occupied bandwidth exceeds (is larger than) the minimum value of the residual allocable bandwidths of each link in the at least one link, the SDN controller allocates bandwidth for the first path from the residual allocable bandwidths of each link in the at least one link according to the path attribute of the first path.
Specifically, if the first path is a standby path, the SDN controller allocates, to the first path, a bandwidth corresponding to a minimum value of remaining allocable bandwidths of each link in the at least one link from the remaining allocable bandwidths of each link in the at least one link.
And if the first path is the main path, the SDN controller determines at least one second link, of the at least one link, in which the remaining allocable bandwidth does not meet the expansion value of the occupied bandwidth.
Specifically, the relationship between the remaining allocable bandwidth and the enlarged value of the required occupied bandwidth of each link in the at least one link may be compared one by one, so as to determine at least one second link, in the at least one link, for which the remaining allocable bandwidth does not satisfy the enlarged value of the required occupied bandwidth.
And the SDN controller releases a fourth path composed of each second link in at least one second link to occupy the bandwidth of each second link, and updates the remaining allocable bandwidth of each second link, wherein the priority of the fourth path is lower than that of the first path.
And the SDN controller updates the minimum value of the residual allocable bandwidth of each link in the at least one link according to the updated residual allocable bandwidth of each second link in the at least one second link.
The above process may refer to the descriptions of steps S10521, S10522, and S10523, or the parts of steps S10531, S10532, and S10533, and will not be described again. The only difference is that: and replacing the first link with a second link, replacing the standby path with a fourth path, and replacing the actual occupied bandwidth of the path of the first path with the minimum value of the residual allocable bandwidths of each link in at least one link.
And the SDN controller allocates a bandwidth corresponding to the minimum value in the residual allocable bandwidths of each link in the at least one link for the first path from the updated residual allocable bandwidths of each link in the at least one link.
After updating the minimum value of the remaining allocable bandwidths of each of the at least one link, the SDN controller allocates a bandwidth corresponding to the minimum value of the remaining allocable bandwidths of each of the at least one link for the first path from the updated remaining allocable bandwidths of each of the at least one link constituting the first path, and updates the updated remaining allocable bandwidths of each of the at least one link again.
Optionally, if the updated remaining allocable bandwidth of each link in the at least one link is still greater than 0, the SDN controller may select an occupied bandwidth of a fourth path that expands the last released bandwidth, where a maximum occupied bandwidth of the expanded fourth path is min (an occupied bandwidth of the fourth path + a remaining allocable bandwidth, min (an actual available bandwidth of the fourth path on other links)), that is, a minimum value of the remaining allocable bandwidth of each link that the fourth path passes through. Other procedures are not described in detail.
Optionally, in a possible implementation manner, the method may further include a step of allocating occupied bandwidth to other paths after the bandwidth occupied by the path is reduced, and the bandwidth utilization rate may be improved through this step. The method specifically comprises the following steps:
when the fact that the bandwidth occupied by the demand of the first path is reduced is detected, the SDN controller releases the bandwidth corresponding to the reduced value of the bandwidth occupied by the demand of the first path.
And the SDN controller updates the residual allocable bandwidth of the current link forming the first path according to the bandwidth corresponding to the released reduction value (the difference between the bandwidth occupied by the original requirement of the first path and the bandwidth occupied by the current requirement).
The SDN controller selects a fifth path with the second state from all paths composed by the current link.
And when the updated remaining allocable bandwidth of the current link is larger than 0, the SDN controller sequentially allocates the bandwidth to the path with the highest priority in the fifth path from the remaining allocable bandwidths, and updates the updated remaining allocable bandwidth again.
When the updated remaining allocable bandwidth is greater than 0, the SDN controller repeatedly executes a process of sequentially allocating a bandwidth to a path with the highest priority in a fifth path from the remaining allocable bandwidths, and updating the updated remaining allocable bandwidth again until the remaining allocable bandwidth of the current link is 0 or the fifth path does not exist in all paths formed by the current link.
The SDN controller may detect a change in bandwidth occupied by a demand of a path through related software, taking the first path as an example, and if detecting that the bandwidth occupied by the demand of the first path is reduced, the SDN controller may determine a reduction value of the bandwidth occupied by the demand, and perform the following process for each of at least one link constituting the first path: releasing the bandwidth corresponding to the reduced value of the bandwidth occupied by the requirement of the first path, and taking the sum of the remaining allocable bandwidth of the current link before release and the bandwidth corresponding to the reduced value as the updated remaining allocable bandwidth of the current link; judging whether a fifth path with the second state exists in all paths formed by the current link, if so, sorting the fifth path according to the priority of the fifth path, and the specific process may refer to the descriptions of steps S204 to S205, which is not described again, and the difference is that: after the bandwidth is allocated to the fifth path, the occupied bandwidth of the fifth path is min (the occupied bandwidth of the fifth path + the remaining allocable bandwidth, min (the path of the fifth path on the other links may actually occupy the bandwidth)), that is, the minimum value of the remaining allocable bandwidth of each link through which the fifth path passes. Other procedures are not described in detail.
Optionally, in a possible implementation manner, the method may further include a step of allocating occupied bandwidth to a path formed by the link after the bandwidth of the link becomes larger, and the bandwidth utilization rate may be improved through the step. The method specifically comprises the following steps:
when detecting that the bandwidth of the third link becomes large, the SDN controller determines the bandwidth increment of the third link.
The third link may be any link in the network, which is not limited in this embodiment of the present invention.
The bandwidth of a link generally refers to the number of bits per second that can be transmitted on the link, depending on the link clock rate and channel coding. The bandwidth change of each link in the network can be detected through the related technology, and when the bandwidth change of the link is detected, the occupied bandwidth of a path formed by the links needs to be adjusted according to the bandwidth change of the link.
For example, as described above, when it is detected that the bandwidth of the third link becomes large, the SDN controller determines a bandwidth increment of the third link to allocate the bandwidth for the path formed by the third link.
And updating the residual allocable bandwidth of the third link according to the bandwidth increment of the third link by the SDN controller.
For example, the SDN controller adds the bandwidth increment of the third link to the remaining allocable bandwidth of the third link before the link bandwidth becomes larger as the updated remaining allocable bandwidth of the third link.
The SDN controller selects a sixth path having the second state from all paths composed of the third links.
Specifically, the SDN controller determines whether a sixth path in a marginal state exists in all paths formed by the third link, and if so, the SDN controller sequences the sixth path according to the priority of the sixth path, where the specific process refers to the description of step S10521, and is not described again. If not, the operation may end.
When the updated remaining allocable bandwidth of the third link is greater than 0, the SDN controller sequentially allocates bandwidths to the path with the highest priority in the sixth path from the remaining allocable bandwidths, and updates the updated remaining allocable bandwidth again.
When the updated remaining allocable bandwidth is greater than 0, the SDN controller repeatedly executes a process of sequentially allocating a bandwidth to a path with the highest priority in the sixth path from among the remaining allocable bandwidths, and updating the updated remaining allocable bandwidth again until the updated remaining allocable bandwidth of the third link is 0, or the sixth path does not exist in all paths formed by the third link.
The above process can be referred to the description in the sections of steps S204 to S205 (the current link is replaced by the third link, and the third path is replaced by the sixth path), and after the bandwidth is allocated to the sixth path, the occupied bandwidth of the sixth path is min (the occupied bandwidth of the sixth path + the remaining allocable bandwidth, min (the path of the sixth path on the other links may actually occupy the bandwidth)), that is, the minimum value of the remaining allocable bandwidth of each link that the sixth path passes through. Other procedures are not described in detail.
Optionally, in a possible implementation manner, the method further includes a step of releasing the occupied bandwidth of the path composed of the link after the bandwidth of the link becomes smaller, and through this step, the utilization rate of the bandwidth may be improved. The method specifically comprises the following steps:
when the bandwidth of the third link is detected to be reduced, the SDN controller determines the bandwidth reduction amount of the third link;
and the SDN controller updates the total bandwidth of the third link according to the bandwidth reduction amount of the third link.
After the above process is completed, the SDN controller may determine all paths formed by the third link, calculate occupied bandwidths of all paths on the third link, determine whether the occupied bandwidths of all paths on the third link can be satisfied according to the updated total bandwidth of the third link, and determine whether to release a part of paths according to a result of the determination.
And if the updated total bandwidth of the third link is smaller than the occupied bandwidth of all paths formed by the third link, the SDN controller sorts all the paths according to the priority of all the paths formed by the third link. The specific sorting process is described above and will not be described in detail.
And when the paths capable of releasing the bandwidth exist in all the paths formed by the third link, the SDN controller sequentially releases the occupied bandwidth of all the paths formed by the third link on the third link according to the sequencing result and updates the occupied bandwidth of all the paths formed by the third link.
Specifically, the SDN controller may sequentially release the occupied bandwidths of the paths with the lowest priority from low to high according to the priority, and update the occupied bandwidths of all the paths composed of the third links according to the released occupied bandwidths of the paths with the lowest priority. For example, the SDN controller may use a difference between an occupied bandwidth of all paths composed of the third link before the release and an occupied bandwidth of a currently released path with the lowest priority as an updated occupied bandwidth of all paths composed of the third link.
When the updated occupied bandwidth of all paths composed of the third link is larger than the updated total bandwidth of the third link, the SDN controller repeatedly executes a process of sequentially releasing the occupied bandwidth of all paths composed of the third link on the third link and updating the occupied bandwidth of all paths composed of the third link according to the sequencing result when a releasable path exists, until the occupied bandwidth of all paths composed of the third link is not larger than the updated total bandwidth of the third link, or no path capable of releasing the bandwidth exists in all paths composed of the third link.
According to the judgment result, if the updated total bandwidth of the third link is smaller than the occupied bandwidth of all paths composed of the third link, the SDN controller needs to release the occupied bandwidth of part of the paths in all the paths composed of the third link to meet the updated total bandwidth of the third link. If the updated total bandwidth of the third link is greater than the occupied bandwidth of all paths composed of the third link or no releasable path exists in all paths composed of the third link, the operation may be ended.
Optionally, after the above process is finished, when the updated total bandwidth of the third link is greater than the updated occupied bandwidth of all paths formed by the third link, the SDN controller expands the occupied bandwidth of the last path that releases the occupied bandwidth of the third link, and after expanding the occupied bandwidth of the path, the occupied bandwidth of the path is min (the occupied bandwidth of the path + the remaining allocable bandwidth of the third link, min (the path may actually occupy the bandwidth on other links)), that is, the minimum value of the remaining allocable bandwidth of each link that the path passes through.
Fig. 9 shows a block diagram of a bandwidth scheduling apparatus according to an embodiment of the present invention. As shown in fig. 9, the apparatus is applied to an SDN controller, and includes:
a first determining module 31, configured to determine, when a first path is loaded in a network, at least one link constituting the first path;
an obtaining module 32, configured to obtain an actual bandwidth that the first path occupies for each link in the at least one link, where the actual bandwidth that the path occupies is a minimum value of remaining allocable bandwidths of each link in the at least one link;
an identifying module 33, configured to identify whether the actual occupiable bandwidth of the path satisfies a required occupied bandwidth of the first path;
a first allocating module 34, configured to allocate an actual occupiable bandwidth of the first path for the first path from the remaining allocable bandwidth of each of the at least one link if the path actual occupiable bandwidth meets the required occupied bandwidth;
a second allocating module 35, configured to allocate bandwidth for the first path from the remaining allocable bandwidth of each link in the at least one link according to the path attribute of the first path if the bandwidth actually occupied by the path does not meet the bandwidth occupied by the demand.
According to the bandwidth scheduling device of the embodiment, when the path of the path can actually occupy the bandwidth and cannot occupy the bandwidth according to the requirement of the path, the bandwidth is allocated to the path according to the loaded path attribute of the path, so that the bandwidth occupied by the backup path can be dynamically reduced, and the utilization rate of the bandwidth is improved under the condition of ensuring the disaster recovery effect as much as possible, so that the network can bear more data stream services.
Fig. 10 shows a block diagram of a bandwidth scheduling apparatus according to an embodiment of the present invention. As shown in fig. 10, in one possible implementation, the second allocating module 35 includes:
a determining unit 351, configured to determine, if the first path is a primary path, at least one first link, of the at least one link, where a remaining allocable bandwidth does not meet the bandwidth occupied by the demand;
a bandwidth releasing unit 352, configured to release that a standby path formed by each first link in the at least one first link occupies a bandwidth of each first link, and update a remaining allocable bandwidth of each first link;
a first updating unit 353, configured to update, according to the updated remaining allocable bandwidth of each first link in the at least one first link, a path actually occupied by the first path for occupying each link in the at least one link;
a first allocating unit 354, configured to allocate, from the updated remaining allocable bandwidth of each link of the at least one link, an updated path actually occupied bandwidth of the first path for the first path.
In one possible implementation manner, the bandwidth releasing unit 352 includes:
a first releasing subunit 3521, configured to sequentially release, if the first link has a standby path that can release occupied bandwidth, a standby path with a lowest priority in the standby paths formed by the first link to occupy the first occupied bandwidth of the first link;
a first updating subunit 3522, configured to update the remaining allocable bandwidth of the first link each time according to the released first occupied bandwidth;
a release stopping subunit 3523, configured to stop releasing the standby path composed of the first link from occupying the first occupied bandwidth of the first link if the updated remaining allocable bandwidth of the first link meets the requirement for occupying the bandwidth, or if the first link does not have a standby path that can release the bandwidth.
In one possible implementation manner, the bandwidth releasing unit 352 includes:
a selecting subunit 3531, configured to, in a case that there is a link that does not release the occupied bandwidth of the standby path in the at least one first link, select, as a current first link, one link that does not release the occupied bandwidth of the standby path in the at least one first link;
a second releasing subunit 3532, configured to, if the current first link has a standby path that can release bandwidth, sequentially release, that the standby path with the lowest priority in the standby paths formed by the current first link occupies the second occupied bandwidth of the current first link, and update, each time according to the released second occupied bandwidth, the remaining allocable bandwidth of the current first link until the updated remaining allocable bandwidth of the current first link meets the required occupied bandwidth, or the current first link does not have a standby path that can release bandwidth;
the update unit 353 includes:
a second updating subunit 3533, configured to update, when the updated remaining allocable bandwidth of the current first link does not meet the bandwidth occupied by the demand, and there is no spare path of the releasable bandwidth on the current first link, the actual available bandwidth of the path of the first path according to the updated remaining allocable bandwidth of the current first link.
In a possible implementation manner, the second allocating module 35 further includes:
an identifying unit 355, configured to identify whether an updated path actually occupiable bandwidth of the first path satisfies a required occupancy bandwidth of the first path;
a first marking unit 356, configured to mark the bandwidth status of the first path as a first status if the updated path actually occupies the bandwidth of the first path and satisfies the bandwidth occupied by the first path;
a second marking unit 357, configured to mark the bandwidth status of the first path as a second status if the updated path actually available bandwidth of the first path does not meet the bandwidth occupied by the demand of the first path;
a second updating unit 358, configured to update the occupied total bandwidth of the primary path formed by each link in the at least one link, and update the updated remaining allocable bandwidth of each link in the at least one link again.
In a possible implementation manner, the second allocating module 35 includes:
a second allocating unit 359, configured to, if the first path is a standby path, allocate, from the remaining allocable bandwidth of each link in the at least one link, an actual available bandwidth of the path of the first path to the first path;
second labeling unit 357 is further configured to label the bandwidth status of the first path as a second status;
a third updating unit 360, configured to update the remaining allocable bandwidth of each link in the at least one link, and the total bandwidth occupied by the backup path formed by each link in the at least one link.
In one possible implementation, the apparatus further includes:
a second determining module 41, configured to, when a second path is removed from the network, determine a path attribute of the second path, and release a third occupied bandwidth of the second path;
an updating module 42, configured to update, according to the released third occupied bandwidth, a remaining allocable bandwidth of a current link that forms the second path;
a selecting module 43, configured to select, according to the path attribute of the second path, a third path having the second state from all paths formed by the current link;
a third allocating module 44, configured to, when the updated remaining allocable bandwidth of the current link is greater than 0, sequentially allocate bandwidth to a path with a highest priority in the third paths from the remaining allocable bandwidth, and update the updated remaining allocable bandwidth again;
and a distribution stopping module 45, configured to, when the remaining allocable bandwidth after the second update is greater than 0, repeatedly perform a process of sequentially distributing bandwidth to a path with the highest priority in the third paths from among the remaining allocable bandwidth, and updating the updated remaining allocable bandwidth again until the remaining allocable bandwidth after the second update of the current link is 0, or the third path does not exist in all paths formed by the current link.
In one possible implementation, the selection module 43 includes one or more of the following units:
a first selecting unit 431, configured to select, when the path attribute indicates that the second path is a primary path, a third path having the second state and a priority lower than that of the second path from all paths composed of the current link;
a second selecting unit 432, configured to select a third path with the second state from all paths formed by the current link when the path attribute indicates that the second path is a standby path.
Fig. 11 is a block diagram illustrating a bandwidth scheduling apparatus 900 according to an example embodiment. Referring to fig. 11, the apparatus 900 may include a processor 901, a machine-readable storage medium 902 having stored thereon machine-executable instructions. The processor 901 and the machine-readable storage medium 902 may communicate via a system bus 903. Also, the processor 901 performs the bandwidth scheduling method described above by reading machine executable instructions in the machine readable storage medium 902 corresponding to the bandwidth scheduling logic.
The machine-readable storage medium 902 referred to herein may be any electronic, magnetic, optical, or other physical storage device that can contain or store information such as executable instructions, data, and the like. For example, the machine-readable storage medium may be: random Access Memory (RAM), volatile Memory, non-volatile Memory, flash Memory, a storage drive (e.g., a hard drive), a solid state drive, any type of storage disk (e.g., an optical disk, dvd, etc.), or similar storage media, or a combination thereof.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein were chosen in order to best explain the principles of the embodiments, the practical application, or technical improvements to the techniques in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims (14)

1. A method of bandwidth scheduling, the method comprising:
determining at least one link constituting a first path when the first path is loaded in a network;
acquiring the actual path occupied bandwidth of each link in the at least one link occupied by the first path, wherein the actual path occupied bandwidth is the minimum value of the residual allocable bandwidth of each link in the at least one link;
identifying whether the actual occupiable bandwidth of the path meets the required occupied bandwidth of the first path;
if the path actual available bandwidth meets the required available bandwidth, allocating the first path actual available bandwidth from the remaining allocable bandwidth of each link in the at least one link;
if the actual available bandwidth of the path does not meet the bandwidth occupied by the demand, allocating bandwidth for the first path from the residual available bandwidth of each link in the at least one link according to the path attribute of the first path;
if the actual available bandwidth of the path does not meet the required available bandwidth, allocating bandwidth for the first path from the remaining available bandwidth of each link in the at least one link according to the path attribute of the first path, including:
if the first path is a main path, determining at least one first link, of the at least one link, in which the remaining allocable bandwidth does not meet the bandwidth occupied by the demand;
releasing the standby path formed by each first link in the at least one first link to occupy the bandwidth of each first link, and updating the residual allocable bandwidth of each first link;
updating the actual bandwidth occupied by the first path for each link in the at least one link according to the updated remaining allocable bandwidth of each first link in the at least one first link;
allocating an updated path actual available bandwidth of the first path for the first path from the updated remaining available bandwidth of each of the at least one link.
2. The method according to claim 1, wherein the releasing the backup path formed by each of the at least one first link occupies the bandwidth of each first link, and the updating the remaining allocable bandwidth of each first link comprises:
if the first link has a standby path which can release occupied bandwidth, sequentially releasing a standby path with the lowest priority from the standby paths consisting of the first link to occupy the first occupied bandwidth of the first link;
updating the remaining allocable bandwidth of the first link according to the released first occupied bandwidth each time;
and if the updated remaining allocable bandwidth of the first link meets the requirement for occupying the bandwidth, or the first link does not have a standby path capable of releasing the bandwidth, stopping releasing the standby path composed of the first link to occupy the first occupied bandwidth of the first link.
3. The method according to claim 1, wherein the releasing the backup path formed by each of the at least one first link occupies the bandwidth of each first link, and the updating the remaining allocable bandwidth of each first link comprises:
selecting one link which does not release the occupied bandwidth of the standby path from the at least one first link as a current first link under the condition that the link which does not release the occupied bandwidth of the standby path exists in the at least one first link;
if the current first link has a standby path capable of releasing the bandwidth, sequentially releasing a standby path with the lowest priority in the standby paths composed of the current first link to occupy the second occupied bandwidth of the current first link, and updating the remaining allocable bandwidth of the current first link each time according to the released second occupied bandwidth until the updated remaining allocable bandwidth of the current first link meets the required occupied bandwidth, or the current first link does not have a standby path capable of releasing the bandwidth;
the updating, according to the updated remaining allocable bandwidth of each first link in the at least one first link, that the first path occupies the actual bandwidth that the path of each link in the at least one link may occupy includes:
and when the updated residual allocable bandwidth of the current first link does not meet the requirement occupied bandwidth and the current first link does not have a standby path capable of releasing the bandwidth, updating the actual available bandwidth of the path of the first path according to the updated residual allocable bandwidth of the current first link.
4. The method according to claim 2 or 3, wherein if the actual bandwidth available for the path does not satisfy the bandwidth required for the path, allocating bandwidth for the first path from the remaining available bandwidth of each of the at least one link according to the path attribute of the first path, further comprises:
identifying whether the updated path actually available bandwidth of the first path meets the bandwidth occupied by the first path;
if the updated path of the first path can actually occupy the bandwidth to meet the requirement of the first path to occupy the bandwidth, marking the bandwidth state of the first path as a first state;
if the updated path actually available bandwidth of the first path does not meet the bandwidth occupied by the demand of the first path, marking the bandwidth state of the first path as a second state;
and updating the occupied total bandwidth of the main path formed by each link in the at least one link, and updating the updated residual allocable bandwidth of each link in the at least one link again.
5. The method of claim 1, wherein if the actual bandwidth available for the path does not satisfy the bandwidth required for the path, allocating bandwidth for the first path from the remaining available bandwidth of each of the at least one link according to the path attribute of the first path, further comprises:
if the first path is a standby path, allocating an actual available bandwidth of the path of the first path to the first path from the remaining available bandwidth of each link in the at least one link;
marking a bandwidth status of the first path as a second state;
and updating the residual allocable bandwidth of each link in the at least one link and the occupied total bandwidth of the standby path formed by each link in the at least one link.
6. The method of claim 4, further comprising:
when a second path is removed from the network, determining the path attribute of the second path, and releasing a third occupied bandwidth of the second path;
updating the residual allocable bandwidth of the current link forming the second path according to the released third occupied bandwidth;
selecting a third path with the second state from all paths formed by the current link according to the path attribute of the second path;
when the updated remaining allocable bandwidth of the current link is greater than 0, sequentially allocating bandwidth to the path with the highest priority in the third paths from the remaining allocable bandwidth, and updating the updated remaining allocable bandwidth again;
when the updated remaining allocable bandwidth is greater than 0, repeatedly executing a process of sequentially allocating bandwidth to the path with the highest priority in the third paths from the remaining allocable bandwidth, and updating the updated remaining allocable bandwidth again until the updated remaining allocable bandwidth of the current link is 0 or the third path does not exist in all paths formed by the current link.
7. The method according to claim 6, wherein selecting the third path having the second status from all paths composed by the current link according to the path attribute of the second path comprises:
when the path attribute represents that the second path is a main path, selecting a third path which has the second state and has a priority lower than that of the second path from all paths formed by the current link;
or,
and when the path attribute represents that the second path is a standby path, selecting a third path with the second state from all paths formed by the current link.
8. A bandwidth scheduling apparatus, the apparatus comprising:
a first determining module for determining at least one link constituting a first path when the first path is loaded in a network;
an obtaining module, configured to obtain an actual bandwidth that the first path occupies for each link in the at least one link, where the actual bandwidth that the path occupies is a minimum value of remaining allocable bandwidths of each link in the at least one link;
the identification module is used for identifying whether the actual occupied bandwidth of the path meets the bandwidth occupied by the requirement of the first path;
a first allocating module, configured to allocate, if the path actual available bandwidth meets the required available bandwidth, the actual available bandwidth of the first path for the first path from the remaining available bandwidth of each link in the at least one link;
a second allocating module, configured to allocate bandwidth to the first path from the remaining allocable bandwidth of each link in the at least one link according to a path attribute of the first path if the actual allocable bandwidth of the path does not meet the bandwidth occupied by the demand;
the second allocating module includes:
a determining unit, configured to determine, if the first path is a primary path, at least one first link, of the at least one link, where a remaining allocable bandwidth does not meet the bandwidth occupied by the demand;
the bandwidth releasing unit is used for releasing the standby path formed by each first link in the at least one first link to occupy the bandwidth of each first link and updating the residual allocable bandwidth of each first link;
a first updating unit, configured to update, according to the updated remaining allocable bandwidth of each first link in the at least one first link, a path actually occupied by the first path for each link in the at least one link;
a first allocating unit, configured to allocate an updated path actual available bandwidth of the first path to the first path from the updated remaining available bandwidth of each link in the at least one link.
9. The apparatus of claim 8, wherein the bandwidth releasing unit comprises:
a first release subunit, configured to sequentially release, if the first link has a standby path that can release occupied bandwidth, a standby path with a lowest priority in the standby paths formed by the first link to occupy the first occupied bandwidth of the first link;
a first updating subunit, configured to update the remaining allocable bandwidth of the first link according to the released first occupied bandwidth each time;
and a release stopping subunit, configured to stop releasing the standby path composed of the first link from occupying the first occupied bandwidth of the first link if the updated remaining allocable bandwidth of the first link meets the requirement for occupying the bandwidth, or if the first link does not have a standby path that can release the bandwidth.
10. The apparatus of claim 8, wherein the bandwidth releasing unit comprises:
a selecting subunit, configured to select, when there is a link that does not release the bandwidth occupied by the standby path in the at least one first link, one link that does not release the bandwidth occupied by the standby path in the at least one first link as a current first link;
a second releasing subunit, configured to, if the current first link has a standby path that can release bandwidth, sequentially release a standby path with a lowest priority in the standby paths formed by the current first link to occupy a second occupied bandwidth of the current first link, and update the remaining allocable bandwidth of the current first link each time according to the released second occupied bandwidth until the updated remaining allocable bandwidth of the current first link meets the required occupied bandwidth, or the current first link does not have a standby path that can release bandwidth;
the update unit includes:
and a second updating subunit, configured to update, when the updated remaining allocable bandwidth of the current first link does not meet the requirement occupied bandwidth and the current first link does not have a standby path capable of releasing the bandwidth, an actual available bandwidth of the path of the first path according to the updated remaining allocable bandwidth of the current first link.
11. The apparatus of claim 9 or 10, wherein the second allocating module further comprises:
the identification unit is used for identifying whether the updated path actually occupied bandwidth of the first path meets the bandwidth occupied by the first path;
a first marking unit, configured to mark the bandwidth status of the first path as a first status if the updated path actually available bandwidth of the first path satisfies the bandwidth occupied by the first path;
a second marking unit, configured to mark the bandwidth status of the first path as a second status if the updated path actually available bandwidth of the first path does not meet the bandwidth occupied by the demand of the first path;
a second updating unit, configured to update a total occupied bandwidth of the primary path formed by each link in the at least one link, and update the updated remaining allocable bandwidth of each link in the at least one link again.
12. The apparatus of claim 8, wherein the second allocation module comprises:
a second allocating unit, configured to allocate, if the first path is a standby path, an actual available bandwidth of the path of the first path to the first path from the remaining available bandwidth of each link in the at least one link;
the second marking unit is further used for marking the bandwidth state of the first path as a second state;
and the third updating unit is used for updating the residual allocable bandwidth of each link in the at least one link and the occupied total bandwidth of the standby path formed by each link in the at least one link.
13. The apparatus of claim 11, further comprising:
a second determining module, configured to determine a path attribute of a second path when the second path is removed from the network, and release a third occupied bandwidth of the second path;
an updating module, configured to update remaining allocable bandwidth of a current link forming the second path according to the released third occupied bandwidth;
a selection module, configured to select, according to the path attribute of the second path, a third path having the second state from all paths formed by the current link;
a third allocating module, configured to, when the updated remaining allocable bandwidth of the current link is greater than 0, sequentially allocate a bandwidth to a path with a highest priority in the third paths from the remaining allocable bandwidths, and update the updated remaining allocable bandwidth again;
and a distribution stopping module, configured to, when the remaining allocable bandwidth after the re-update is greater than 0, repeatedly perform a process of sequentially distributing bandwidth to a path with the highest priority in the third paths from among the remaining allocable bandwidth, and updating the updated remaining allocable bandwidth again until the remaining allocable bandwidth after the re-update of the current link is 0, or the third path does not exist in all paths formed by the current link.
14. The apparatus of claim 13, wherein the selection module comprises one or more of:
a first selecting unit, configured to select, when the path attribute indicates that the second path is a primary path, a third path having the second state and a priority lower than that of the second path from all paths formed by the current link;
and a second selecting unit, configured to select a third path in the second state from all paths formed by the current link when the path attribute indicates that the second path is a standby path.
CN201810082837.3A 2018-01-29 2018-01-29 Bandwidth scheduling method and device Active CN108337189B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810082837.3A CN108337189B (en) 2018-01-29 2018-01-29 Bandwidth scheduling method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810082837.3A CN108337189B (en) 2018-01-29 2018-01-29 Bandwidth scheduling method and device

Publications (2)

Publication Number Publication Date
CN108337189A CN108337189A (en) 2018-07-27
CN108337189B true CN108337189B (en) 2021-11-05

Family

ID=62926622

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810082837.3A Active CN108337189B (en) 2018-01-29 2018-01-29 Bandwidth scheduling method and device

Country Status (1)

Country Link
CN (1) CN108337189B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111031516B (en) * 2018-10-10 2024-04-09 上汽通用汽车有限公司 Method and apparatus for providing security protection for an in-vehicle communication system
CN112468403B (en) * 2019-09-06 2022-08-26 华为技术有限公司 Communication method and related equipment thereof
CN111817890B (en) * 2020-07-07 2023-04-18 国家电网有限公司 Data synchronization processing method and device, computer equipment and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101127689A (en) * 2006-08-15 2008-02-20 中国电信股份有限公司 Method for minimizing priority seizing cost
CN101155131A (en) * 2006-09-29 2008-04-02 中国电信股份有限公司 Method for establishing label switched path of minimized path preemption cost
CN102130787A (en) * 2011-03-02 2011-07-20 华为技术有限公司 Network management equipment and optical channel data unit (ODU) flex frame idle timeslot management method thereof
CN103647711A (en) * 2013-12-20 2014-03-19 大连大学 Priority mechanism based satellite network rerouting method
CN103780474A (en) * 2014-01-13 2014-05-07 杭州华三通信技术有限公司 Traffic engineering tunnel resource preemption method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8179786B2 (en) * 2004-05-19 2012-05-15 Mosaid Technologies Incorporated Dynamic traffic rearrangement and restoration for MPLS networks with differentiated services capabilities

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101127689A (en) * 2006-08-15 2008-02-20 中国电信股份有限公司 Method for minimizing priority seizing cost
CN101155131A (en) * 2006-09-29 2008-04-02 中国电信股份有限公司 Method for establishing label switched path of minimized path preemption cost
CN102130787A (en) * 2011-03-02 2011-07-20 华为技术有限公司 Network management equipment and optical channel data unit (ODU) flex frame idle timeslot management method thereof
CN103647711A (en) * 2013-12-20 2014-03-19 大连大学 Priority mechanism based satellite network rerouting method
CN103780474A (en) * 2014-01-13 2014-05-07 杭州华三通信技术有限公司 Traffic engineering tunnel resource preemption method and device

Also Published As

Publication number Publication date
CN108337189A (en) 2018-07-27

Similar Documents

Publication Publication Date Title
CN108337189B (en) Bandwidth scheduling method and device
CN112015583B (en) Data storage method, device and system
CN114356557B (en) Cluster capacity expansion method and device
CN110244901B (en) Task allocation method and device and distributed storage system
US9058166B2 (en) Method and apparatus for managing processing resources in a distributed processing system
CN109032533B (en) Data storage method, device and equipment
JP2007049336A (en) Method for restoring fault, communication node and network
CN108337190B (en) Shunting method and device
CN110858986A (en) Bandwidth adjusting method, device, communication equipment and computer readable storage medium
CN110795234A (en) Resource scheduling method and device
CN113193996A (en) Power optical transmission network optimization method, device, equipment and storage medium
CN112306692B (en) Method and device for group-in-group balancing of distributed storage system
CN107623711B (en) Distribution method and device for master node and slave nodes in cluster
CN108073445B (en) Backpressure processing method and system based on distributed flow calculation
CN104486401A (en) Partitioned data control method and system
WO2021012588A1 (en) Method and system for establishing flexible ethernet management channel
WO2018161754A1 (en) Method and apparatus for recovering tunnel bandwidth resource
CN105763348B (en) Method and system for recovering optical layer service
CN107153570B (en) Method and device for controlling number of virtual machines
CN109587218A (en) A kind of method and apparatus of cluster election
US9420583B2 (en) Combined voice and data communications in a distributed hybrid allocation and reservation multiple access mobile wireless network
CN106330748A (en) Flow control method and apparatus
CN106612519B (en) A kind of single board backup method and device of radio network controller (RNC) equipment
TW201433127A (en) Method and apparatus for transmission and determination of timeslot status
US20180077607A1 (en) Method for authorising transmission requests

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant