WO2014173367A2 - 实现QoS的方法、系统、设备及计算机存储介质 - Google Patents

实现QoS的方法、系统、设备及计算机存储介质 Download PDF

Info

Publication number
WO2014173367A2
WO2014173367A2 PCT/CN2014/079723 CN2014079723W WO2014173367A2 WO 2014173367 A2 WO2014173367 A2 WO 2014173367A2 CN 2014079723 W CN2014079723 W CN 2014079723W WO 2014173367 A2 WO2014173367 A2 WO 2014173367A2
Authority
WO
WIPO (PCT)
Prior art keywords
qos
configuration information
module
flow table
qos configuration
Prior art date
Application number
PCT/CN2014/079723
Other languages
English (en)
French (fr)
Other versions
WO2014173367A3 (zh
Inventor
梁乾灯
廖婷
于波
肖敏
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2014173367A2 publication Critical patent/WO2014173367A2/zh
Publication of WO2014173367A3 publication Critical patent/WO2014173367A3/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/24Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2425Traffic characterised by specific attributes, e.g. priority or QoS for supporting services specification, e.g. SLA
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/24Traffic characterised by specific attributes, e.g. priority or QoS

Definitions

  • the present invention relates to communication technologies, and in particular, to a method, system, device and computer storage medium for implementing QoS. Background technique
  • FIG. 1 is a schematic diagram of a service flow processing process for implementing QoS in a network device in the prior art. As shown in FIG. 1 , the QoS of the network in the prior art mainly includes the following four steps: 1. Flow classification and flow labeling. A flow is a group of data packets with the same characteristics.
  • Traffic classification classifies traffic entering the network according to certain flow division rules (such as port-based or packet-based field values).
  • the flow label is used to set the service level and priority mark in the network for the service packets after the traffic classification, so as to implement QoS differentiation of different service flows.
  • Rate Rate
  • TrTCM Transmission Three Color Marker
  • IETF Internet Engineering Task Force
  • the algorithm marks the packets that do not exceed the CIR as green, marks the packets that exceed the CIR and does not exceed the PIR as yellow, marks the packets that exceed the PIR as red, and then discards them. Red packets implement rate limiting for traffic.
  • queue congestion management and scheduling Traffic that is measured, shaded, and supervised enters different levels of cache queues. This process is called queuing.
  • the queue congestion management is to monitor the queue buffer, and to discard the packet by means of TD, Tail Drop, RED, Random Early Detection, or Weighted Random Early Detection (WRED).
  • Queue scheduling is to schedule the dequeue sequence of packets buffered in different queues to avoid congestion and meet the dequeue of high-priority traffic.
  • Common queue scheduling algorithms include: Strict Priority Queue (SPQ) scheduling algorithm, Weighted Fair Queue (WFQ) scheduling algorithm, and Deficit Weighted Round Robin (DWRR) scheduling algorithm.
  • SPQ Priority Queue
  • WFQ Weighted Fair Queue
  • DWRR Deficit Weighted Round Robin
  • traffic shaping and priority rewriting After the queue is scheduled, the traffic is generally configured through the leaky bucket mechanism. Traffic shaping is the same as traffic policing. The difference between the two is that traffic shaping does not directly drop traffic that exceeds the limiting rate.
  • the traffic is cached first, and the cached messages are sent out when appropriate. This allows traffic to maintain a relatively uniform rate, which helps reduce the probability of packet loss on downstream network nodes due to bursty traffic.
  • the packet priority can be rewritten to affect the QoS processing of the downstream network node.
  • the QoS function of the network can be used to implement single layer QoS or Hierarchical QoS (HQoS).
  • Single-layer QoS can only be based on ports for service bandwidth control and traffic management, while hierarchical QoS can be implemented.
  • 2 is a schematic diagram of a hierarchical QoS layered traffic management model in the prior art. As shown in FIG. 2, hierarchical QoS implements four levels of traffic management at a port level, a user group level, a user level, and a service level, and is implemented by a physical port.
  • Level-level scheduling The second-level scheduling by user group is realized by the user group; the third-level scheduling by user level is realized by the user; the fourth-level scheduling according to the user service level is realized by the service of each user.
  • Hierarchical QoS implements QoS mechanisms such as bandwidth control and queue scheduling through layering. It can finely control the bandwidth allocation of each layer in complex networking and layered service models to achieve end-to-end service QoS.
  • . 3 is a schematic diagram of comparison between single-layer QoS and hierarchical QoS scheduling mechanism in the prior art. As shown in FIG. 3( a ), single-layer QoS scheduling refers to that only one scheduler is used for the entire QoS processing process, and only Complete a layer of QoS scheduling.
  • hierarchical QoS scheduling refers to the use of multiple schedulers for the entire QoS process, which can perform multi-layer QoS scheduling.
  • hierarchical QoS in Figure 3 (b) uses three schedulers. Two levels of scheduling have been completed.
  • TR-101 Second Edition
  • BBF International Standards Organization Broadband Forum
  • the traffic measurement, coloring, and supervision functions in the QoS function can be performed as instruction instruction in the middle of the OpenFlow switch packet processing flow, but the queuing and queue scheduling functions can only be reported in the OpenFlow switch.
  • the end of the text processing flow is executed as the action action, which makes the OpenFlow switch only implement port-based single-layer queue scheduling, and cannot implement multi-layer queue scheduling required by multi-layer QoS.
  • embodiments of the present invention provide a method, system, device, and computer storage medium for implementing QoS, which can implement multi-layer queue scheduling required by multi-layer QoS.
  • the technical solution of the embodiment of the present invention is implemented as follows:
  • An embodiment of the present invention provides a method for implementing quality of service QoS.
  • the method includes: a configuration point generates QoS configuration information according to a QoS policy, sends the QoS configuration information to a switch, and sends the QoS configuration information to a control. Or for the controller to query QoS configuration information;
  • the switch generates a multi-layer QoS configuration tree model according to the QoS configuration information; the controller provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and the flow table entry carrying the packet classification identifier Sent to the switch;
  • the switch performs a multi-layer QoS traffic management policy based on the flow table entry and the multi-layer QoS configuration tree model.
  • the configuration point generates the QoS configuration information according to the QoS policy, including: the configuration point forms a QOS policy according to the QoS requirement submitted by the controller or the application layer APP, allocates a traffic management resource for the QoS policy, and sets the resource to be hanged.
  • the connection relationship generates QoS configuration information according to the resource hook relationship.
  • the controller provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, including:
  • the controller generates a flow table entry according to the service requirement, provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and sets an associated flow queue identifier in an instruction action of the flow table entry.
  • the method before the configuration point generates the QoS configuration information according to the QoS policy, the method further includes:
  • the resource information of the traffic management TM sub-module of the switch is obtained through the Get-config message of the OF-Config protocol;
  • the resource information of the TM submodule includes at least one of the following information: a number of flow queues, and a number of schedulers.
  • the switch Preferably, the switch generates a multi-layer QoS configuration tree model according to the QoS configuration information, including:
  • the TM sub-module of the switch generates a multi-layer QoS configuration tree model based on the QoS configuration information.
  • the switch performs a multi-layer QoS traffic management policy according to the flow table entry and the multi-layer QoS configuration tree model, including:
  • the switch matches the input packet and the packet classification identifier of the flow table entry, and determines that the matching is successful, and obtains the associated flow queue identifier in the corresponding flow table entry, and enters the TM submodule according to the flow queue identifier.
  • the multi-layer QoS configuration tree model performs a QoS policy, performs traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion packet loss operations, and the input packet is output after the TM sub-module processes according to the QoS policy, or Cache, or discard.
  • the embodiment of the present invention further provides a method for implementing QoS, where the method includes: configuring a QoS configuration information according to a QoS policy, transmitting the QoS configuration information to a switch, and sending the QoS configuration information to a controller .
  • the configuration point generates the QoS configuration information according to the QoS policy, including: the configuration point forms a QOS policy according to the QoS requirement submitted by the controller or the application layer APP, allocates a traffic management resource for the QoS policy, and sets the resource.
  • the hook relationship generates QoS configuration information according to the resource hook relationship.
  • the method before the configuration point generates the QoS configuration information according to the QoS policy, the method further includes:
  • the resource information of the traffic management TM sub-module of the switch is obtained through the Get-config message of the OF-Config protocol;
  • the resource information of the TM sub-module includes at least one of the following information: a number of flow queues, a number of schedulers.
  • An embodiment of the present invention further provides a method for implementing QoS, where the method includes: The switch receives the QoS configuration information sent by the configuration point, and generates a multi-layer QoS configuration 4-pair model according to the QoS configuration information;
  • a multi-layer QoS traffic management policy is performed according to the flow table entry sent by the controller and the multi-layer QoS configuration tree model.
  • the generating the multi-layer QoS configuration tree model according to the QoS configuration information comprises: generating, by the traffic management TM sub-module of the switch, a multi-layer QoS configuration tree model according to the QoS configuration information.
  • the executing the multi-layer QoS traffic management policy according to the controller sending the flow table entry and the multi-layer QoS configuration tree model includes:
  • the switch matches the packet classification identifier of the input packet and the flow table entry, and determines that the matching is successful, and obtains the associated flow queue identifier in the corresponding flow table entry, and enters the TM sub-module according to the flow queue identifier.
  • the multi-layer QoS configuration tree model performs a QoS policy, performs traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion packet loss operations, and the input packet is output after the TM sub-module processes according to the QoS policy. , or cache, or discard.
  • the embodiment of the present invention further provides a method for implementing QoS, where the method includes: receiving, by the controller, QoS configuration information sent by a configuration point, or querying QoS configuration information from the configuration point;
  • the providing the QoS configuration information to a flow table entry having a corresponding QoS policy requirement includes:
  • the controller generates a flow table entry according to the service requirement, provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and sets an associated flow queue identifier in an instruction action of the flow table entry.
  • An embodiment of the present invention further provides a switch, where the switch includes: QoS configuration receiving a module, a QoS configuration mapping module, and a QoS policy execution module; wherein
  • the QoS configuration receiving module is configured to receive the QoS configuration information sent by the configuration point, and send the QoS configuration information to the QoS configuration mapping module, and configured to receive a flow table entry that is sent by the controller and carries the packet classification identifier. ;
  • the QoS configuration mapping module is configured to generate a multi-layer QoS configuration tree model corresponding to the QoS policy execution module according to the QoS configuration information;
  • the QoS policy execution module is configured to perform a multi-layer QoS traffic management policy according to the flow table entry of the carried message classification identifier received by the QoS configuration receiving module and the multi-layer QoS configuration tree model generated by the QoS configuration mapping module.
  • the QoS policy execution module includes: a flow table matching submodule, an identifier obtaining submodule, and a traffic management TM submodule; wherein
  • the flow table matching sub-module is configured to match a packet classification identifier of the input packet and a packet classification identifier of the flow table entry received by the receiving module;
  • the identifier acquisition sub-module is configured to: when the matching of the flow table matching sub-modules is consistent, perform an instruction action of the corresponding flow table entry, and obtain an associated flow queue identifier from the instruction action of setting the flow queue identifier, according to the
  • the flow queue identifier is configured to import the packet from the corresponding flow queue to the traffic management sub-module, where the flow queue identifier directly corresponds to a corresponding pre-generated QoS configuration tree model on the traffic management sub-module;
  • the traffic management sub-module is configured to perform a QoS policy according to a multi-layer QoS configuration tree model generated by the QoS configuration mapping module to perform traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion.
  • the packet loss operation determines that the input message is output, cached, or discarded from the specified interface.
  • the embodiment of the present invention further provides a configuration point, where the configuration point includes: a generating module and a sending module;
  • the generating module is configured to generate QoS configuration information according to the QoS policy;
  • the sending module is configured to send the QoS configuration information generated by the generating module to the switch, and configured to send the QoS configuration information generated by the generating module to the controller.
  • the configuration point further includes an obtaining module, configured to acquire resource information of the traffic management TM sub-module of the switch by using a Get-config message of the OF-Config protocol; the resource information of the TM sub-module includes the following information at least One: The number of stream queues, the number of schedulers.
  • the generating module is configured to form a QoS policy according to the QoS requirement submitted by the controller or the application layer APP, allocate a traffic management resource for the QoS policy, and set a hook relationship of the resource, according to the resource hook The relationship generates QoS configuration information.
  • the embodiment of the present invention further provides a controller, where the controller includes: an obtaining module, a flow table entry generating module, and a sending module;
  • the acquiring module is configured to receive the QoS configuration information sent by the configuration point, or query the configuration point for related QoS configuration information, and send the QoS configuration information to the flow table entry generation module; the flow table entry generation module, configured Providing the QoS configuration information to a flow table entry having a corresponding QoS policy requirement and generating a flow table entry;
  • the sending module is configured to send, to the switch, a flow table entry that carries the packet classification identifier generated by the flow table entry generating module.
  • the flow table entry generating module is configured to generate a flow table entry according to a service requirement, and provide the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and in an instruction action of the flow table entry Set the associated stream queue ID.
  • the embodiment of the present invention further provides a system for implementing QoS, where the system includes: a switch, a configuration point, and a controller;
  • the configuration point is configured to generate QoS configuration information according to the QoS policy, send the QoS configuration information to the switch, and send the QoS configuration information to the controller;
  • the controller is configured to receive QoS configuration information sent by the configuration point, or query QoS configuration information from the configuration point; and provide the QoS configuration information to a corresponding QoS policy And obtaining a flow table entry, and sending a flow table entry carrying the message classification identifier to the switch; the switch, configured to generate a multi-layer QoS configuration tree model according to the QoS configuration information sent by the configuration point; The flow table entry sent by the controller and the multi-layer QoS configuration tree model perform a multi-layer QoS traffic management policy.
  • the embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to perform QoS applied to a configuration point according to an embodiment of the present invention. method.
  • the embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to implement the QoS applied to the controller according to the embodiment of the present invention. method.
  • the embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the method for implementing QoS applied to a switch according to an embodiment of the present invention. .
  • the configuration point generates QoS configuration information according to the QoS requirement, sends the QoS configuration information to the switch, and sends the QoS configuration information to the controller.
  • the controller queries the relevant QoS configuration information;
  • the switch generates a multi-layer QoS configuration tree model according to the QoS configuration information;
  • the controller provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and Transmitting a flow table entry carrying a message classification identifier to the switch; the switch performing a multi-layer QoS traffic management policy according to the flow table entry and the multi-layer QoS configuration tree model.
  • the single-layer or multi-layer QoS configuration information is written into the TM sub-module, and a corresponding single-layer or multi-layer QoS configuration tree model is generated, which realizes more The multi-layer queue scheduling required by the layer QoS; and the flow table pipeline and the TM sub-module are separately implemented and processed in series, thereby avoiding the consumption of the high-speed bus bandwidth caused by the interleaving of the flow table pipeline and the TM sub-module.
  • FIG. 1 is a schematic diagram of a service flow processing process for implementing QoS in a network device in the prior art
  • FIG. 2 is a schematic diagram of a layered QoS hierarchical traffic management model in the prior art
  • Figure 3 (a) and Figure 3 (b) show a comparison of single-layer QoS and hierarchical QoS scheduling mechanisms in the prior art
  • FIG. 5 is a schematic diagram of a QoS configuration tree model according to Embodiment 1 of the present invention.
  • FIG. 6 is a schematic structural diagram of a system for implementing QoS according to Embodiment 1 of the present invention
  • FIG. 7 is a schematic flowchart of a method for implementing QoS according to Embodiment 2 of the present invention
  • FIG. 8 is a schematic structural diagram of a configuration point of a configuration point based on Embodiment 2 of the present invention.
  • FIG. 9 is a schematic flowchart of a method for implementing QoS according to Embodiment 3 of the present invention.
  • FIG. 10 is a schematic structural diagram of a switch according to Embodiment 3 of the present invention.
  • FIG. 11 is a schematic flowchart of a method for implementing QoS according to Embodiment 4 of the present invention.
  • FIG. 12 is a schematic structural diagram of a controller according to Embodiment 4 of the present invention.
  • FIG. 13 is a schematic diagram of a network topology of an application scenario for implementing multi-layer QoS according to Embodiment 5 of the present invention.
  • FIG. 14 is a schematic diagram of a configuration tree model of a multi-layer QoS application according to Embodiment 5 of the present invention.
  • FIG. 15 is a schematic diagram of a configuration tree model of a multi-layer QoS application according to Embodiment 6 of the present invention.
  • FIG. 4 is a schematic flowchart of a method for implementing QoS according to Embodiment 1 of the present invention. As shown in FIG. 4, the method includes the following steps:
  • Step 401 The configuration point generates QoS configuration information according to the QoS policy, sends the QoS configuration information to the switch, and sends the QoS configuration information to the controller or for query by the controller. QoS configuration information.
  • the method further includes:
  • the resource information of the traffic management (TM, Traffic Manage) module of the switch is obtained through the Get-config message of the OF-Config protocol;
  • the resource information of the TM submodule includes but is not limited to: a number of flow queues, and a number of schedulers.
  • the configuration point generates QoS configuration information according to the QoS policy, including:
  • the configuration point forms a QoS policy according to the QoS requirement submitted by the controller or the application layer (APP), allocates a traffic management resource to the QoS policy, sets a hook relationship of the resource, and generates QoS configuration information according to the resource hook relationship.
  • APP application layer
  • Step 402 The switch generates a multi-layer QoS configuration tree model according to the QoS configuration information.
  • the OpenFlow interface inside the switch sends the QoS configuration information to a TM sub-module downstream of the switch, and the TM sub-module generates a QoS configuration according to the QoS configuration information. Tree model.
  • FIG. 5 the QoS configuration tree model, converging branch node number scheduler (Scheduler ID) to identify another 1 J, a schematic view of a leaf node of the tree model QoS configured in the embodiment of the present invention.
  • FIG. 5 It is identified by the flow queue number (Flow-Queue ID).
  • Flow-Queue ID the flow queue number
  • the Scheduler ID is nested under the Scheduler ID and the scheduled Flow-Queue ID is carried in the Scheduler ID. For example, as shown in FIG.
  • the total scheduling of the ports of the scheduler 3 is to ensure that the bandwidth of each user when the port is congested, the flow queue 3 is attached to the high priority of the Scheduler 3; Scheduler 1) and Scheduler 2 (Scheduler 2) are respectively attached to the low priority of Scheduler 3, and Flow Queue 1 and Flow Queue 2 are respectively attached to the first queue of Scheduler 1 and On the second queue, the related parameters of the flow queue 1 and the flow queue 2 are allocated according to the service type or setting of the flow queue 1 and the flow queue 2; the flow queue 4 and the flow queue 5 are respectively connected to the first queue and the second of the Scheduler 2 On the queue, related parameters of the flow queue 4 and the flow queue 5 are allocated according to the service type or setting of the flow queue 4 and the flow queue 5.
  • Step 403 The controller sends the QoS configuration information to a flow table entry that has a corresponding QoS policy requirement, and sends a flow table entry that carries the packet classification identifier to the switch.
  • the controller provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, including:
  • the controller generates a flow table entry according to a specific service requirement, provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and sets an associated flow queue identifier in an instruction action of the flow table entry.
  • the flow queue identifier corresponds to the specified leaf node entry of the multi-layer QoS configuration tree, and specifically, the Flow-Queue ID is carried in the action (Action), and the Flow-Queue ID corresponds to the QoS of the first embodiment of the present invention. Configure the leaf nodes in the schematic of the tree model.
  • Table 1 is a schematic diagram of an OpenFlow flow table format according to Embodiment 1 of the present invention.
  • the flow table may include multiple Flow Table Entrys, including: Match Fields, Priorlity, Counters, Instructions, Timeouts, Status Messages, etc.
  • the matching field may be any field in the packet header, including: an address field, a user identifier field, a service priority field, a message classification identifier field, and the like.
  • Step 404 The switch performs a multi-layer QoS traffic management policy according to the flow table entry and the multi-layer QoS configuration tree model.
  • the switch performs according to the flow table entry and the multi-layer QoS configuration tree model Multi-layer QoS traffic management strategies, including:
  • the switch matches the input packet and the packet classification identifier of the flow table entry, and determines that the matching is successful, and obtains the associated flow queue identifier in the corresponding flow table entry, and enters the TM submodule according to the flow queue identifier.
  • Performing a QoS policy according to the multi-layer QoS configuration tree model performing traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion packet loss operation, where the input packet is processed by the TM submodule according to the QoS policy Output, or cache, or discard.
  • the subsequent instruction action (Action) is executed, and the action is set in the instruction action.
  • the Flow-Queue ID a multi-layer QoS traffic management policy is executed in the TM sub-module according to the set Flow-Queue ID, and the packet is output according to the setting of the multi-layer QoS configuration tree model; or when congestion occurs.
  • the traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion and packet loss operations are performed according to the packet loss policy of the scheduler and/or the flow queue, and the packet loss is performed according to the priority from low to high. Discard, or cache, or output.
  • FIG. 6 is a schematic structural diagram of a system for implementing QoS in an OpenFlow network according to Embodiment 1 of the present invention.
  • the system includes: The configuration point 20 is configured to generate QoS configuration information according to the QoS policy, send the QoS configuration information to the switch 10, and send the QoS configuration information to the controller. 30;
  • the controller 30 is configured to receive QoS configuration information sent by the configuration point 20, or query QoS configuration information from the configuration point 20; and provide the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, And sending a flow table entry carrying the message classification identifier to the switch 10;
  • the switch 10 is configured to generate a multi-layer QoS configuration tree model according to the QoS configuration information sent by the configuration point 20; perform multi-layer QoS traffic according to the flow table entry sent by the controller and the multi-layer QoS configuration tree model Management strategy.
  • the configuration point 20 is configured to form a QOS policy according to the QoS requirement submitted by the controller or the application layer APP, allocate a traffic management resource to the QoS policy, and set a hook relationship of the resource, according to the resource The connection generates QoS configuration information.
  • the configuration point 20 is further configured to acquire the resource information of the TM submodule of the switch 10 by using a Get-config message of the OF-Config protocol before establishing the QoS configuration information according to the QoS policy. ;
  • the resource information of the TM submodule includes at least one of the following information: a number of flow queues, and a number of schedulers.
  • the TM submodule of the switch 10 generates a multi-layer QoS configuration tree model according to the QoS configuration information.
  • the switch 10 is configured to match the packet classification identifier of the input packet and the flow table entry, and after determining that the matching is successful, obtain the associated flow queue identifier in the corresponding flow table entry, according to the flow queue identifier.
  • the module processes, outputs, or caches, or discards according to the QoS policy.
  • the controller 30 is configured to generate a flow table entry according to the service requirement, provide the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and set the command action in the flow table entry.
  • the associated stream queue ID is configured to generate a flow table entry according to the service requirement, provide the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and set the command action in the flow table entry.
  • FIG. 7 is a schematic flowchart of a method for implementing QoS according to Embodiment 2 of the present invention. As shown in FIG. 7, the method includes:
  • Step 701 The configuration point generates QoS configuration information according to the QoS policy.
  • the configuration point generates QoS configuration information according to the QoS policy, including:
  • the configuration point forms a QOS policy according to a QoS requirement submitted by a controller or an application layer (APP)
  • APP application layer
  • the traffic management resource is allocated to the QoS policy and the connection relationship of the resource is set, and the QoS configuration information is generated according to the resource connection relationship.
  • the method further includes:
  • the resource information of the traffic management TM sub-module of the switch is obtained through the Get-config message of the OF-Config protocol;
  • the resource information of the TM sub-module includes at least one of the following information: a number of flow queues, a number of schedulers.
  • Step 702 Send the QoS configuration information to the switch, and send the QoS configuration information to the controller.
  • the embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are configured to perform the method for implementing QoS according to the embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a configuration point of the second embodiment according to the present invention.
  • the configuration point 20 includes: a generating module 21 and sending Module 22; wherein
  • the generating module 21 is configured to generate QoS configuration information according to the QoS policy
  • the sending module 22 is configured to send the QoS configuration information generated by the generating module 21 to the switch 10; and configured to send the QoS configuration information generated by the generating module 21 to the controller 30.
  • the configuration point 20 further includes an obtaining module 23 configured to acquire resource information of the TM submodule of the switch 10 by using a Get-config message of the OF-Config protocol; the resource information of the TM submodule 133 includes the following: At least one of the information: the number of stream queues, the number of schedulers.
  • the generating module 21 is configured to form a QoS policy according to the requirements submitted by the controller 30 or the APP, allocate traffic management resources for the QoS policy, and set the hook of the resource. Relationship, generating QoS configuration information according to the resource hook relationship.
  • the generating module 21 and the obtaining module 23 may be used by a central processing unit (CPU), a digital signal processor (DSP) or a digital signal processor (DSP) in the actual application.
  • CPU central processing unit
  • DSP digital signal processor
  • DSP digital signal processor
  • FPGA Field-Programmable Gate Array
  • the sending module 22 can be implemented by a transmitter in the configuration point.
  • FIG. 9 is a schematic flowchart of a method for implementing QoS according to Embodiment 3 of the present invention. As shown in FIG. 9, the method includes:
  • Step 901 The switch receives the QoS configuration information sent by the configuration point, and generates a multi-layer QoS configuration tree model according to the QoS configuration information.
  • the generating the multi-layer QoS configuration tree model according to the QoS configuration information comprises: generating, by the traffic management TM sub-module of the switch, a multi-layer QoS configuration tree model according to the QoS configuration information.
  • the OpenFlow interface in the switch sends the QoS configuration information to a TM sub-module downstream of the switch, and the TM sub-module generates a QoS according to the QoS configuration information.
  • Step 902 Perform a multi-layer QoS traffic management policy according to the flow table entry sent by the controller and the multi-layer QoS configuration tree model.
  • the performing according to the controller sending the flow table entry and the multi-layer QoS configuration tree model Multi-layer QoS traffic management strategies including:
  • the switch matches the packet classification identifier of the input packet and the flow table entry, and determines that the matching is successful, and obtains the associated flow queue identifier in the corresponding flow table entry, and enters the TM sub-module according to the flow queue identifier.
  • the multi-layer QoS configuration tree model performs a QoS policy, performs traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion packet loss operations, and the input packet is output after the TM sub-module processes according to the QoS policy. , or cache, or discard.
  • the embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the method for implementing QoS according to the embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of a switch based on the third implementation of the present invention; as shown in FIG. 10, the switch 10 includes: a QoS configuration receiving module 11, and a QoS configuration. a mapping module 12 and a QoS policy execution module 13; wherein
  • the QoS configuration receiving module 11 is configured to receive the QoS configuration information sent by the configuration point 20, and send the QoS configuration information to the QoS configuration mapping module 12;
  • the flow table entry; the QoS configuration receiving module 11 is generally a processing module of the OF-Config protocol and the OpenFlow protocol;
  • the QoS configuration mapping module 12 is configured to generate a multi-layer QoS configuration tree model corresponding to the QoS policy execution module 13 according to the QoS configuration information; the QoS configuration mapping module 12 functions as a QoS configuration receiving module 11 and a QoS policy.
  • the QoS policy execution module 13 is configured to: according to the flow table entry carrying the message classification identifier received by the QoS configuration receiving module 11 and the multi-layer QoS generated by the QoS configuration mapping module 12
  • the configuration tree model implements a multi-layer QoS traffic management policy.
  • the QoS policy execution module 13 can be a dedicated hardware module and chip.
  • the QoS policy execution module 13 includes: a flow table matching submodule 131, an identifier Obtaining a submodule 132, a TM submodule 133; wherein
  • the flow table matching sub-module 131 is configured to match the packet classification identifier of the input packet with the packet classification identifier of the flow table entry received by the receiving module 11;
  • the identifier acquisition sub-module 132 is configured to: when the flow table matching sub-module 131 matches, perform an instruction action of the corresponding flow table entry, and obtain an associated flow queue identifier from the instruction action of the set flow queue identifier, Importing the packet from the corresponding flow queue to the TM sub-module 133 according to the flow queue identifier, where the flow queue identifier directly corresponds to the pre-generated QoS configuration tree model on the TM sub-module 133;
  • the TM sub-module 133 is configured to perform a QoS policy according to the multi-layer QoS configuration tree model generated by the QoS configuration mapping module 12, and perform traffic rate limiting and/or traffic scheduling and/or traffic shaping and/or traffic congestion loss. Packet operation, and determines that the input message is output, cached, or discarded from the specified interface.
  • the QoS configuration mapping module 12 and the QoS policy execution module 13 may be implemented by a CPU, a DSP, or an FPGA in the switch in an actual application; the QoS configuration receiving module 11 may be implemented by the switch in an actual application. Receiver implementation in .
  • FIG. 11 is a schematic flowchart of a method for implementing QoS according to Embodiment 4 of the present invention. As shown in FIG. 11, the method includes:
  • Step 1101 The controller receives the QoS configuration information sent by the configuration point, or queries the QoS configuration information from the configuration point.
  • Step 1102 The QoS configuration information is provided to a flow table entry having a corresponding QoS policy requirement, and the flow table entry carrying the packet classification identifier is sent to the switch.
  • the providing the QoS configuration information to the flow table entry having the corresponding QoS policy requirement includes:
  • the controller generates a flow table entry according to the service requirement, provides the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and sets an associated flow queue identifier in an instruction action of the flow table entry.
  • the embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the method for implementing QoS according to the embodiment of the present invention.
  • FIG. 12 is a schematic structural diagram of a controller according to the fourth embodiment of the present invention.
  • the controller 30 includes: an obtaining module 31, a flow table entry generating module 32 and a sending module 33; wherein
  • the obtaining module 31 is configured to receive the QoS configuration information sent by the configuration point 20, or query the configuration point 20 for the relevant QoS configuration information, and send the QoS configuration information to the flow table entry generation module 32;
  • the flow table entry generating module 32 is configured to provide the QoS configuration information to a flow table entry having a corresponding QoS policy requirement and generate a flow table entry;
  • the sending module 33 is configured to send the flow table entry carrying the packet classification identifier generated by the flow table entry generating module 32 to the switch 10.
  • the flow table entry generating module 32 is configured to generate a flow table entry according to a service requirement, provide the QoS configuration information to a flow table entry having a corresponding QoS policy requirement, and perform an action action on the flow table entry. Set the associated stream queue ID in .
  • processing units in the controller of the embodiment of the present invention can be understood by referring to the foregoing description of the method for implementing QoS, and the control of the embodiment of the present invention
  • the processing units in the device may be implemented by an analog circuit that implements the functions described in the embodiments of the present invention, or may be implemented by running software running the functions described in the embodiments of the present invention on the smart terminal.
  • the flow table entry generating module 32 may be implemented by a CPU, a DSP or an FPGA in the controller in an actual application; the sending module 33 may be implemented by a transmitter in the controller in an actual application; The acquisition module 31 may be implemented by a receiver in the controller in an actual application, or may be implemented by a transceiver in the controller in combination with a CPU, a DSP, or an FPGA.
  • the controller, the switch, and the configuration point may be implemented by a network device that implements QoS services in practical applications.
  • FIG. 13 is a schematic diagram of a network topology of an application scenario for implementing multi-layer QoS according to Embodiment 5 of the present invention. As shown in FIG. 13, interface 1 and interface 2 are input terminals for accessing with user terminals; interface 3 is output. The interface bandwidth is 1G.
  • FIG. 14 is a schematic diagram of a configuration tree model of a multi-layer QoS application according to Embodiment 5 of the present invention.
  • interface 1 and interface 2 access a total of four users, namely: user l (userl), user 2 ( user2 ), user 3 ( user3 ), and user 4 ( user4 ).
  • the QoS requirements of the four users are as follows:
  • the maximum bandwidth is 500M
  • User4 guarantees a rate of 100M and a burst rate of 150M.
  • the method for implementing QoS includes the following steps:
  • Step 1 After the configuration point and the switch establish a link, the Get-config of the OF-Config protocol is adopted.
  • the message acquires resource information of the TM sub-module of the switch, and the resource information of the TM sub-module includes the number of flow queues, the number of various types of schedulers, and the like.
  • the configuration point acquires resource information of the TM sub-module to facilitate subsequent generation of QoS configuration information.
  • Step 2 The configuration point allocates the traffic management resource according to the QoS requirements of the four users submitted by the controller or the APP, and sets the connection relationship of the resource, and generates according to the resource hook relationship.
  • the QoS configuration information is sent to the switch and the controller respectively.
  • the scheduler 1 is SPQ type.
  • Flow-Queue has a speed limit of 500M and is attached to the Scheduler high priority.
  • the CIR parameter of the Flow-Queue of User2 is set to 300M, and is attached to the first queue of the Schedule.
  • the weight parameter is set to 60%.
  • the CIR parameter of the Flow-Queue of User3 is set to 100M, and the PI parameter is configured to 150M, which is attached to the first queue of Scheduled.
  • the CIR parameter of the Flow-Queue of User4 is set to 100M, and the PI parameter is configured to 150M, which is connected to the second queue of Scheduled.
  • Step 3 The switch receives the QoS configuration information of the four users, and forms a QoS configuration 4-pair model on the TM sub-module.
  • Step 4 The controller generates a flow table entry according to the characteristics of the four users, and sends a flow table entry carrying the packet classification identifier to the switch.
  • the server sends the flow-entry, the Flow-Queue is carried in the action (Action). ID.
  • Step 5 The packet entering the switch first matches the packet classification identifier of the packet and the flow table entry. After the matching succeeds, the subsequent instruction action (Action) is executed, and when the instruction action has "Set Flow-Queue" In the ID field, the multi-layer QoS traffic management policy is executed in the TM sub-module according to the set Flow-Queue ID, and the packet output is performed according to the setting of the multi-layer QoS configuration tree model; when congestion occurs, According to the packet loss policy of the scheduler and/or the flow queue, the user packets of the lower-to-high priority are discarded, cached, and outputted on the interface Int3.
  • Multi-layer QoS needs to be completed on the switch to implement each user. Two-layer scheduling between service traffic of different service levels and service traffic of different users, and fair scheduling of traffic between different users.
  • FIG. 15 is a schematic diagram of a configuration tree model of a multi-layer QoS application according to Embodiment 3 of the present invention.
  • the method for implementing the QoS in the OpenFlow network includes the following steps: Step 1: After the configuration point and the switch are established, the resource information of the TM submodule of the switch is obtained by using the Get-config message of the OF-Config protocol, where The resource information of the TM sub-module includes the number of flow queues, the number of various types of schedulers, and the like.
  • the configuration point acquires resource information of the TM sub-module to facilitate subsequent generation of QoS configuration information.
  • Step 2 The configuration point allocates the traffic management resource according to the QoS requirement of the user submitted by the controller or the APP, and sets the connection relationship of the resource, generates QoS configuration information according to the resource connection relationship, and sends the configuration information separately. Give the switch and controller.
  • Step 3 The switch receives the QoS configuration information of the four flows of the two users, and forms a QoS configuration tree model on the TM submodule.
  • Step 4 The controller generates a flow table entry according to the characteristics of the four flows of the two users, and sends a flow table entry carrying the packet classification identifier to the switch; when the flow table entry is sent, the action is performed (Action) Carry the Flow-Queue ID field.
  • Step 5 The packet entering the switch first matches the packet classification identifier of the packet and the flow table entry. After the matching succeeds, the subsequent instruction action (Action) is executed, and when the instruction action has "Set Flow-Queue" When ID", according to the set Flow-Queue ID, a multi-layer QoS traffic management policy is executed in the TM sub-module, and the packet is output according to the setting of the multi-layer QoS configuration tree model; when congestion occurs, according to The packet loss policy of the scheduler and/or the flow queue is processed, and the user packets with low priority to high priority are discarded, cached, or outputted on the interface Int3.
  • the present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus, and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • the computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor, or other programmable data processing device to produce a machine such that a process or a process and/or a block diagram of a block or A device that has multiple functions specified in the box.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
  • the existing OF-Config protocol is extended, and the configuration information of the single layer or multiple layers of QoS is written into the TM sub-module to generate a corresponding single-layer or multi-layer QoS configuration tree model, which realizes the requirements of the multi-layer QoS.
  • Multi-layer queue scheduling; and, the flow table pipeline and the TM sub-module are separately implemented and processed in series, which avoids the consumption of high-speed bus bandwidth caused by the interleaving of the flow table pipeline and the TM sub-module.

Landscapes

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

Abstract

本发明实施例公开了一种实现服务质量QoS的方法、系统、设备及计算机存储介质;其中,所述方法包括:配置点根据QoS需求生成QoS配置信息,将QoS配置信息发送给交换机,并将QoS配置信息发送给控制器或供控制器查询相关QoS配置信息;交换机根据QoS配置信息生成多层QoS 配置树模型;控制器将所述QoS配置信息提供给有相应QoS策略需求的流表条目,并将携带报文分类标识的流表条目发送给交换机;交换机根据流表条目和所述多层QoS配置树模型执行多层QoS流量管理策略。

Description

实现 QoS的方法、 系统、 设备及计算机存储介质 技术领域
本发明涉及通信技术, 具体涉及一种实现 QoS的方法、 系统、 设备及 计算机存储介质。 背景技术
在当今的数据通信网络中, 不管是局域网、 城域网还是广域网, 服务 质量(QoS, Quality of Service )机制都是非常重要并且常用的。 网络的 QoS 机制是指网络针对各种业务应用的不同需求, 为其提供不同的服务质量保 证, 如带宽、 丟包率、 延迟和延迟抖动等, 以实现同时承载数据、 语音和 视频等多种业务的综合网络。 图 1为现有技术中网络设备实现 QoS的业务 流量处理流程示意图, 如图 1所示, 现有技术中网络的 QoS主要包括以下 四个步骤: 一、 流分类和流标记。 流是一组具有相同特性的数据报文, 流 分类就是按照一定的流划分规则(例如基于端口或基于报文的某些字段值 ) 对进入网络的业务流量进行分类。 流标记就是给流分类后的业务报文设置 网络内的服务等级和优先级标记, 以实现不同业务流的 QoS区分。 二、 流 量测量、 着色和监管。 该步骤就是在完成流分类和流标记之后, 对业务流 进行速率测量(Metering ), 同时根据流量的到达速率、 配置的保证通过速 率( CI , Committed Information Rate )和最大允许速率( PIR, Peak Information Rate )给输入的报文进行着色和速率限制。 具体来说, 报文着色和速率限制 通常会釆用互联网工程任务组(IETF, Internet Engineering Task Force ) 的 标准 RFC 2698所规定的双速率三色标记 ( TrTCM, Two rate Three Color Marker )算法, 该算法将不超过 CIR的报文标记为绿色, 将超过 CIR而不 超过 PIR的报文标记为黄色, 将超过 PIR的报文标记为红色, 再通过丟弃 红色报文实现业务流的速率限制。 三、 队列拥塞管理和调度。 经过流量测 量、 着色和监管的业务流量会进入不同等级的緩存队列, 这个过程被称为 排队(Queuing )。 队列拥塞管理就是监控队列緩存, 通过尾丟弃(TD, Tail Drop )、 随机早期探测 (RED, Random Early Detection )或加权随机早期探 测 (WRED, Weighted Random Early Detection )等报文丟弃方式(也被称 为拥塞策略)来预见并避免拥塞的发生。 队列调度 ( Scheduling )就是从代 出队列中, 即针对不同队列所緩存报文的出队顺序进行调度, 以最大程度 避免拥塞和满足高优先级业务流量的出队, 常用的队列调度算法包括: 严 格优先级队列( SPQ, Strict Priority Queue )调度算法、加权公平队列( WFQ, Weighted Fair Queue )调度算法和差额加权轮询 ( DWRR, Deficit Weighted Round Robin )调度算法等。 四、 流量整形和优先级重写。 经过队列调度后 的流量一般会通过漏桶机制实现流量整形, 流量整形与流量监管一样完成 流量速率限制的功能, 二者的区别在于流量整形不会像流量监管一样对于 超出限制速率的流量直接丟弃, 而是会先将这些流量緩存起来, 并在合适 的时候再将緩存的报文发送出去。 这使得流量能保持一个比较均勾的速率, 有助于降低下游网络节点由于突发流量导致丟包的概率。 在业务流量从出 端口发送出去之前, 除了完成流量整形, 还可以对报文的优先级进行重写, 以影响下游网络节点的 QoS处理。
网络的 QoS功能既可以用于实现单层 QoS,也可以用于实现层次化 QoS ( HQoS, Hierarchical QoS )。 单层 QoS只能基于端口进行业务带宽控制和 流量管理, 而层次化 QoS则可以实现。 图 2为现有技术中层次化 QoS分层 流量管理模型示意图, 如图 2所示, 层次化 QoS实现了端口级、 用户组级、 用户级和业务级的四级流量管理, 通过物理端口实现了按端口级别的第一 级调度; 通过用户组实现了按用户组级别的第二级调度; 通过用户实现了 按用户级别的第三级调度; 通过每个用户的业务实现了按用户业务级别的 第四级调度。 层次化 QoS通过分层实现带宽控制和队列调度等 QoS机制, 可以在复杂的组网和分层的业务模型中, 对每一层的带宽分配进行精细控 制, 达到保证端到端业务 QoS的目的。 图 3为现有技术中单层 QoS和层次 化 QoS调度机制对比示意图, 如图 3 ( a )所示, 单层 QoS调度指的是整个 QoS处理过程只使用一个调度器( Scheduler ), 只能完成一层 QoS调度。 如 图 3 ( b )所示,层次化 QoS调度指的是整个 QoS处理过程使用多个调度器, 可以完成多层 QoS调度, 例如图 3 ( b ) 中层次化 QoS使用了三个调度器, 完成了两层调度。 基于上述层次化 QoS具备相对于单层 QoS的明显技术优 势, 在国际标准组织宽带论坛( BBF, Broadband Forum )于 2011年 7月发 布的 TR-101 (第二版)标准中, 明确提出了要求网络设备支持层次化 QoS 的能力, 而且目前已经有越来越多的商用网络设备能够支持层次化 QoS的 能力。
但是,根据目前 OpenFlow交换机规范的规定, QoS功能中的流量测量、 着色和监管功能都可以在 OpenFlow 交换机报文处理流程的中间作为指令 Instruction执行, 但排队和队列调度功能都只能在 OpenFlow交换机报文处 理流程的最后作为动作 Action执行, 这就使得 OpenFlow交换机只能实现 基于端口的单层队列调度, 无法实现多层 QoS所要求的多层队列调度。
针对上述 OpenFlow交换机无法实现多层 QoS的问题, 目前尚未有现 有技术方案可供参考。 发明内容
为解决现有存在的技术问题, 本发明实施例提供了一种实现 QoS的方 法、 系统、 设备及计算机存储介质, 能够实现多层 QoS所要求的多层队列 调度。 为达到上述目的, 本发明实施例的技术方案是这样实现的:
本发明实施例提供了一种实现服务质量 QoS的方法; 所述方法包括: 配置点根据 QoS策略生成 QoS配置信息,将所述 QoS配置信息发送给 交换机, 并将所述 QoS配置信息发送给控制器或供控制器查询 QoS配置信 息;
所述交换机根据所述 QoS配置信息生成多层 QoS配置树模型; 所述控制器将所述 QoS配置信息提供给有相应 QoS策略需求的流表条 目, 并将携带报文分类标识的流表条目发送给所述交换机;
所述交换机根据所述流表条目和所述多层 QoS 配置树模型执行多层 QoS流量管理策略。
优选地, 所述配置点根据 QoS策略生成 QoS配置信息, 包括: 配置点根据控制器或应用层 APP提交的 QoS需求形成 QOS策略, 为 所述 QoS策略分配流量管理资源并设置所述资源的挂接关系, 根据所述资 源挂接关系生成 QoS配置信息。
优选地, 所述控制器将所述 QoS配置信息提供给有相应 QoS策略需求 的流表条目, 包括:
所述控制器根据业务需求生成流表条目, 将所述 QoS配置信息提供给 有相应 QoS策略需求的流表条目, 并在所述流表条目的指令动作中设定关 联的流队列标识。
优选地, 所述配置点根据 QoS策略生成 QoS配置信息之前, 所述方法 还包括:
配置点与交换机建立链接后,通过 OF-Config协议的 Get-config消息获 取所述交换机的流量管理 TM子模块的资源信息;
所述 TM子模块的资源信息包括以下信息至少之一: 流队列数量、 调 度器数量。 优选地,所述交换机根据所述 QoS配置信息生成多层 QoS配置树模型, 包括:
所述交换机的 TM子模块根据所述 QoS配置信息生成多层 QoS配置树 模型。
优选地, 所述交换机根据所述流表条目和所述多层 QoS配置树模型执 行多层 QoS流量管理策略, 包括:
交换机匹配输入报文与所述流表条目的报文分类标识, 确定匹配成功 后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标识进入 所述 TM子模块按照所述多层 QoS配置树模型执行 QoS策略, 进行流量限 速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作, 所述输入报文在 TM子模块根据 QoS策略处理后输出、 或緩存、 或丟弃。
本发明实施例还提供了一种实现 QoS的方法, 所述方法包括: 配置点根据 QoS策略生成 QoS配置信息,将所述 QoS配置信息发送给 交换机, 并将所述 QoS配置信息发送给控制器。
优选地, 所述配置点根据 QoS策略生成 QoS配置信息, 包括: 所述配置点根据控制器或应用层 APP提交的 QoS需求形成 QOS策略, 为所述 QoS策略分配流量管理资源并设置所述资源的挂接关系, 根据所述 资源挂接关系生成 QoS配置信息。
优选地, 所述配置点根据 QoS策略生成 QoS配置信息之前, 所述方法 还包括:
配置点与交换机建立链接后,通过 OF-Config协议的 Get-config消息获 取所述交换机的流量管理 TM子模块的资源信息;
所述 TM子模块的资源信息包括以下信息至少之一: 流队列数量、 调 度器数量。
本发明实施例还提供了一种实现 QoS的方法, 所述方法包括: 交换机接收到配置点发送的 QoS配置信息, 根据所述 QoS配置信息生 成多层 QoS配置 4对模型;
根据控制器发送的流表条目和所述多层 QoS配置树模型执行多层 QoS 流量管理策略。
优选地,所述根据所述 QoS配置信息生成多层 QoS配置树模型,包括: 所述交换机的流量管理 TM子模块根据所述 QoS 配置信息生成多层 QoS配置树模型。
优选地, 所述根据控制器发送流表条目和所述多层 QoS配置树模型执 行多层 QoS流量管理策略, 包括:
所述交换机匹配输入报文与所述流表条目的报文分类标识, 确定匹配 成功后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标识 进入所述 TM子模块按照所述多层 QoS配置树模型执行 QoS策略, 进行流 量限速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作,所述输入报文 在 TM子模块根据 QoS策略处理后输出、 或緩存、 或丟弃。
本发明实施例还提供了一种实现 QoS的方法, 所述方法包括: 控制器接收配置点发送的 QoS配置信息, 或从所述配置点查询 QoS配 置信息;
将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目, 并将携 带报文分类标识的流表条目发送给所述交换机。
优选地, 所述将所述 QoS配置信息提供给有相应 QoS策略需求的流表 条目, 包括:
所述控制器根据业务需求生成流表条目, 将所述 QoS配置信息提供给 有相应 QoS策略需求的流表条目, 并在所述流表条目的指令动作中设定关 联的流队列标识。
本发明实施例还提供了一种交换机, 所述交换机包括: QoS 配置接收 模块、 QoS配置映射模块和 QoS策略执行模块; 其中,
所述 QoS配置接收模块, 配置为接收配置点发送的 QoS配置信息, 将 所述 QoS配置信息发送给所述 QoS配置映射模块; 还配置为接收控制器发 送的携带报文分类标识的流表条目;
所述 QoS配置映射模块, 配置为才艮据所述 QoS配置信息在对应的 QoS 策略执行模块上对应生成多层 QoS配置树模型;
所述 QoS策略执行模块, 配置为根据所述 QoS配置接收模块接收的携 带报文分类标识的流表条目和所述 QoS配置映射模块生成的多层 QoS配置 树模型执行多层 QoS流量管理策略。
优选地, 所述 QoS策略执行模块包括: 流表匹配子模块、 标识获取子 模块和流量管理 TM子模块; 其中,
所述流表匹配子模块, 配置为匹配输入报文的报文分类标识与所述接 收模块接收的流表条目的报文分类标识;
所述标识获取子模块, 配置为当所述流表匹配子模块匹配一致后, 执 行相应的流表条目的指令动作, 并从设置流队列标识的指令动作中获取关 联的流队列标识, 根据所述流队列标识将报文从相应的流队列导入到流量 管理子模块, 所述流队列标识直接对应流量管理子模块上的相应的预先生 成的 QoS配置树模型;
所述流量管理子模块, 配置为 4艮据所述 QoS配置映射模块预先生成的 多层 QoS配置树模型执行 QoS策略, 进行流量限速和 /或流量调度和 /或流 量整形和 /或流量拥塞丟包操作, 并决定所述输入报文从指定接口输出、 或 緩存、 或丟弃。
本发明实施例还提供了一种配置点, 所述配置点包括: 生成模块和发 送模块; 其中,
所述生成模块, 配置为根据 QoS策略生成 QoS配置信息; 所述发送模块, 配置为将所述生成模块生成的 QoS配置信息发送给交 换机; 配置为将所述生成模块生成的 QoS配置信息发送给控制器。
优选地, 所述配置点还包括获取模块, 配置为通过 OF-Config协议的 Get-config 消息获取所述交换机的流量管理 TM子模块的资源信息; 所述 TM子模块的资源信息包括以下信息至少之一: 流队列数量、 调度器数量。
优选地, 所述生成模块, 配置为根据控制器或应用层 APP提交的 QoS 需求形成 QoS策略, 为所述 QoS策略分配流量管理资源并设置所述资源的 挂接关系, 根据所述资源挂接关系生成 QoS配置信息。
本发明实施例还提供了一种控制器, 所述控制器包括: 获取模块、 流 表条目生成模块和发送模块; 其中,
所述获取模块, 配置为接收配置点发送的 QoS配置信息, 或向配置点 查询相关的 QoS配置信息,将所述 QoS配置信息发送给流表条目生成模块; 所述流表条目生成模块, 配置为将所述 QoS 配置信息提供给有相应 QoS策略需求的流表条目并生成流表条目;
所述发送模块, 配置为将所述流表条目生成模块生成的携带报文分类 标识的流表条目发送给交换机。
优选地, 所述流表条目生成模块, 配置为根据业务需求生成流表条目, 将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目, 并在所述流 表条目的指令动作中设定关联的流队列标识。
本发明实施例还提供了一种实现 QoS 的系统, 所述系统包括: 交换 机、 配置点和控制器; 其中,
所述配置点, 配置为根据 QoS策略生成 QoS配置信息, 将所述 QoS 配置信息发送给交换机, 并将所述 QoS配置信息发送给控制器;
所述控制器, 配置为接收所述配置点发送的 QoS配置信息, 或从所述 配置点查询 QoS配置信息;将所述 QoS配置信息提供给有相应 QoS策略需 求的流表条目, 并将携带报文分类标识的流表条目发送给所述交换机; 所述交换机, 配置为根据所述配置点发送的 QoS 配置信息生成多层 QoS配置树模型; 根据所述控制器发送的流表条目和所述多层 QoS配置树 模型执行多层 QoS流量管理策略。
本发明实施例还提供了一种计算机存储介质, 所述计算机存储介质中 存储有计算机可执行指令, 所述计算机可执行指令用于执行本发明实施例 所述的应用于配置点的实现 QoS的方法。
本发明实施例还提供了一种计算机存储介质, 所述计算机存储介质中 存储有计算机可执行指令, 所述计算机可执行指令用于执行本发明实施例 所述的应用于控制器的实现 QoS的方法。
本发明实施例还提供了一种计算机存储介质, 所述计算机存储介质中 存储有计算机可执行指令, 所述计算机可执行指令用于执行本发明实施例 所述的应用于交换机的实现 QoS的方法。
本发明实施例提供的实现 QoS的方法、 系统、设备及计算机存储介质, 配置点根据 QoS需求生成 QoS配置信息,将所述 QoS配置信息发送给交换 机, 并将所述 QoS配置信息发送给控制器或供控制器查询相关 QoS配置信 息; 所述交换机根据所述 QoS配置信息生成多层 QoS配置树模型; 所述控 制器将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目, 并将携 带报文分类标识的流表条目发送给所述交换机; 所述交换机根据所述流表 条目和所述多层 QoS配置树模型执行多层 QoS流量管理策略。 釆用本发明 的技术方案, 通过扩展现有的 OF-Config协议, 将单层或多层 QoS的配置 信息写入 TM子模块, 生成相应的单层或多层 QoS配置树模型, 实现了多 层 QoS所要求的多层队列调度; 并且, 将流表管道和 TM子模块分开实现, 串联处理, 避免了流表管道和 TM子模块交织导致的高速总线带宽的消耗。 附图说明
图 1为现有技术中网络设备实现 QoS的业务流量处理流程示意图; 图 2为现有技术中层次化 QoS分层流量管理模型示意图;
图 3 ( a )和图 3 ( b ) 为现有技术中单层 QoS和层次化 QoS调度机制 对比示意图;
图 4为本发明实施例一的实现 QoS的方法流程示意图;
图 5为本发明实施例一的 QoS配置树模型的示意图;
图 6为本发明基于实施例一的实现 QoS的系统的组成结构示意图; 图 7为本发明实施例二的实现 QoS的方法流程示意图;
图 8为本发明基于实施二的配置点的组成结构示意图;
图 9为本发明实施例三的实现 QoS的方法流程示意图;
图 10为本发明基于实施例三的交换机的组成结构示意图;
图 11为本发明实施例四的实现 QoS的方法流程示意图;
图 12为本发明基于实施例四的控制器的组成结构示意图;
图 13 为本发明实施例五的实现多层 QoS的应用场景的网络拓朴示意 图;
图 14为本发明实施例五的多层 QoS应用的配置树模型的示意图; 图 15为本发明实施例六的多层 QoS应用的配置树模型的示意图。 具体实施方式
下面结合附图及具体实施例对本发明作进一步详细的说明。
实施例一
图 4为本发明实施例一的实现 QoS的方法流程示意图, 如图 4所示, 包括以下步骤:
步骤 401 : 配置点根据 QoS策略生成 QoS配置信息, 将所述 QoS配置 信息发送给交换机, 并将所述 QoS配置信息发送给控制器或供控制器查询 QoS配置信息。
这里, 所述配置点根据 QoS策略生成 QoS配置信息之前, 所述方法还 包括:
所述配置点和交换机建链后,通过 OF-Config协议的 Get-config消息获 取所述交换机的流量管理(TM, Traffic Manage )模块的资源信息;
其中, 所述 TM子模块的资源信息包括但不限于: 流队列数量、 调度 器数量。
这里, 所述配置点根据 QoS策略生成 QoS配置信息, 包括:
配置点根据控制器或应用层 ( APP )提交的 QoS需求形成 QoS策略, 为所述 QoS策略分配流量管理资源并设置所述资源的挂接关系, 根据所述 资源挂接关系生成 QoS配置信息。
步骤 402: 所述交换机根据所述 QoS配置信息生成多层 QoS配置树模 型。
这里, 所述交换机内部的 OpenFlow接口接收到所述 QoS配置信息后, 将所述 QoS配置信息发送至所述交换机内部下游的 TM子模块, 所述 TM 子模块根据所述 QoS配置信息生成 QoS配置树模型。
图 5为本发明实施例一的 QoS配置树模型的示意图, 如图 5所示, 所 述 QoS配置树模型中, 汇聚的树枝节点以调度器号码( Scheduler ID )来识 另1 J, 叶子节点以流队列号码(Flow-Queue ID )来识别。 多层 QoS配置树模 型在 OF-Config协议中下发时表现为在 Scheduler ID下嵌套 Scheduler ID, 并在 Scheduler ID下携带有所调度的 Flow-Queue ID, 由此实现多层 QoS功 能。 例如图 5中所示的, 调度器 3 ( Scheduler 3 ) 的端口的总调度来保证当 端口拥塞时每个用户的带宽, 流队列 3挂接到 Scheduler 3的高优先级上; 调度器 1 ( Scheduler 1 )和调度器 2 ( Scheduler 2 )分别挂接到 Scheduler 3 的低优先级上, 流队列 1和流队列 2分别挂接到 Scheduler 1的第一队列和 第二队列上, 根据流队列 1和流队列 2的业务类型或设置分配流队列 1和 流队列 2的相关参数; 流队列 4和流队列 5分别挂接到 Scheduler 2的第一 队列和第二队列上, 根据流队列 4和流队列 5的业务类型或设置分配流队 列 4和流队列 5的相关参数。
步骤 403: 所述控制器将所述 QoS配置信息提供给有相应 QoS策略需 求的流表条目, 并将携带报文分类标识的流表条目发送给所述交换机。
这里, 所述控制器将所述 QoS配置信息提供给有相应 QoS策略需求的 流表条目, 包括:
所述控制器根据特定的业务需求生成流表条目, 将所述 QoS配置信息 提供给有相应 QoS策略需求的流表条目, 并在所述流表条目的指令动作中 设定关联的流队列标识, 所述流队列标识对应多层 QoS配置树的指定叶子 节点入口, 具体的, 在指令动作 (Action ) 中携带 Flow-Queue ID , 所述 Flow-Queue ID对应图 5本发明实施例一的 QoS配置树模型的示意图中的叶 子节点。
具体的, 表 1为本发明实施例一的 OpenFlow流表格式示意表, 如表 1 所示, 流表中可以包含多个流表条目 ( Flow Table Entry ), 包括: 匹配字段 ( Match Fields )、优先级( Priorlity )、计数器( Counters )、指令( Instructions )、 超时时限(Timeouts ), 状态信息 (Cookie )等。 其中, 所述匹配字段可以 是报文头中的任意字段, 包括: 地址字段、 用户标识字段、 服务优先级字 段、 报文分类标识字段等。
Figure imgf000014_0001
表 1
步骤 404: 所述交换机根据所述流表条目和所述多层 QoS配置树模型 执行多层 QoS流量管理策略。
这里, 所述交换机根据所述流表条目和所述多层 QoS配置树模型执行 多层 QoS流量管理策略, 包括:
所述交换机匹配输入的报文与所述流表条目的报文分类标识, 确定匹 配成功后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标 识进入所述 TM子模块按照所述多层 QoS配置树模型执行 QoS策略, 进行 流量限速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作,所述输入报 文在 TM子模块根据 QoS策略处理后输出, 或緩存, 或丟弃。
具体的, 进入所述交换机的报文, 所述报文的报文分类标识与流表条 目的报文分类标识匹配成功后, 执行后续的指令动作 (Action ), 当所述指 令动作中有 Set Flow-Queue ID时,则根据设置的 Flow-Queue ID,在 TM子 模块中执行多层 QoS流量管理策略, 按所述多层 QoS配置树模型的设置进 行报文的输出; 或是当发生拥塞时, 根据调度器和 /或流队列的丟包策略, 进行流量限速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作,按优先 级由低到高执行报文的丟弃, 或緩存, 或输出。
基于上述方法, 本发明实施例还提供了一种实现 QoS的系统, 图 6为 本发明实施例一的 OpenFlow网络中实现 QoS的系统组成结构示意图, 如 图 6所示, 所述系统包括: 交换机 10、 配置点 20和控制器 30; 其中, 所述配置点 20, 配置为根据 QoS策略生成 QoS配置信息,将所述 QoS 配置信息发送给交换机 10, 并将所述 QoS配置信息发送给控制器 30;
所述控制器 30, 配置为接收所述配置点 20发送的 QoS配置信息, 或 从所述配置点 20查询 QoS配置信息; 将所述 QoS配置信息提供给有相应 QoS 策略需求的流表条目, 并将携带报文分类标识的流表条目发送给所述 交换机 10;
所述交换机 10, 配置为根据所述配置点 20发送的 QoS配置信息生成 多层 QoS配置树模型; 根据所述控制器发送的流表条目和所述多层 QoS配 置树模型执行多层 QoS流量管理策略。 具体的, 所述配置点 20, 配置为根据控制器或应用层 APP提交的 QoS 需求形成 QOS策略,为所述 QoS策略分配流量管理资源并设置所述资源的 挂接关系, 根据所述资源挂接关系生成 QoS配置信息。
具体的, 所述配置点 20,还配置为根据 QoS策略生成 QoS配置信息之 前, 与交换机 10建立链接后, 通过 OF-Config协议的 Get-config消息获取 所述交换机 10的 TM子模块的资源信息;
其中, 所述 TM子模块的资源信息包括以下信息至少之一: 流队列数 量、 调度器数量。
具体的,所述交换机 10的 TM子模块根据所述 QoS配置信息生成多层 QoS配置树模型。
具体的, 所述交换机 10, 配置为匹配输入报文与所述流表条目的报文 分类标识, 确定匹配成功后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标识进入所述 TM子模块按照所述多层 QoS配置树模型执 行 QoS策略, 进行流量限速和 /或流量调度和 /或流量整形和 /或流量拥塞丟 包操作, 所述输入报文在 TM子模块根据 QoS策略处理后输出、 或緩存、 或丟弃。
具体的, 所述控制器 30, 配置为根据业务需求生成流表条目, 将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目, 并在所述流表条目 的指令动作中设定关联的流队列标识。
实施例二
本发明实施例还提供了一种实现 QoS的方法; 图 7为本发明实施例二 的实现 QoS的方法流程示意图; 如图 7所示, 所述方法包括:
步骤 701 : 配置点根据 QoS策略生成 QoS配置信息。
这里, 所述配置点根据 QoS策略生成 QoS配置信息, 包括:
所述配置点根据控制器或应用层( APP )提交的 QoS需求形成 QOS策 略, 为所述 QoS策略分配流量管理资源并设置所述资源的挂接关系, 根据 所述资源挂接关系生成 QoS配置信息。
这里, 所述配置点根据 QoS策略生成 QoS配置信息之前, 所述方法还 包括:
配置点与交换机建立链接后,通过 OF-Config协议的 Get-config消息获 取所述交换机的流量管理 TM子模块的资源信息;
所述 TM子模块的资源信息包括以下信息至少之一: 流队列数量、 调 度器数量。
步骤 702: 将所述 QoS配置信息发送给交换机, 并将所述 QoS配置信 息发送给控制器。
本发明实施例还提供了一种计算机存储介质, 所述计算机存储介质中 存储有计算机可执行指令, 所述计算机可执行指令配置为执行本发明实施 例所述的实现 QoS的方法。
基于上述方法, 本发明实施例还提供了一种配置点; 图 8 为本发明基 于实施二的配置点的组成结构示意图; 如图 8所示, 所述配置点 20包括: 生成模块 21和发送模块 22; 其中,
所述生成模块 21, 配置为根据 QoS策略生成 QoS配置信息;
所述发送模块 22, 配置为将所述生成模块 21生成的 QoS配置信息发 送给交换机 10; 配置为将所述生成模块 21生成的 QoS配置信息发送给控 制器 30。
优选地, 所述配置点 20还包括获取模块 23, 配置为通过 OF-Config协 议的 Get-config消息获取所述交换机 10的 TM子模块的资源信息;所述 TM 子模块 133的资源信息包括以下信息至少之一: 流队列数量、 调度器数量。
优选地, 所述生成模块 21, 配置为根据控制器 30或 APP提交的需求 形成 QoS策略, 为所述 QoS策略分配流量管理资源并设置所述资源的挂接 关系, 根据所述资源挂接关系生成 QoS配置信息。
本领域技术人员应当理解, 本发明实施例的配置点中各处理单元的功 能, 可参照前述实现 QoS的方法的相关描述而理解, 本发明实施例的配置 点中各处理单元, 可通过实现本发明实施例所述的功能的模拟电路而实现, 也可以通过执行本发明实施例所述的功能的软件在智能终端上的运行而实 现。
其中, 所述生成模块 21和所述获取模块 23在实际应用中, 均可由所 述配置点中的中央处理器(CPU, Central Processing Unit ), 数字信号处理 器 (DSP, Digital Signal Processor ) 或可编程门阵列 (FPGA, Field - Programmable Gate Array ) 实现; 所述发送模块 22在实际应用中, 可由所 述配置点中的发射机实现。
实施例三
本发明实施例还提供了一种实现 QoS的方法; 图 9为本发明实施例三 的实现 QoS的方法流程示意图; 如图 9所示, 所述方法包括:
步骤 901 : 交换机接收到配置点发送的 QoS配置信息, 根据所述 QoS 配置信息生成多层 QoS配置树模型。
这里, 所述根据所述 QoS配置信息生成多层 QoS配置树模型, 包括: 所述交换机的流量管理 TM子模块根据所述 QoS 配置信息生成多层 QoS配置树模型。
具体的, 所述交换机内部的 OpenFlow接口接收到所述 QoS配置信息 后, 将所述 QoS配置信息发送至所述交换机内部下游的 TM子模块, 所述 TM子模块根据所述 QoS配置信息生成 QoS配置树模型。
步骤 902: 根据控制器发送的流表条目和所述多层 QoS配置树模型执 行多层 QoS流量管理策略。
这里, 所述根据控制器发送流表条目和所述多层 QoS配置树模型执行 多层 QoS流量管理策略, 包括:
所述交换机匹配输入报文与所述流表条目的报文分类标识, 确定匹配 成功后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标识 进入所述 TM子模块按照所述多层 QoS配置树模型执行 QoS策略, 进行流 量限速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作,所述输入报文 在 TM子模块根据 QoS策略处理后输出、 或緩存、 或丟弃。
本发明实施例还提供了一种计算机存储介质, 所述计算机存储介质中 存储有计算机可执行指令, 所述计算机可执行指令用于执行本发明实施例 所述的实现 QoS的方法。
基于上述方法, 本发明实施例还提供了一种交换机; 图 10为本发明基 于实施三的交换机的组成结构示意图; 如图 10所示, 所述交换机 10包括: QoS配置接收模块 11、 QoS配置映射模块 12和 QoS策略执行模块 13 ; 其 中,
所述 QoS配置接收模块 11, 配置为接收配置点 20发送的 QoS配置信 息,将所述 QoS配置信息发送给所述 QoS配置映射模块 12; 还配置为接收 控制器 30发送的携带报文分类标识的流表条目; 所述 QoS配置接收模块 11一般是 OF-Config协议和 OpenFlow协议的处理模块;
所述 QoS配置映射模块 12, 配置为根据所述 QoS配置信息在对应的 QoS策略执行模块 13上对应生成多层 QoS配置树模型; 所述 QoS配置映 射模块 12充当 QoS配置接收模块 11和 QoS策略执行模块 13的适配层; 所述 QoS策略执行模块 13, 配置为根据所述 QoS配置接收模块 11接 收的携带报文分类标识的流表条目和所述 QoS配置映射模块 12生成的多层 QoS配置树模型执行多层 QoS流量管理策略。 所述 QoS策略执行模块 13 可以是专门的硬件模块和芯片。
优选地, 所述 QoS策略执行模块 13包括: 流表匹配子模块 131、 标识 获取子模块 132、 TM子模块 133; 其中,
所述流表匹配子模块 131,配置为匹配输入报文的报文分类标识与所述 接收模块 11接收的流表条目的报文分类标识;
所述标识获取子模块 132,配置为当所述流表匹配子模块 131匹配一致 后, 执行相应的流表条目的指令动作, 并从设置流队列标识的指令动作中 获取关联的流队列标识, 根据所述流队列标识将报文从相应的流队列导入 到 TM子模块 133,所述流队列标识直接对应 TM子模块 133上的预先生成 的 QoS配置树模型;
所述 TM子模块 133, 配置为根据所述 QoS配置映射模块 12预先生成 的多层 QoS配置树模型执行 QoS策略, 进行流量限速和 /或流量调度和 /或 流量整形和 /或流量拥塞丟包操作, 并决定所述输入报文从指定接口输出、 或緩存、 或丟弃。
本领域技术人员应当理解, 本发明实施例的交换机中各处理单元的功 能, 可参照前述实现 QoS的方法的相关描述而理解, 本发明实施例的交换 机中各处理单元, 可通过实现本发明实施例所述的功能的模拟电路而实现, 也可以通过执行本发明实施例所述的功能的软件在智能终端上的运行而实 现。
其中, 所述 QoS配置映射模块 12和 QoS策略执行模块 13在实际应用 中, 均可由所述交换机中的 CPU、 DSP或 FPGA实现; 所述 QoS配置接收 模块 11在实际应用中, 可由所述交换机中的接收机实现。
实施例四
本发明实施例还提供了一种实现 QoS的方法;图 11为本发明实施例四 的实现 QoS的方法流程示意图; 如图 11所示, 所述方法包括:
步骤 1101 : 控制器接收配置点发送的 QoS配置信息, 或从所述配置点 查询 QoS配置信息。 步骤 1102:将所述 QoS配置信息提供给有相应 QoS策略需求的流表条 目, 并将携带报文分类标识的流表条目发送给所述交换机。
这里, 所述将所述 QoS配置信息提供给有相应 QoS策略需求的流表条 目, 包括:
所述控制器根据业务需求生成流表条目, 将所述 QoS配置信息提供给 有相应 QoS策略需求的流表条目, 并在所述流表条目的指令动作中设定关 联的流队列标识。
本发明实施例还提供了一种计算机存储介质, 所述计算机存储介质中 存储有计算机可执行指令, 所述计算机可执行指令用于执行本发明实施例 所述的实现 QoS的方法。
基于上述方法, 本发明实施例还提供了一种控制器; 图 12为本发明基 于实施例四的控制器的组成结构示意图; 如图 12所示, 所述控制器 30包 括: 获取模块 31、 流表条目生成模块 32和发送模块 33 ; 其中,
所述获取模块 31, 配置为接收配置点 20发送的 QoS配置信息, 或向 配置点 20查询相关的 QoS配置信息, 将所述 QoS配置信息发送给流表条 目生成模块 32;
所述流表条目生成模块 32, 配置为将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目并生成流表条目;
所述发送模块 33,配置为将所述流表条目生成模块 32生成的携带报文 分类标识的流表条目发送给交换机 10。
优选地, 所述流表条目生成模块 32, 配置为根据业务需求生成流表条 目, 将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目, 并在所 述流表条目的指令动作中设定关联的流队列标识。
本领域技术人员应当理解, 本发明实施例的控制器中各处理单元的功 能, 可参照前述实现 QoS的方法的相关描述而理解, 本发明实施例的控制 器中各处理单元, 可通过实现本发明实施例所述的功能的模拟电路而实现, 也可以通过执行本发明实施例所述的功能的软件在智能终端上的运行而实 现。
其中, 所述流表条目生成模块 32在实际应用中, 可由所述控制器中的 CPU, DSP或 FPGA实现; 所述发送模块 33在实际应用中, 可由所述控制 器中的发射机实现; 所述获取模块 31在实际应用中, 可由所述控制器中的 接收机实现, 或者可由所述控制器中的收发机结合 CPU、 DSP或 FPGA实 现。
在本发明上述实施例中, 所述控制器、 交换机和配置点, 在实际应用 中, 可由实现 QoS业务的网络设备实现。
下面以具体的实施例对本发明再作进一步详细的说明
实施例五
图 13 为本发明实施例五的实现多层 QoS的应用场景的网络拓朴示意 图, 如图 13所示, 接口 1和接口 2为输入端, 用于与用户终端接入; 接口 3为输出端, 接口带宽为 1G。
图 14为本发明实施例五的多层 QoS应用的配置树模型的示意图,结合 图 13和图 14,接口 1和接口 2总共接入四个用户,分别为:用户 l ( userl )、 用户 2 ( user2 )、 用户 3 ( user3 )和用户 4 ( user4 )。 所述四个用户的 QoS 要求如下:
userl优先通过, 最大带宽 500M;
user2保证速率 300M, 可以大于 300M;
user3保证速率 100M, 突发速率 150M;
user4保证速率 100M, 突发速率 150M。
本实施例中, 实现 QoS的方法包括如下步骤:
步骤一:配置点和交换机建立链接后,通过 OF-Config协议的 Get-config 消息获取所述交换机的 TM子模块的资源信息, 所述 TM子模块的资源信 息包括流队列数量以及各种类型调度器数量等等。所述配置点获取所述 TM 子模块的资源信息以便于后续生成 QoS配置信息。
步骤二: 配置点根据控制器或 APP提交的上述四个用户的 QoS要求分 配流量管理资源并设置所述资源的挂接关系, 根据所述资源挂接关系生成
QoS配置信息, 将所述配置信息分别发送给交换机和控制器。
具体的, 通过多层 QoS实现上述要求, 实现方式如下:
1、 在接口 3 下行配置多层 QoS, 以调度器 1 ( Schedulerl ) 的接口 3 端口的总调度来保证当端口拥塞时可以保证每个用户的带宽,所述调度器 1 为 SPQ类型。
2、 Userl的流队列 ( Flow-Queue )限速 500M, 挂接到 Schedulerl高优 先级上。
3、 分配一个调度器 2 ( Schedule )挂接到 Schedulerl低优先级上, 所 述 Schedule 为 WFQ类型。
4、 User2的 Flow-Queue 的 CIR参数配置为 300M, 挂接到 Schedule 第一个队列上, 权重参数设置为 60%。
5、 分配一个调度器 3 ( Scheduled )挂接到 Schedule 第二个队列上, 权重设置为 40%; 所述 Scheduled为公平队列 (FQ, Fair Queuing )类型。
6、 User3 的 Flow-Queue 的 CIR参数配置为 100M, PI 参数配置为 150M, 挂接到 Scheduled的第一个队列上。
7、 User4的 Flow-Queue 的 CIR参数配置为 100M, PI 参数配置为 150M, 挂接到 Scheduled的第二个队列上。
步骤三, 交换机接收到上述四个用户的 QoS配置信息, 在 TM子模块 上形成 QoS配置 4对模型。
具体的, 所述 QoS 配置树模型的可扩展标记语言 (XML, Extensible
Figure imgf000024_0001
。z
U <V33bOnn.-
Π
Figure imgf000025_0001
/_7 ΛνΛνιπι!ΕιΙΙιΙΙιΙΕιΙΙηΧηΧ--.
¾ν¾09ΑΑν¾¾ ΛνΛν <Weitht>40</Weitht>
</Weithts>
</properties>
</scheduler>
<scheduler>
<resource-id>scheduler3</resource-id>
<id>3</id>
<type>FQ</type>
<scheduler>2</ scheduler
<attach>2</attach>
<properties>
<max-num>4</ max-num>
</properties>
</scheduler>
步骤四: 所述控制器根据上述四个用户的特征生成流表条目, 并向交 换机下发携带有报文分类标识的流表条目, 下发时, 在指令动作 (Action ) 中携带 Flow-Queue ID。
步骤五: 进入交换机的报文, 首先匹配所述报文与流表条目的报文分 类标识, 匹配成功后, 执行后续的指令动作 (Action ), 当所述指令动作中 有 "Set Flow-Queue ID" 字段时, 则根据设置的 Flow-Queue ID, 在 TM子 模块中执行多层 QoS流量管理策略, 按所述多层 QoS配置树模型的设置进 行报文的输出; 当有拥塞发生时, 根据调度器和 /或流队列的丟包策略进行 处理, 在接口 Int3上进行优先级由低到高的用户报文的丟弃、 緩存、 输出。
实施例六
本实施例中, 分配到虚拟局域网 10的用户 A ( VLAN— ID = 10 )有两个 服务等级的业务流量经过 OpenFlow网络, 其 VLAN字段中 CoS字段分别 为 CoS = 1和 CoS = 2; 分配到虚拟局域网 20的用户 B ( VLAN— ID = 20 ) 也有两个服务等级的业务流量经过 OpenFlow网络, 其 VLAN字段中 CoS 字段分别为 CoS = 1和 CoS = 2, 并从同一个交换机出接口转发, 需要在交 换机上完成多层 QoS, 实现各用户中不同服务等级的业务流量之间和不同 用户的业务流量之间的两层调度, 不同用户之间流量公平调度。 图 15为本 发明实施例三的多层 QoS应用的配置树模型的示意图。
本实施例中, OpenFlow网络实现 QoS的方法包括如下步骤: 步骤一: 配置点和交换机建链后, 通过 OF-Config协议的 Get-config消 息获取所述交换机的 TM子模块的资源信息, 所述 TM子模块的资源信息 包括流队列数量、 各种类型调度器数量等等。 所述配置点获取所述 TM子 模块的资源信息以便于后续生成 QoS配置信息。
步骤二: 配置点根据控制器或 APP提交的上述用户的 QoS要求分配流 量管理资源并设置所述资源的挂接关系, 根据所述资源挂接关系生成 QoS 配置信息, 将所述配置信息分别发送给交换机和控制器。
具体的, 通过多层 QoS实现上述要求, 实现方式如下:
1、 在接口 1下行配置多层 QoS, 以调度器 1 ( Scheduler 1 )的 Int3端口 的总调度来保证当端口拥塞时可以保证每个用户的带宽。
2、 分配一个调度器 2 ( Schedule )挂接到 Schedulerl上。
3、 分配一个调度器 3 ( Scheduled )挂接到 Schedule 第一个队列上。
4、 用户 A CoS = 1的 Flow-Queue挂接到 Scheduled的第一个队列上。 5、 用户 A CoS = 2的 Flow-Queue挂接到 Scheduled的第二个队列上。
6、 分配一个调度器 4 ( Scheduler4 )挂接到 Schedule 第二个队列上。
7、 用户 B CoS = 1的 Flow-Queue挂接到 Scheduler4的第一个队列上。
8、 用户 B CoS = 2的 Flow-Queue挂接到 Scheduler4的第二个队列上。 步骤三, 交换机接收到上述两个用户四种流的 QoS配置信息, 在 TM 子模块上形成 QoS配置树模型。
U <V33bOnn.-
Figure imgf000028_0001
ρνρΐ Λν Λν ξτ
Figure imgf000029_0001
/ζΛν ΛνιαιΙΙιΙΕιΙΙιΙΙδβηΧΓχ-- J/l <〉s3J3doJd <scheduler>
<resource-id>scheduler4</resource-id>
<id>4</id>
<type>SPQ</type>
<scheduler>2</ scheduler>
<attach>2</attach>
<properties>
<max-num>2</ max-num>
</properties>
</scheduler>
步骤四: 所述控制器根据上述两个用户四种流的特征生成流表条目, 并向交换机下发携带有报文分类标识的流表条目; 下发流表条目时, 在指 令动作 ( Action ) 中携带 Flow-Queue ID字段。
步骤五: 进入交换机的报文, 首先匹配所述报文与流表条目的报文分 类标识, 匹配成功后, 执行后续的指令动作 (Action ), 当所述指令动作中 有 "Set Flow-Queue ID" 时, 则根据设置的 Flow-Queue ID, 在 TM子模块 中执行多层 QoS流量管理策略, 按所述多层 QoS配置树模型的设置进行报 文的输出; 当有拥塞发生时,根据调度器和 /或流队列的丟包策略进行处理, 在接口 Int3上进行优先级由低到高的用户报文的丟弃、 或緩存、 或输出。
本发明是参照根据本发明实施例的方法、 装置、 和计算机程序产品的 流程图和 /或方框图来描述的。 应理解可由计算机程序指令实现流程图和 / 或方框图中的每一流程和 /或方框、以及流程图和 /或方框图中的流程和 /或方 框的结合。 可提供这些计算机程序指令到通用计算机、 专用计算机、 嵌入 式处理机或其他可编程数据处理设备的处理器以产生一个机器, 使得通过 程图一个流程或多个流程和 /或方框图一个方框或多个方框中指定的功能的 装置。 这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理 设备以特定方式工作的计算机可读存储器中, 使得存储在该计算机可读存 储器中的指令产生包括指令装置的制造品, 该指令装置实现在流程图一个 流程或多个流程和 /或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备 上, 使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机 实现的处理, 从而在计算机或其他可编程设备上执行的指令提供用于实现 在流程图一个流程或多个流程和 /或方框图一个方框或多个方框中指定的功 能的步骤。
以上所述仅是本发明实施例的实施方式, 应当指出, 对于本技术领域 的普通技术人员来说, 在不脱离本发明实施例原理的前提下, 还可以作出 若干改进和润饰, 这些改进和润饰也应视为本发明实施例的保护范围。 工业实用性
本发明实施例通过扩展现有的 OF-Config协议, 将单层或多层 QoS的 配置信息写入 TM子模块, 生成相应的单层或多层 QoS配置树模型, 实现 了多层 QoS所要求的多层队列调度; 并且, 将流表管道和 TM子模块分开 实现, 串联处理, 避免了流表管道和 TM子模块交织导致的高速总线带宽 的消耗。

Claims

权利要求书
1、 一种实现服务质量 QoS的方法, 所述方法包括:
配置点根据 QoS策略生成 QoS配置信息,将所述 QoS配置信息发送给 交换机, 并将所述 QoS配置信息发送给控制器或供控制器查询 QoS配置信 息;
所述交换机根据所述 QoS配置信息生成多层 QoS配置树模型; 所述控制器将所述 QoS配置信息提供给有相应 QoS策略需求的流表条 目, 并将携带报文分类标识的流表条目发送给所述交换机;
所述交换机根据所述流表条目和所述多层 QoS 配置树模型执行多层
QoS流量管理策略。
2、 根据权利要求 1所述的方法, 其中, 所述配置点根据 QoS策略生成 QoS配置信息, 包括:
配置点根据控制器或应用层 APP提交的 QoS需求形成 QOS策略, 为 所述 QoS策略分配流量管理资源并设置所述资源的挂接关系, 根据所述资 源挂接关系生成 QoS配置信息。
3、 根据权利要求 1所述的方法, 其中, 所述控制器将所述 QoS配置信 息提供给有相应 QoS策略需求的流表条目, 包括:
所述控制器根据业务需求生成流表条目, 将所述 QoS配置信息提供给 有相应 QoS策略需求的流表条目, 并在所述流表条目的指令动作中设定关 联的流队列标识。
4、 根据权利要求 1所述的方法, 其中, 所述配置点根据 QoS策略生成 QoS配置信息之前, 所述方法还包括:
配置点与交换机建立链接后,通过 OF-Config协议的 Get-config消息获 取所述交换机的流量管理 TM子模块的资源信息;
所述 TM子模块的资源信息包括以下信息至少之一: 流队列数量、 调 度器数量。
5、 根据权利要求 4所述的方法, 其中, 所述交换机根据所述 QoS配置 信息生成多层 QoS配置树模型, 包括:
所述交换机的 TM子模块根据所述 QoS配置信息生成多层 QoS配置树 模型。
6、 根据权利要求 1至 5任一项所述的方法, 其中, 所述交换机根据所 述流表条目和所述多层 QoS配置树模型执行多层 QoS流量管理策略,包括: 交换机匹配输入报文与所述流表条目的报文分类标识, 确定匹配成功 后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标识进入 所述 TM子模块按照所述多层 QoS配置树模型执行 QoS策略, 进行流量限 速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作, 所述输入报文在 TM子模块根据 QoS策略处理后输出、 或緩存、 或丟弃。
7、 一种实现 QoS的方法, 所述方法包括:
配置点根据 QoS策略生成 QoS配置信息,将所述 QoS配置信息发送给 交换机, 并将所述 QoS配置信息发送给控制器。
8、 根据权利要求 7所述的方法, 其中, 所述配置点根据 QoS策略生成 QoS配置信息, 包括:
所述配置点根据控制器或应用层 APP提交的 QoS需求形成 QOS策略, 为所述 QoS策略分配流量管理资源并设置所述资源的挂接关系, 根据所述 资源挂接关系生成 QoS配置信息。
9、 根据权利要求 7所述的方法, 其中, 所述配置点根据 QoS策略生成 QoS配置信息之前, 所述方法还包括:
配置点与交换机建立链接后,通过 OF-Config协议的 Get-config消息获 取所述交换机的流量管理 TM子模块的资源信息;
所述 TM子模块的资源信息包括以下信息至少之一: 流队列数量、 调 度器数量。
10、 一种实现 QoS的方法, 所述方法包括:
交换机接收到配置点发送的 QoS配置信息, 根据所述 QoS配置信息生 成多层 QoS配置 4对模型;
根据控制器发送的流表条目和所述多层 QoS配置树模型执行多层 QoS 流量管理策略。
11、 根据权利要求 10所述的方法, 其中, 所述根据所述 QoS配置信息 生成多层 QoS配置树模型, 包括:
所述交换机的流量管理 TM子模块根据所述 QoS 配置信息生成多层 QoS配置树模型。
12、 根据权利要求 10所述的方法, 其中, 所述根据控制器发送流表条 目和所述多层 QoS配置树模型执行多层 QoS流量管理策略, 包括:
所述交换机匹配输入报文与所述流表条目的报文分类标识, 确定匹配 成功后, 获取相应的流表条目中关联的流队列标识, 根据所述流队列标识 进入所述 TM子模块按照所述多层 QoS配置树模型执行 QoS策略, 进行流 量限速和 /或流量调度和 /或流量整形和 /或流量拥塞丟包操作,所述输入报文 在 TM子模块根据 QoS策略处理后输出、 或緩存、 或丟弃。
13、 一种实现 QoS的方法, 所述方法包括:
控制器接收配置点发送的 QoS配置信息, 或从所述配置点查询 QoS配 置信息;
将所述 QoS配置信息提供给有相应 QoS策略需求的流表条目, 并将携 带报文分类标识的流表条目发送给所述交换机。
14、 根据权利要求 13所述的方法, 其中, 所述将所述 QoS配置信息提 供给有相应 QoS策略需求的流表条目, 包括:
所述控制器根据业务需求生成流表条目, 将所述 QoS配置信息提供给 有相应 QoS策略需求的流表条目, 并在所述流表条目的指令动作中设定关 联的流队列标识。
15、 一种交换机, 所述交换机包括: QoS配置接收模块、 QoS 配置映 射模块和 QoS策略执行模块; 其中,
所述 QoS配置接收模块, 配置为接收配置点发送的 QoS配置信息, 将 所述 QoS配置信息发送给所述 QoS配置映射模块; 还配置为接收控制器发 送的携带报文分类标识的流表条目;
所述 QoS配置映射模块, 配置为才艮据所述 QoS配置信息在对应的 QoS 策略执行模块上对应生成多层 QoS配置树模型;
所述 QoS策略执行模块, 配置为根据所述 QoS配置接收模块接收的携 带报文分类标识的流表条目和所述 QoS配置映射模块生成的多层 QoS配置 树模型执行多层 QoS流量管理策略。
16、根据权利要求 15所述交换机,其中,所述 QoS策略执行模块包括: 流表匹配子模块、 标识获取子模块和流量管理 TM子模块; 其中,
所述流表匹配子模块, 配置为匹配输入报文的报文分类标识与所述接 收模块接收的流表条目的报文分类标识;
所述标识获取子模块, 配置为当所述流表匹配子模块匹配一致后, 执 行相应的流表条目的指令动作, 并从设置流队列标识的指令动作中获取关 联的流队列标识, 根据所述流队列标识将报文从相应的流队列导入到流量 管理子模块, 所述流队列标识直接对应流量管理子模块上的相应的预先生 成的 QoS配置树模型;
所述流量管理子模块, 配置为 4艮据所述 QoS配置映射模块预先生成的 多层 QoS配置树模型执行 QoS策略, 进行流量限速和 /或流量调度和 /或流 量整形和 /或流量拥塞丟包操作, 并决定所述输入报文从指定接口输出、 或 緩存、 或丟弃。
17、 一种配置点, 所述配置点包括: 生成模块和发送模块; 其中, 所述生成模块, 配置为根据 QoS策略生成 QoS配置信息;
所述发送模块, 配置为将所述生成模块生成的 QoS配置信息发送给交 换机; 配置为将所述生成模块生成的 QoS配置信息发送给控制器。
18、 根据权利要求 17所述的配置点, 其中, 所述配置点还包括获取模 块,配置为通过 OF-Config协议的 Get-config消息获取所述交换机的流量管 理 TM子模块的资源信息; 所述 TM子模块的资源信息包括以下信息至少 之一: 流队列数量、 调度器数量。
19、 根据权利要求 17所述的配置点, 其中, 所述生成模块, 配置为根 据控制器或应用层 APP提交的 QoS需求形成 QoS策略, 为所述 QoS策略 分配流量管理资源并设置所述资源的挂接关系, 根据所述资源挂接关系生 成 QoS配置信息。
20、 一种控制器, 所述控制器包括: 获取模块、 流表条目生成模块和 发送模块; 其中,
所述获取模块, 配置为接收配置点发送的 QoS配置信息, 或向配置点 查询相关的 QoS配置信息,将所述 QoS配置信息发送给流表条目生成模块; 所述流表条目生成模块, 配置为将所述 QoS 配置信息提供给有相应 QoS策略需求的流表条目并生成流表条目;
所述发送模块, 配置为将所述流表条目生成模块生成的携带报文分类 标识的流表条目发送给交换机。
21、 根据权利要求 20所述的控制器, 其中, 所述流表条目生成模块, 配置为根据业务需求生成流表条目, 将所述 QoS 配置信息提供给有相应 QoS 策略需求的流表条目, 并在所述流表条目的指令动作中设定关联的流 队列标识。
22、 一种实现服务质量 QoS的系统, 所述系统包括: 交换机、 配置点 和控制器; 其中,
所述配置点, 配置为根据 QoS策略生成 QoS配置信息, 将所述 QoS 配置信息发送给交换机, 并将所述 QoS配置信息发送给控制器;
所述控制器, 配置为接收所述配置点发送的 QoS配置信息, 或从所述 配置点查询 QoS配置信息;将所述 QoS配置信息提供给有相应 QoS策略需 求的流表条目, 并将携带报文分类标识的流表条目发送给所述交换机; 所述交换机, 配置为根据所述配置点发送的 QoS 配置信息生成多层 QoS配置树模型; 根据所述控制器发送的流表条目和所述多层 QoS配置树 模型执行多层 QoS流量管理策略。
23、 一种计算机存储介质, 所述计算机存储介质中存储有计算机可执 行指令, 所述计算机可执行指令用于执行权利要求 7至 9任一项所述的实 现 QoS的方法。
24、 一种计算机存储介质, 所述计算机存储介质中存储有计算机可执 行指令, 所述计算机可执行指令用于执行权利要求 10至 12任一项所述的 实现 QoS的方法。
25、 一种计算机存储介质, 所述计算机存储介质中存储有计算机可执 行指令,所述计算机可执行指令用于执行权利要求 13或 14所述的实现 QoS 的方法。
PCT/CN2014/079723 2013-08-16 2014-06-12 实现QoS的方法、系统、设备及计算机存储介质 WO2014173367A2 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310359387.5 2013-08-16
CN201310359387.5A CN104378309B (zh) 2013-08-16 2013-08-16 OpenFlow网络中实现QoS的方法、系统和相关设备

Publications (2)

Publication Number Publication Date
WO2014173367A2 true WO2014173367A2 (zh) 2014-10-30
WO2014173367A3 WO2014173367A3 (zh) 2014-11-27

Family

ID=51792442

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/079723 WO2014173367A2 (zh) 2013-08-16 2014-06-12 实现QoS的方法、系统、设备及计算机存储介质

Country Status (2)

Country Link
CN (1) CN104378309B (zh)
WO (1) WO2014173367A2 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106453138A (zh) * 2016-11-25 2017-02-22 杭州华三通信技术有限公司 一种报文处理方法和装置
CN111294291A (zh) * 2020-01-16 2020-06-16 新华三信息安全技术有限公司 一种协议报文的处理方法和装置
EP4175253A4 (en) * 2020-07-16 2023-11-29 Huawei Technologies Co., Ltd. QUEUE SCHEDULING METHOD, DEVICE AND SYSTEM

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106487713A (zh) * 2015-08-26 2017-03-08 深圳市中兴微电子技术有限公司 一种服务质量复用方法及装置
CN105553710A (zh) * 2015-12-14 2016-05-04 北京信息科技大学 在软件定义网络中的交换机中控制QoS的设备和方法
CN107483361B (zh) * 2016-06-08 2022-04-15 中兴通讯股份有限公司 一种调度模型构建方法和装置
CN106254271B (zh) * 2016-08-08 2019-07-19 北京邮电大学 一种用于软件定义网络的可编程队列配置方法及装置
CN106789640A (zh) * 2016-11-10 2017-05-31 上海斐讯数据通信技术有限公司 一种基于sdn的优先级分类方法、装置和系统
US10331485B2 (en) 2016-11-18 2019-06-25 Huawei Technologies Co., Ltd. Method and system for meeting multiple SLAS with partial QoS control
CN108449279B (zh) * 2016-12-13 2021-10-26 中国移动通信有限公司研究院 拥塞控制方法及装置
CN106685849B (zh) * 2016-12-22 2020-11-27 台州市吉吉知识产权运营有限公司 基于sdn的wred报文丢弃方法及系统
CN109644078B (zh) * 2017-03-23 2021-05-25 Oppo广东移动通信有限公司 一种上行数据传输方法、终端、网络侧设备及系统
CN109644377B (zh) * 2017-03-23 2021-01-12 Oppo广东移动通信有限公司 一种上行数据传输方法、终端、网络侧设备及系统
CN107332686B (zh) * 2017-05-23 2019-08-06 烽火通信科技股份有限公司 一种在SPTN设备上实现QoS功能的方法及系统
CN107171852B (zh) * 2017-06-15 2019-11-08 烽火通信科技股份有限公司 基于OpenFlow的二层业务配置系统及其方法
CN109218060B (zh) * 2017-07-07 2023-04-07 中兴通讯股份有限公司 一种业务配置驱动流表的方法及装置
CN109768931B (zh) * 2017-11-09 2020-10-13 中国移动通信集团公司 处理数据包的方法、交换机、装置及计算机可读存储介质
CN107896192B (zh) * 2017-11-20 2020-09-25 电子科技大学 一种SDN网络中区分业务优先级的QoS控制方法
US10686714B2 (en) * 2018-04-27 2020-06-16 Avago Technologies International Sales Pte. Limited Traffic management for high-bandwidth switching
CN109450798B (zh) * 2018-12-13 2022-07-12 郑州云海信息技术有限公司 路由表信息的管理方法和计算机可读存储介质
CN110808916B (zh) * 2019-10-31 2022-12-09 武汉二进制半导体有限公司 一种基于分簇设计的Qos实现方法及系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101123534A (zh) * 2007-09-29 2008-02-13 华中科技大学 应用于合法监听系统的网络策略架构及其策略处理方法
CN101583160A (zh) * 2009-06-19 2009-11-18 中兴通讯股份有限公司 一种实现分层服务质量业务的装置及方法
CN101667974A (zh) * 2009-10-12 2010-03-10 中兴通讯股份有限公司 一种实现分层服务质量的方法及网络处理器

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7974241B2 (en) * 2007-09-28 2011-07-05 At&T Intellectual Property I, L.P. Methods, systems, and computer-readable media for providing cellular handoff
CN102148764B (zh) * 2011-05-09 2013-12-11 杭州华三通信技术有限公司 一种基于QoS业务的数据处理方法和设备
CN102868645A (zh) * 2012-09-26 2013-01-09 盛科网络(苏州)有限公司 Openflow交换机系统及其报文处理方法
CN103236944A (zh) * 2013-04-08 2013-08-07 北京天地互连信息技术有限公司 基于OpenFlow网络的QoS集中管理系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101123534A (zh) * 2007-09-29 2008-02-13 华中科技大学 应用于合法监听系统的网络策略架构及其策略处理方法
CN101583160A (zh) * 2009-06-19 2009-11-18 中兴通讯股份有限公司 一种实现分层服务质量业务的装置及方法
CN101667974A (zh) * 2009-10-12 2010-03-10 中兴通讯股份有限公司 一种实现分层服务质量的方法及网络处理器

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106453138A (zh) * 2016-11-25 2017-02-22 杭州华三通信技术有限公司 一种报文处理方法和装置
CN111294291A (zh) * 2020-01-16 2020-06-16 新华三信息安全技术有限公司 一种协议报文的处理方法和装置
EP4175253A4 (en) * 2020-07-16 2023-11-29 Huawei Technologies Co., Ltd. QUEUE SCHEDULING METHOD, DEVICE AND SYSTEM

Also Published As

Publication number Publication date
WO2014173367A3 (zh) 2014-11-27
CN104378309B (zh) 2019-05-21
CN104378309A (zh) 2015-02-25

Similar Documents

Publication Publication Date Title
WO2014173367A2 (zh) 实现QoS的方法、系统、设备及计算机存储介质
US10554570B2 (en) Method and device for implementing QoS in OpenFlow network
WO2021227245A1 (zh) 一种面向tsn和非tsn互联的工业异构网络调度方法
US11316795B2 (en) Network flow control method and network device
US8797867B1 (en) Generating and enforcing a holistic quality of service policy in a network
US8787375B2 (en) Multicast to unicast conversion technique
US20220217051A1 (en) Method, Device, and System for Determining Required Bandwidth for Data Stream Transmission
US10432554B2 (en) Bandwidth providing method based on multi-flow grouping
WO2008095397A1 (fr) Procédé de planification de trafic et appareil apparenté
CN108471629A (zh) 传输网络中业务服务质量的控制方法、设备及系统
WO2006069527A1 (fr) Procede, dispositif et reseau permettant de garantir la qualite de service d&#39;un service pour un acces a large bande
WO2016082603A1 (zh) 一种调度器及调度器的动态复用方法
CN112600684B (zh) 云业务的带宽管理及配置方法及相关装置
CN105915468B (zh) 一种业务的调度方法及装置
KR20170033179A (ko) 소프트웨어 정의 네트워크 기반 가상 네트워크 사용 대역폭 관리 방법 및 가상 네트워크 관리장치
JP2012182605A (ja) ネットワーク制御システム及び管理サーバ
WO2016177058A1 (zh) 数据流调度方法及装置
CN110417656A (zh) 一种基于sdn网络的服务质量保障系统以及方法
KR101587379B1 (ko) 큐 사이즈의 동적 제어 방법 및 이를 수행하는 장치
Tokmakov et al. A case for data centre traffic management on software programmable ethernet switches
Li et al. Design and implementation of a new queue scheduling scheme in DiffServ networks
Lan et al. A real-time network traffic analysis and QoS management platform
Kaur et al. Implementation of Differential Services Based on Priority, Token Bucket, and Round Robin Algorithms
Addeo et al. On QoS mechanism profiling in MPLS-TP transport networks
Fidler On the impacts of traffic shaping on end-to-end delay bounds in aggregate scheduling networks

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14788637

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14788637

Country of ref document: EP

Kind code of ref document: A2