WO2017211287A1 - Method and device for constructing scheduling model - Google Patents

Method and device for constructing scheduling model Download PDF

Info

Publication number
WO2017211287A1
WO2017211287A1 PCT/CN2017/087406 CN2017087406W WO2017211287A1 WO 2017211287 A1 WO2017211287 A1 WO 2017211287A1 CN 2017087406 W CN2017087406 W CN 2017087406W WO 2017211287 A1 WO2017211287 A1 WO 2017211287A1
Authority
WO
WIPO (PCT)
Prior art keywords
scheduling
level
unit
units
scheduling unit
Prior art date
Application number
PCT/CN2017/087406
Other languages
French (fr)
Chinese (zh)
Inventor
傅正龙
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 WO2017211287A1 publication Critical patent/WO2017211287A1/en

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
    • H04L47/62—Queue scheduling characterised by scheduling criteria
    • H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
    • H04L47/6275—Queue scheduling characterised by scheduling criteria for service slots or service orders based on priority
    • 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
    • H04L47/60—Queue scheduling implementing hierarchical scheduling
    • 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
    • H04L47/62—Queue scheduling characterised by scheduling criteria
    • H04L47/6295—Queue scheduling characterised by scheduling criteria using multiple queues, one for each individual QoS, connection, flow or priority

Definitions

  • the present application relates to the field of communication technologies, for example, to a scheduling model construction method and apparatus.
  • HQOS Hierarchical Quality Of Service
  • the current HQOS scheduling model for network devices is mainly the user model and the business model.
  • both the user model and the business model are fixed, that is, the link relationships between the scheduling levels and levels in the two models are fixed.
  • different service scenarios generally have different service requirements.
  • SP Strict Priority
  • the scheduling model supports three types of SP-scheduled network devices.
  • each user has four types of SP-scheduled services and four types of service-weighted round-robin (WRR) scheduled services.
  • WRR service-weighted round-robin
  • the embodiments of the present disclosure provide a scheduling model construction method and apparatus, which solves the problem that the service requirements of different service scenarios cannot be met.
  • the embodiment of the present disclosure provides a scheduling model construction method, including:
  • the scheduling level of the current HQOS scheduling model to be built is an N level, where the N is a positive integer greater than or equal to 2;
  • the embodiment of the present disclosure further provides a scheduling model construction apparatus, which includes:
  • the determining module is configured to determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
  • a selection module configured to select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource;
  • the first configuration module is configured to configure a link relationship of the scheduling unit of each scheduling level.
  • Embodiments of the present disclosure also provide a computer storage medium having stored therein one or more programs executable by a computer, the one or more programs being executed by the computer to cause the computer to perform as described above
  • a scheduling model construction method is provided.
  • An embodiment of the present disclosure further provides an electronic device, including:
  • At least one processor At least one processor
  • the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the method described above.
  • the scheduling level of the current HQOS scheduling model to be built is determined to be N level according to the service requirement of the current service scenario; the scheduling unit of each of the N scheduling levels is selected from the storage resources; and the scheduling unit of each scheduling level is configured. Link relationship.
  • the HQOS scheduling model can be configured according to the service requirements of different service scenarios, so as to meet the service requirements of different service scenarios.
  • FIG. 1 is a schematic flowchart diagram of a method for constructing a scheduling model according to an embodiment of the present disclosure
  • FIG. 2 is a schematic flowchart diagram of another scheduling model construction method according to an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram of a scheduling model according to an embodiment of the present disclosure.
  • FIG. 4 is a schematic structural diagram of a scheduling model construction apparatus according to an embodiment of the present disclosure.
  • FIG. 5 is a schematic structural diagram of another scheduling model construction apparatus according to an embodiment of the present disclosure.
  • FIG. 6 is a schematic structural diagram of another scheduling model construction apparatus according to an embodiment of the present disclosure.
  • FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
  • an embodiment of the present disclosure provides a scheduling model construction method, including the following steps:
  • Step S101 Determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, where the N is a positive integer greater than or equal to 2.
  • the scheduling level of the current HQOS scheduling model to be constructed may be determined according to the service requirement of the current service scenario, where the N may correspond to the service requirement, that is, the service requirement and the level of the scheduling level of the HQOS scheduling model are Correspondingly, the constructed HQOS scheduling model can meet the business requirements of the current business scenario.
  • the correspondence here may be preset, that is, the step S101 may determine the number of levels of the scheduling level corresponding to the service requirement of the current service scenario in the preset relationship, or may also receive the user input in step S101. Determined by operation.
  • the scheduling level of the HQOS scheduling model can be understood as the level of the HQOS scheduling model, and one level corresponds to one level.
  • the HQOS scheduling model includes a user level and a port level, that is, the HQOS scheduling model includes the user. Two levels of level and port level.
  • Step S102 Select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource.
  • multiple types of scheduling units may be set in the storage resource before the step S102 is performed, and one or more scheduling units may be set for each type of scheduling unit, which is not limited in this embodiment.
  • the scheduling unit may be understood as a register in a storage resource, for example, a scheduling unit may include one or more registers.
  • the foregoing storage resource may be a local storage resource of the network device, and the network device may be a device that implements the method.
  • the scheduling unit belongs to the storage resource that is, multiple types of scheduling units are preset in the foregoing storage resource, it can be understood that multiple types of scheduling units are divided in the storage resource.
  • the type of the scheduling unit is not limited.
  • the SP may be configured with an SP scheduling unit, a Weighted Fair Queuing (WFQ) scheduling unit, a WFQ scheduling unit, and a WFQ2 scheduling unit.
  • each type of scheduling unit can be set to one or more.
  • the SP scheduling unit can implement strict priority scheduling between multiple streams
  • the WFQ scheduling unit can implement weight scheduling between multiple streams
  • the WFQ2 scheduling unit includes a combination of two WFQ scheduling units and one SP scheduling unit, and implements two priority scheduling and weight scheduling between multiple flows under each priority;
  • the WFQ4 scheduling unit includes four WFQ scheduling units and one SP scheduling unit combination to implement four priority scheduling and weight scheduling between multiple streams under each priority level;
  • the WFQ8 scheduling unit includes eight WFQ scheduling units and one SP scheduling unit combination, and implements eight priority scheduling and weight scheduling between multiple flows under each priority;
  • the FQ scheduling unit can implement the same weight among multiple streams, that is, a 1:1 scheduling
  • the FQ2 scheduling unit includes two FQ scheduling units and one SP scheduling unit combination to implement two priority scheduling, and the weights between the multiple streams under each priority are the same, that is, the 1:1 scheduling;
  • the FQ4 scheduling unit includes four FQ scheduling units and one SP scheduling unit combination to implement four priority scheduling, and the weights of the multiple flows under each priority are the same, that is, the 1:1 scheduling;
  • the FQ8 scheduling unit includes 8 FQ scheduling units and 1 SP scheduling unit combination to implement 8 priority scheduling, and the weights of multiple streams under each priority are the same, that is, 1:1 scheduling.
  • the scheduling unit of each level may be selected according to the service requirement of the current service scenario, for example, selecting the scheduling unit of each level according to the correspondence between the preset service requirement and the scheduling unit, or may be receiving the user input.
  • the operation selects the scheduling unit of each level and the like, which is not limited thereto.
  • Step S103 Configure a link relationship of the scheduling unit of each scheduling level.
  • the link relationship of the scheduling unit of each scheduling level in this step may be understood as setting a link relationship of each scheduling unit in each scheduling level, where the link relationship may include a link relationship of the level internal scheduling unit, or may include a level between The connection relationship of the scheduling unit.
  • the link relationship of the scheduling unit can be understood as a message scheduling relationship of the scheduling unit.
  • a scheduling unit at the user level has a link relationship with a scheduling unit at the port level.
  • the link relationship can represent a scheduling unit at the user level.
  • the message is scheduled to a scheduling unit at the port level.
  • the link relationship of the scheduling unit of each scheduling level may be configured according to the service requirements of the current service scenario.
  • the scheduling level of the HQOS scheduling model can be determined, the scheduling unit of each scheduling level is selected, and the link relationship of the scheduling units of each scheduling level is configured, thereby completing the construction of the HQOS scheduling model to meet the business requirements of the current business scenario.
  • the implementation of the method in this embodiment is very flexible, and different models can be constructed according to different scheduling scenarios, thereby meeting different user requirements and improving product competitiveness.
  • the foregoing method may be applied to any network device with a scheduling function in the HQOS technology, for example, a root node network device, a parent node network device, or a child node network device, and the like, which is not limited in this embodiment.
  • the scheduling level of the current HQOS scheduling model to be built is determined to be N level according to the service requirement of the current service scenario; and the scheduling unit of each of the N scheduling levels is selected from the storage resources; The link relationship of the scheduling unit of the scheduling level.
  • the HQOS scheduling model can be configured according to the service requirements of different service scenarios, so as to meet the service requirements of different service scenarios.
  • another scheduling model construction method includes the following steps:
  • Step S201 Determine, according to the service requirement of the current service scenario, that the scheduling level of the hierarchical quality of service HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2.
  • Step S202 Select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource.
  • step S202 may include:
  • the scheduling unit type and the number of scheduling units of each of the N scheduling levels may be calculated according to the service requirements of the current service scenario.
  • the calculation herein may include selecting, for each scheduling level, a traffic scheduling category including at least the service requirement of the current service scenario, and calculating a number of scheduling units of each scheduling unit type, where scheduling of each scheduling unit type The number of units satisfies at least the service scheduling that implements the service requirements of the current business scenario.
  • each user includes four types of services, and there are four types of service SPs. Then, at least the SP scheduling unit is included in the user level, and the SP scheduling unit has at least four.
  • step S202 in addition to selecting the scheduling unit of each of the N scheduling levels in the manner of the foregoing calculation, in this embodiment, the scheduling unit of each scheduling level may be determined by receiving an operation input by the user. The selection and the like are completed, and the embodiment is not limited.
  • Step S203 configuring a link relationship of the scheduling unit of each scheduling level.
  • step S203 may include:
  • the link relationship between the scheduling units in the respective levels may be set first, and then the link relationship between the scheduling units in the levels may be set, or the two relationships may be set at the same time, or the scheduling unit may be set between the levels first.
  • Link relationships and then set the link relationship of the scheduling unit within their respective levels.
  • the above-mentioned link relationship in which the scheduling units of each scheduling level are set in respective levels can be understood as setting a link relationship of each scheduling unit within respective levels, for example, a certain scheduling level includes three scheduling units, then, here It is possible to set a link relationship between the three scheduling units.
  • the above-mentioned link relationship between the levels of the scheduling units of each scheduling level may be understood as setting a link relationship between the levels of the scheduling unit, for example, one scheduling level includes three scheduling units, and another scheduling level includes two Scheduling units, then, here may be the scheduling relationship between the three scheduling units and the two scheduling units.
  • the foregoing method may further include the following steps:
  • Step S204 configuring priorities and/or weights of scheduling units of each scheduling level; and/or
  • Step S205 Configure traffic shaping of the scheduling unit of each scheduling level.
  • the configuration priorities and/or weights may be defined as the priority of each scheduling unit, and/or the weight of each scheduling unit.
  • the configuration herein may be configured according to service requirements.
  • the traffic shaping can be configured as the traffic shaping of each scheduling level, or the traffic shaping of each scheduling unit can be configured.
  • the configuration here can be configured according to service requirements or network conditions.
  • the priority of the configuration and/or the weight can be used to perform the scheduling according to the priority and/or the weight.
  • the traffic shaping of the scheduling unit can be controlled and controlled by the traffic shaping.
  • steps S204 and S205 may be performed after step S202.
  • steps S204 and S205 may be executed simultaneously with step S203, or may be performed before or after it. This embodiment is not limited in this embodiment.
  • the scheduling level of the HQOS scheduling model includes a queue level, where the queue level includes T scheduling units, where the T is greater than or equal to the number of services of each user in the service scenario;
  • the foregoing setting the link relationship of the scheduling units of each scheduling level in the respective levels may include:
  • a Committed Information Rate (CIR) chain and an Excess Information Rate (EIR) chain are set for each scheduling unit of the queue level.
  • the number of scheduling units that determine the queue level according to the number of services of each user in the current service scenario may be implemented.
  • the foregoing T may be set according to the number of services of each user in the foregoing service scenario. For example, each user includes five types of services, and then the T may be 5, 6, 7, or 8, and the like. Wait.
  • the scheduling level of the HQOS scheduling model may further include a user level, where the foregoing setting a link relationship between the scheduling units of each scheduling level includes:
  • the foregoing part of the scheduling unit may be a partial sub-scheduling within a scheduling unit of the user level.
  • the unit for example, the FQ8 scheduling unit is included in the foregoing user level.
  • the part of the scheduling unit may be a part of the FQ scheduling unit in the FQ8 scheduling unit, and the other part of the scheduling unit may be another part of the FQ8 scheduling unit.
  • the foregoing part of the scheduling unit may also be understood as a group of scheduling units at the user level.
  • the user level includes two FQ8 scheduling units and two WFQ scheduling units, that is, the user level includes two scheduling units, where each group of scheduling units At least one scheduling unit is included.
  • the foregoing part of the scheduling unit may be one FQ8 scheduling unit and one WFQ scheduling unit, and the foregoing part of the scheduling unit may be another FQ8 scheduling unit and another WFQ scheduling unit. That is, the foregoing steps may be implemented to set a link relationship between a CIR chain of T scheduling units of the queue group and a group of scheduling units of the user level, and set an EIR chain of the T scheduling units of the queue group and the user. The link relationship of another set of scheduling units of the level.
  • a link relationship between a CIR chain of each scheduling unit in the queue level and a group of scheduling units at the user level may be implemented, and an EIR chain of each scheduling unit and another group scheduling of the user level may be set.
  • each group of scheduling units may include one or more scheduling units, which may be set according to the service requirements of the current service scenario, for example, when there is a service SP in the service of the user. Then, at each group scheduling unit, at least an SP scheduling unit is included to implement scheduling of the SP. Or, when the user's service includes WRR in the presence, the scheduling unit in each group includes at least a WFQ scheduling unit to implement scheduling of the WRR.
  • one user level may correspond to multiple queue levels.
  • the scheduling level of the HQOS scheduling model includes a port level
  • the setting a link relationship between the levels of the scheduling unit of each scheduling level further includes:
  • a part of the scheduling unit may be a partial sub-scheduling unit in a scheduling unit of the user level.
  • the user level includes an FQ8 scheduling unit, and then part of the scheduling unit of the user level may be part of the FQ8 scheduling unit.
  • the FQ scheduling unit, and a part of the scheduling unit at the port level may be a part of the scheduling unit of a certain scheduling unit, for example, the scheduling unit at the port level is WFQ2.
  • the scheduling unit, then, the port level part of the scheduling unit may be a WFQ scheduling unit in the WFQ2 scheduling unit, and the other part of the scheduling unit is another WFQ scheduling unit in the WFQ2 scheduling unit, the two parts of the scheduling unit, and then the FQ scheduling unit Schedule.
  • the port level may include two sets of scheduling units and one port scheduling unit, where each group of scheduling units includes at least one scheduling unit, then the setting of the scheduling relationship between the scheduling units of each scheduling level may be include:
  • the setting a link relationship of the scheduling units of each scheduling level in respective levels includes:
  • a group of scheduling units at the port level may be implemented corresponding to a group of scheduling units at the user level, and another group of scheduling units at the port level may be corresponding to another group of scheduling units at the user level.
  • one port level may correspond to multiple user levels.
  • the foregoing port scheduling unit may be a scheduling unit including an SP, WFQ, or FQ scheduling unit to implement CIR and EIR for guaranteeing user services.
  • Each user in the service scenario has eight services and supports SP+WRR scheduling, among which four service SPs and four types of service WRR.
  • the port level CIR and EIR are SP, and the CIR is higher than EIR.
  • scheduling is performed according to the priority of the service, and flows of different users in the same priority are allocated according to the weight ratio.
  • the foregoing method is implemented for the service requirement of the service scenario, where the number of the various scheduling units is divided in a certain case of the total storage resource, and after the division, the various scheduling units ensure that the service scenarios are sufficient.
  • step S201 three levels of levels may be determined according to the above service requirements, which are queue level, user level, and port level.
  • step S202 can set that the queue level can include 8 queues, that is, 8 scheduling units can be selected, and in step S203, 2 chains can be set for each scheduling unit of the queue level, one of which is 1 chain. It is a CIR chain and the other chain is an EIR chain.
  • the flow shaping can be configured for the CIR chain and the EIR chain in step S205.
  • step S202 according to the above service needs, as shown in FIG. 3, setting user level needs 2 WFQ scheduling unit and 2 FQ8 scheduling units, wherein the first WFQ scheduling unit and the FQ8 scheduling unit can be understood as a group of scheduling units, and the CIR traffic is passed, and the second WFQ scheduling unit and the FQ8 scheduling unit can be understood as another A set of scheduling units, and is through EIR traffic.
  • step S203 may set the first WFQ scheduling unit to link to the lowest priority FQ scheduling unit in the first FQ8 scheduling unit, and set the second WFQ scheduling unit to link to the second FQ8 scheduling unit.
  • the lowest priority FQ scheduling unit within.
  • the flow shaping of the FQ8 scheduling unit may be configured in step S205 to implement user-level CIR and EIR shaping.
  • step S204 the priority and/or the weight of each scheduling unit at the user level may be configured.
  • step S203 may also set a CIR chain at the queue level, according to the priority level. The sequence is linked to several FQ scheduling units of high priority within the FQ8 scheduling unit. It should be noted that, in FIG. 3, the internal 8 FQ scheduling units of the FQ8 scheduling unit are represented by RR0 to RR7 and SP.
  • step S203 may also set a CIR chain at the queue level, link to the first WFQ, and configure weights at the same time.
  • step S203 can configure the link relationship of the EIR part, that is, step S203 sets the CIR chain out of the queue for several queues that need SP scheduling, and links to the high priority in the FQ8 scheduling unit according to the priority order. Several FQ scheduling units. And for several queues that need WRR scheduling, set the EIR chain out of the queue, link to the second WFQ scheduling unit, and configure the corresponding weights.
  • the scheduling unit at the port level needs to include two WFQ8 scheduling units and one FQ2 scheduling unit.
  • Each WFQ8 scheduling unit can be understood as a group of scheduling units, and the FQ2 scheduling unit can be understood as a port scheduling unit.
  • the internal 8 WFQ scheduling units of the WFQ8 scheduling unit are represented by RR0 to RR7 and SP.
  • step S203 may link the first WFQ8 scheduling unit of the user level to the first high priority FQ scheduling unit in the FQ2 scheduling unit, and the second WFQ8 scheduling unit of the user level is linked to the FQ2 scheduling unit.
  • step S205 can be configured to configure traffic shaping of the FQ2 scheduling unit at the port level.
  • shaping can also be understood as setting a peak information rate (PIR).
  • PIR peak information rate
  • step S203 may link the internal 8 FQ scheduling units of the first FQ8 scheduling unit of the user-level CIR traffic to the first WFQ8 scheduling unit in the port level according to the priority level.
  • 8 WFQ scheduling units, and step S204 can be configured for different users Different weights.
  • the eight internal FQ scheduling units of the second FQ8 scheduling unit of the user-level EIR traffic may be linked to the eight WFQ schedulings in the second WFQ8 scheduling unit of the port level according to the priority level. Units, and step S204 can configure different weights for different users.
  • HQOS scheduling models may be constructed according to different service scenarios.
  • an embodiment of the present disclosure provides a scheduling model construction apparatus, and a scheduling model construction apparatus 400, including:
  • the determining module 401 is configured to determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
  • the selecting module 402 is configured to select a scheduling unit of each of the N scheduling levels from the storage resource, where multiple types of scheduling units are preset in the storage resource;
  • the first configuration module 403 is configured to configure a link relationship of the scheduling unit of each scheduling level.
  • the selecting module 402 may be configured to calculate a scheduling unit type and a scheduling unit number of each of the N scheduling levels, and select a corresponding one of each scheduling level in the storage resource. Scheduling unit.
  • the first configuration module 403 may include:
  • the first setting unit 4031 is configured to set a link relationship of the scheduling units of each scheduling level within respective levels;
  • the second setting unit 4032 is configured to set a link relationship between the levels of the scheduling units of each scheduling level.
  • the foregoing apparatus further includes:
  • a second configuration module 404 configured to configure a priority and/or a weight of a scheduling unit of each scheduling level
  • the third configuration module 405 is configured to configure traffic shaping of the scheduling unit of each scheduling level.
  • the scheduling level of the HQOS scheduling model includes a queue level, where the queue level includes T scheduling units, where the T is greater than or equal to the number of services of each user in the service scenario;
  • the first setting unit 4031 may be configured to set a CIR chain and an EIR chain for each scheduling unit of the queue level.
  • the scheduling level of the HQOS scheduling model includes a user level
  • the second setting unit 4032 may be configured to set a CIR chain of the T scheduling units of the queue group and a part of the user level. a link relationship of the scheduling unit, and a link relationship between an EIR chain of T scheduling units of the queue group and another part of the scheduling unit of the user level.
  • the scheduling level of the HQOS scheduling model includes a port level
  • the second setting unit 4032 may be configured to set a part of the scheduling unit of the user level and a part of the scheduling unit of the port level. Linking relationship, and setting a link relationship between another part of the scheduling unit of the user level and another part of the scheduling unit of the port level.
  • the scheduling model construction device 400 may be the device in any of the embodiments of FIG. 1 to FIG. 3, for example, the device may be the device in the network device described in the foregoing embodiment.
  • the scheduling level of the current HQOS scheduling model to be built is determined to be N level according to the service requirement of the current service scenario; and the scheduling unit of each of the N scheduling levels is selected from the storage resources; The link relationship of the scheduling unit of the scheduling level.
  • the HQOS scheduling model can be configured according to the service requirements of different service scenarios, so as to meet the service requirements of different service scenarios.
  • N is a positive integer greater than or equal to 2;
  • the scheduling unit that selects each of the N scheduling levels from the storage resource includes:
  • the configuring a link relationship of the scheduling unit of each scheduling level includes:
  • the method further includes:
  • the scheduling level of the HQOS scheduling model includes a queue level, where the queue level includes T scheduling units, where the T is greater than or equal to the number of services of each user in the service scenario;
  • the setting a link relationship of the scheduling units of each scheduling level in respective levels includes:
  • a committed information rate CIR chain and an additional information rate EIR chain are set for each scheduling unit of the queue level.
  • the scheduling level of the HQOS scheduling model includes a user level
  • the setting a link relationship between the levels of the scheduling unit of each scheduling level includes:
  • the scheduling level of the HQOS scheduling model includes a port level, where the setting of a scheduling relationship between the scheduling units of each scheduling level includes:
  • the storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
  • the embodiment of the present disclosure further provides a schematic structural diagram of an electronic device.
  • the electronic device include:
  • At least one processor 70 which is exemplified by a processor 70 in FIG. 7; and a memory 71, may further include a communication interface 72 and a bus 73.
  • the processor 70, the communication interface 72, and the memory 71 can complete communication with each other through the bus 73.
  • Communication interface 72 can be used for information transfer.
  • Processor 70 can invoke logic instructions in memory 71 to perform the methods of the above-described embodiments.
  • logic instructions in the memory 71 described above may be implemented in the form of a software functional unit and sold or used as a stand-alone product, and may be stored in a computer readable storage medium.
  • the memory 71 is a computer readable storage medium, and can be used to store a software program, a computer executable program, a program instruction/module corresponding to the method in the embodiment of the present disclosure.
  • the processor 70 executes the function application and the data processing by executing software programs, instructions, and modules stored in the memory 71, that is, implementing the scheduling model construction method in the above method embodiments.
  • the memory 71 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to usage of the terminal device, and the like. Further, the memory 71 may include a high speed random access memory, and may also include a nonvolatile memory.
  • the technical solution of the embodiments of the present disclosure may be embodied in the form of a software product stored in a storage medium, including one or more instructions for causing a computer device (which may be a personal computer, a server, or a network) The device or the like) performs all or part of the steps of the method described in the embodiments of the present disclosure.
  • the foregoing storage medium may be a non-transitory storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like.
  • the scheduling model construction method and device provided by the present application can meet the business requirements of different business scenarios.

Landscapes

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

Abstract

Provided in the present application are a method and device for constructing a scheduling model. The method comprises: determining, according to a service requirement of a current service scenario, a scheduling level number of an HQOS scheduling model to be constructed as N, wherein N is a positive integer greater than or equal to 2; selecting, from storage resources, scheduling units for respective scheduling levels of the N scheduling levels, wherein each storage resource is provided with various types of scheduling units in advance; and configuring linking relationships between the scheduling units of the respective scheduling levels. Embodiments of the present disclosure can meet service requirements of different service scenarios.

Description

一种调度模型构建方法和装置Method and device for constructing scheduling model 技术领域Technical field
本申请涉及通信技术领域，例如涉及一种调度模型构建方法和装置。The present application relates to the field of communication technologies, for example, to a scheduling model construction method and apparatus.
背景技术Background technique
在通信技术领域中，随着用户规模的扩大和业务种类的增多，层次化服务质量(Hierarchical Quality Of Service，HQOS)调度技术被得到广泛应用。然而，目前网络设备的HQOS调度模型主要是用户模型和业务模型。然而，无论是用户模型，还是业务模型都是固定的，即这两种模型中调度级别和级别之间的链接关系都是固定不变的。然而，在实际应用中，不同业务场景一般存在不同的业务需求，例如：一场景中，每个用户有3种严格优先级(Strict Priority，SP)调度的业务，那么，在调度时，需要使用调度模型支持3种SP调度的网络设备；另一场景中，每个用户有4种SP调度的业务和4种业务加权循环(Weighted Round Robin，WRR)调度的业务，那么，在调度时，需要使用调度模型支持4种SP调度和4种WRR调度的网络设备。可见，目前由于网络设备的HQOS调度模型固定，导致在实际应用中，会可能存在无法满足不同业务场景的业务需求的问题。In the field of communication technology, with the expansion of user scale and the increase of service types, Hierarchical Quality Of Service (HQOS) scheduling technology has been widely used. However, the current HQOS scheduling model for network devices is mainly the user model and the business model. However, both the user model and the business model are fixed, that is, the link relationships between the scheduling levels and levels in the two models are fixed. However, in actual applications, different service scenarios generally have different service requirements. For example, in a scenario, each user has three types of Strict Priority (SP) scheduled services, so when scheduling, you need to use The scheduling model supports three types of SP-scheduled network devices. In another scenario, each user has four types of SP-scheduled services and four types of service-weighted round-robin (WRR) scheduled services. A network device supporting four SP schedulings and four WRR schedulings using a scheduling model. It can be seen that the HQOS scheduling model of the network device is fixed, which may cause problems that cannot meet the service requirements of different service scenarios.
发明内容Summary of the invention
本公开实施例提供一种调度模型构建方法和装置，解决了无法满足不同业务场景的业务需求的问题。The embodiments of the present disclosure provide a scheduling model construction method and apparatus, which solves the problem that the service requirements of different service scenarios cannot be met.
本公开实施例提供一种调度模型构建方法，包括：The embodiment of the present disclosure provides a scheduling model construction method, including:
根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数；Determining, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is an N level, where the N is a positive integer greater than or equal to 2;
从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元； Selecting, by the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource;
配置每个调度级别的调度单元的链接关系。Configure the link relationship for the scheduling unit for each scheduling level.
本公开实施例还提供一种调度模型构建装置，其特征在于，包括：The embodiment of the present disclosure further provides a scheduling model construction apparatus, which includes:
确定模块，被配置为根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数；The determining module is configured to determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
选择模块，被配置为从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元；a selection module, configured to select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource;
第一配置模块，被配置为配置每个调度级别的调度单元的链接关系。The first configuration module is configured to configure a link relationship of the scheduling unit of each scheduling level.
本公开实施例还提供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的一个或多个程序，所述一个或多个程序被所述计算机执行时使所述计算机执行如上述提供的一种调度模型构建方法。Embodiments of the present disclosure also provide a computer storage medium having stored therein one or more programs executable by a computer, the one or more programs being executed by the computer to cause the computer to perform as described above A scheduling model construction method is provided.
本公开实施例还提供了一种电子设备，包括：An embodiment of the present disclosure further provides an electronic device, including:
至少一个处理器；以及At least one processor;
与所述至少一个处理器通信连接的存储器；其中，a memory communicatively coupled to the at least one processor; wherein
所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器执行上述的方法。The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the method described above.
上述技术方案中的一个技术方案具有如下优点或有益效果：One of the above technical solutions has the following advantages or benefits:
通过根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级；从存储资源中选择N个调度级别中每个调度级别的调度单元；配置每个调度级别的调度单元的链接关系。可以实现根据不同业务场景的业务需求构建对应的HQOS调度模型，从而实现满足不同业务场景的业务需求。The scheduling level of the current HQOS scheduling model to be built is determined to be N level according to the service requirement of the current service scenario; the scheduling unit of each of the N scheduling levels is selected from the storage resources; and the scheduling unit of each scheduling level is configured. Link relationship. The HQOS scheduling model can be configured according to the service requirements of different service scenarios, so as to meet the service requirements of different service scenarios.
附图概述BRIEF abstract
图1为本公开实施例提供一种调度模型构建方法的流程示意图；FIG. 1 is a schematic flowchart diagram of a method for constructing a scheduling model according to an embodiment of the present disclosure;
图2为本公开实施例提供的另一种调度模型构建方法的流程示意图；FIG. 2 is a schematic flowchart diagram of another scheduling model construction method according to an embodiment of the present disclosure;
图3为本公开实施例提供的一种调度模型的示意图；FIG. 3 is a schematic diagram of a scheduling model according to an embodiment of the present disclosure;
图4为本公开实施例提供的一种调度模型构建装置的结构示意图； FIG. 4 is a schematic structural diagram of a scheduling model construction apparatus according to an embodiment of the present disclosure;
图5为本公开实施例提供的另一种调度模型构建装置的结构示意图；FIG. 5 is a schematic structural diagram of another scheduling model construction apparatus according to an embodiment of the present disclosure;
图6为本公开实施例提供的另一种调度模型构建装置的结构示意图；以及FIG. 6 is a schematic structural diagram of another scheduling model construction apparatus according to an embodiment of the present disclosure;
图7为本公开实施例提供的电子设备的结构示意图。FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
具体实施方式detailed description
为使本公开要解决的技术问题、技术方案和优点更加清楚，下面将结合附图及实施例进行详细描述。In order to make the technical problems, technical solutions and advantages of the present disclosure more clear, the following detailed description will be made with reference to the accompanying drawings and embodiments.
如图1所示，本公开实施例提供一种调度模型构建方法，包括以下步骤：As shown in FIG. 1 , an embodiment of the present disclosure provides a scheduling model construction method, including the following steps:
步骤S101、根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数。Step S101: Determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, where the N is a positive integer greater than or equal to 2.
该步骤中，可以实现根据当前业务场景的业务需求确定当前待构建的HQOS调度模型的调度级别，其中，上述N可以与上述业务需求对应，即业务需求与HQOS调度模型的调度级别的级别数是对应的，从而可以实现构建出的HQOS调度模型满足上述当前业务场景的业务需求。其中，这里的对应可以是预先设置好的，即步骤S101可以在预先设置好的对应关系中确定当前业务场景的业务需求对应的调度级别的级别数；或者步骤S101中还可以是接收用户输入的操作而确定的。In this step, the scheduling level of the current HQOS scheduling model to be constructed may be determined according to the service requirement of the current service scenario, where the N may correspond to the service requirement, that is, the service requirement and the level of the scheduling level of the HQOS scheduling model are Correspondingly, the constructed HQOS scheduling model can meet the business requirements of the current business scenario. Wherein, the correspondence here may be preset, that is, the step S101 may determine the number of levels of the scheduling level corresponding to the service requirement of the current service scenario in the preset relationship, or may also receive the user input in step S101. Determined by operation.
需要说明的是，本实施例中，HQOS调度模型的调度级别可以理解为HQOS调度模型的层次，且一个级别对应一个层次，例如：HQOS调度模型包括用户级和端口级，即HQOS调度模型包括用户级和端口级这两个级别层次。It should be noted that, in this embodiment, the scheduling level of the HQOS scheduling model can be understood as the level of the HQOS scheduling model, and one level corresponds to one level. For example, the HQOS scheduling model includes a user level and a port level, that is, the HQOS scheduling model includes the user. Two levels of level and port level.
步骤S102、从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元。Step S102: Select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource.
本实施例中，可以在执行步骤S102之前，在存储资源中设置多种类型的调度单元，且每个种类型的调度单元可以设置一个或者多个等等，对此本实施例不作限定。另外，本实施例中，调度单元可以理解为存储资源中的寄存器，例如：一个调度单元可以包括一个或者多个寄存器。另外，本实施例中，上述存储资源可以是网络设备的本地存储资源，该网络设备可以是实现该方法的设备。 另外，由于调度单元属于存储资源中，即上述存储资源中预先设置有多种类型的调度单元可以理解为，在存储资源中划分多种类型的调度单元。In this embodiment, multiple types of scheduling units may be set in the storage resource before the step S102 is performed, and one or more scheduling units may be set for each type of scheduling unit, which is not limited in this embodiment. In addition, in this embodiment, the scheduling unit may be understood as a register in a storage resource, for example, a scheduling unit may include one or more registers. In addition, in this embodiment, the foregoing storage resource may be a local storage resource of the network device, and the network device may be a device that implements the method. In addition, since the scheduling unit belongs to the storage resource, that is, multiple types of scheduling units are preset in the foregoing storage resource, it can be understood that multiple types of scheduling units are divided in the storage resource.
另外，本实施例中，对调度单元的种类不作限定，例如：在上述存储资源中可以设置有SP调度单元、加权公平队列(Weighted Fair Queuing，WFQ)调度单元、WFQ调度单元、WFQ2调度单元、WFQ4调度单元、WFQ8调度单元、公平队列(Fair Queuing，FQ)调度单元、FQ2调度单元、FQ4调度单元、FQ8调度单元，以及还可以根据需要继续扩充WFQ16、WFQ32......，FQ16、FQ32......等调度单元，每种类型的调度单元可以设置一个或者多个。In addition, in this embodiment, the type of the scheduling unit is not limited. For example, the SP may be configured with an SP scheduling unit, a Weighted Fair Queuing (WFQ) scheduling unit, a WFQ scheduling unit, and a WFQ2 scheduling unit. WFQ4 scheduling unit, WFQ8 scheduling unit, Fair Queuing (FQ) scheduling unit, FQ2 scheduling unit, FQ4 scheduling unit, FQ8 scheduling unit, and WFQ16, WFQ32, etc., FQ16, FQ32 ... and other scheduling units, each type of scheduling unit can be set to one or more.
其中，SP调度单元可以实现多条流之间的严格优先级调度；The SP scheduling unit can implement strict priority scheduling between multiple streams;
WFQ调度单元可以实现多条流之间的权重调度；The WFQ scheduling unit can implement weight scheduling between multiple streams;
WFQ2调度单元包括2个WFQ调度单元和1个SP调度单元的组合，实现2个优先级调度、每个优先级下的多条流之间的权重调度；The WFQ2 scheduling unit includes a combination of two WFQ scheduling units and one SP scheduling unit, and implements two priority scheduling and weight scheduling between multiple flows under each priority;
WFQ4调度单元包括4个WFQ调度单元和1个SP调度单元组合，实现4个优先级调度、每个优先级下的多条流之间的权重调度；The WFQ4 scheduling unit includes four WFQ scheduling units and one SP scheduling unit combination to implement four priority scheduling and weight scheduling between multiple streams under each priority level;
WFQ8调度单元包括8个WFQ调度单元和1个SP调度单元组合，实现8个优先级调度、每个优先级下的多条流之间的权重调度；The WFQ8 scheduling unit includes eight WFQ scheduling units and one SP scheduling unit combination, and implements eight priority scheduling and weight scheduling between multiple flows under each priority;
FQ调度单元可以实现多条流之间的权重相同，即1∶1的调度；The FQ scheduling unit can implement the same weight among multiple streams, that is, a 1:1 scheduling;
FQ2调度单元包括2个FQ调度单元和1个SP调度单元组合，实现2个优先级调度、每个优先级下的多条流之间的权重相同，即1∶1的调度；The FQ2 scheduling unit includes two FQ scheduling units and one SP scheduling unit combination to implement two priority scheduling, and the weights between the multiple streams under each priority are the same, that is, the 1:1 scheduling;
FQ4调度单元包括4个FQ调度单元和1个SP调度单元组合，实现4个优先级调度、每个优先级下的多条流之间的权重相同，即1∶1的调度；The FQ4 scheduling unit includes four FQ scheduling units and one SP scheduling unit combination to implement four priority scheduling, and the weights of the multiple flows under each priority are the same, that is, the 1:1 scheduling;
FQ8调度单元包括8个FQ调度单元和1个SP调度单元组合，实现8个优先级调度、每个优先级下的多条流之间的权重相同，即1∶1的调度。The FQ8 scheduling unit includes 8 FQ scheduling units and 1 SP scheduling unit combination to implement 8 priority scheduling, and the weights of multiple streams under each priority are the same, that is, 1:1 scheduling.
另外，在步骤S102中可以是根据当前业务场景的业务需求选择各级别的调度单元，例如：根据预设的业务需求与调度单元的对应关系选择各级别的调度单元，或者可以是接收用户输入的操作选择各级别的调度单元等等，对此不作限定。 In addition, in step S102, the scheduling unit of each level may be selected according to the service requirement of the current service scenario, for example, selecting the scheduling unit of each level according to the correspondence between the preset service requirement and the scheduling unit, or may be receiving the user input. The operation selects the scheduling unit of each level and the like, which is not limited thereto.
步骤S103、配置每个调度级别的调度单元的链接关系。Step S103: Configure a link relationship of the scheduling unit of each scheduling level.
该步骤中配置每个调度级别的调度单元的链接关系可以理解为，设置各调度级别中各调度单元的链接关系，这里的链接关系可以包括级别内部调度单元的链接关系，也可以包括级别之间的调度单元的连接关系。另外，调度单元的链接关系可以理解为调度单元的报文调度关系，例如：用户级的一调度单元与端口级的一调度单元存在链接关系，那么，该链接关系可以表示用户级的一调度单元的报文会调度给端口级的一调度单元。另外，每个调度级别的调度单元的链接关系可以是根据当前业务场景的业务需求配置的。The link relationship of the scheduling unit of each scheduling level in this step may be understood as setting a link relationship of each scheduling unit in each scheduling level, where the link relationship may include a link relationship of the level internal scheduling unit, or may include a level between The connection relationship of the scheduling unit. In addition, the link relationship of the scheduling unit can be understood as a message scheduling relationship of the scheduling unit. For example, a scheduling unit at the user level has a link relationship with a scheduling unit at the port level. Then, the link relationship can represent a scheduling unit at the user level. The message is scheduled to a scheduling unit at the port level. In addition, the link relationship of the scheduling unit of each scheduling level may be configured according to the service requirements of the current service scenario.
通过上述步骤可以实现确定HQOS调度模型的调度级别，选择各调度级别的调度单元，以及配置各调度级别的调度单元的链接关系，从而完成HQOS调度模型的构建，以满足上述当前业务场景的业务需求。另外，本实施例中方法的实现十分灵活，可以根据不同调度场景，构建不同模型，从而满足不同用户需求，提升产品竞争力。Through the foregoing steps, the scheduling level of the HQOS scheduling model can be determined, the scheduling unit of each scheduling level is selected, and the link relationship of the scheduling units of each scheduling level is configured, thereby completing the construction of the HQOS scheduling model to meet the business requirements of the current business scenario. . In addition, the implementation of the method in this embodiment is very flexible, and different models can be constructed according to different scheduling scenarios, thereby meeting different user requirements and improving product competitiveness.
本实施例中，上述方法可以应用于HQOS技术中任一具备调度功能的网络设备，例如：根节点网络设备、父节点网络设备或者子节点网络设备等等，对此本实施例不作限定。In this embodiment, the foregoing method may be applied to any network device with a scheduling function in the HQOS technology, for example, a root node network device, a parent node network device, or a child node network device, and the like, which is not limited in this embodiment.
本实施例中，通过根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级；从存储资源中选择N个调度级别中每个调度级别的调度单元；配置每个调度级别的调度单元的链接关系。可以实现根据不同业务场景的业务需求构建对应的HQOS调度模型，从而实现满足不同业务场景的业务需求。In this embodiment, the scheduling level of the current HQOS scheduling model to be built is determined to be N level according to the service requirement of the current service scenario; and the scheduling unit of each of the N scheduling levels is selected from the storage resources; The link relationship of the scheduling unit of the scheduling level. The HQOS scheduling model can be configured according to the service requirements of different service scenarios, so as to meet the service requirements of different service scenarios.
如图2所示，本公开实施例提供的另一种调度模型构建方法，包括以下步骤：As shown in FIG. 2, another scheduling model construction method provided by an embodiment of the present disclosure includes the following steps:
步骤S201、根据当前业务场景的业务需求，确定当前待构建的层次化服务质量HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数。Step S201: Determine, according to the service requirement of the current service scenario, that the scheduling level of the hierarchical quality of service HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2.
步骤S202、从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元。 Step S202: Select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource.
作为一种可选的实施方式，步骤S202可以包括：As an optional implementation manner, step S202 may include:
计算N个调度级别中每个调度级别的调度单元种类和调度单元个数，并在所述存储资源中选择每个调度级别的相应调度单元。Calculating a scheduling unit type and a scheduling unit number of each of the N scheduling levels, and selecting a corresponding scheduling unit of each scheduling level among the storage resources.
其中，上述计算N个调度级别中每个调度级别的调度单元种类和调度单元个数可以是根据上述当前业务场景的业务需求进行计算的。另外，这里的计算可以包括为每一个调度级别中选择至少包括上述当前业务场景的业务需求的业务调度种类，以及计算每个调度单元种类的调度单元的个数，这里每个调度单元种类的调度单元的个数至少满足实现上述当前业务场景的业务需求的业务调度。例如：在上述当前业务场景中每个用户包括4种业务，其中，有4种业务SP，那么，在用户级中至少包括SP调度单元，且SP调度单元至少有4个。The scheduling unit type and the number of scheduling units of each of the N scheduling levels may be calculated according to the service requirements of the current service scenario. In addition, the calculation herein may include selecting, for each scheduling level, a traffic scheduling category including at least the service requirement of the current service scenario, and calculating a number of scheduling units of each scheduling unit type, where scheduling of each scheduling unit type The number of units satisfies at least the service scheduling that implements the service requirements of the current business scenario. For example, in the current service scenario, each user includes four types of services, and there are four types of service SPs. Then, at least the SP scheduling unit is included in the user level, and the SP scheduling unit has at least four.
当然，在步骤S202除了通过上述计算的方式选择N个调度级别中每个调度级别的调度单元之外，本实施例中，还可以是通过接收用户输入的操作确定每个调度级别的调度单元，并完成选择等等，对此本实施例不作限定。Of course, in step S202, in addition to selecting the scheduling unit of each of the N scheduling levels in the manner of the foregoing calculation, in this embodiment, the scheduling unit of each scheduling level may be determined by receiving an operation input by the user. The selection and the like are completed, and the embodiment is not limited.
步骤S203、配置每个调度级别的调度单元的链接关系。Step S203, configuring a link relationship of the scheduling unit of each scheduling level.
作为一种可选的实施方式，步骤S203可以包括：As an optional implementation manner, step S203 may include:
设置每个调度级别的调度单元在各自级别内的链接关系；Setting a link relationship of scheduling units of each scheduling level within respective levels;
设置每个调度级别的调度单元在级别之间的链接关系。Set the link relationship between the scheduling units of each scheduling level.
该实施方式中，可以是先设置调度单元在各自级别内的链接关系，再设置调度单元在级别之间的链接关系，或者可以同时设置这两种关系，或者先设置调度单元在级别之间的链接关系，再设置调度单元在各自级别内的链接关系。另外，上述设置每个调度级别的调度单元在各自级别内的链接关系可以理解为，设置每个调度单元在各自级别内部的链接关系，例如：某一调度级别包括三个调度单元，那么，这里可以是设置这三个调度单元之间的链接关系。另外，上述设置每个调度级别的调度单元在级别之间的链接关系可以理解为，设置调度单元在级别之间的链接关系，例如：一调度级别包括三个调度单元，另一调度级别包括两个调度单元，那么，这里可以是调度这三个调度单元与这两个调度单元之间的链接关系。In this implementation manner, the link relationship between the scheduling units in the respective levels may be set first, and then the link relationship between the scheduling units in the levels may be set, or the two relationships may be set at the same time, or the scheduling unit may be set between the levels first. Link relationships, and then set the link relationship of the scheduling unit within their respective levels. In addition, the above-mentioned link relationship in which the scheduling units of each scheduling level are set in respective levels can be understood as setting a link relationship of each scheduling unit within respective levels, for example, a certain scheduling level includes three scheduling units, then, here It is possible to set a link relationship between the three scheduling units. In addition, the above-mentioned link relationship between the levels of the scheduling units of each scheduling level may be understood as setting a link relationship between the levels of the scheduling unit, for example, one scheduling level includes three scheduling units, and another scheduling level includes two Scheduling units, then, here may be the scheduling relationship between the three scheduling units and the two scheduling units.
作为一种可选的实施方式中，上述方法还可以包括如下步骤： As an optional implementation manner, the foregoing method may further include the following steps:
步骤S204、配置每个调度级别的调度单元的优先级和/或权重；和/或Step S204, configuring priorities and/or weights of scheduling units of each scheduling level; and/or
步骤S205、配置每个调度级别的调度单元的流量整形。Step S205: Configure traffic shaping of the scheduling unit of each scheduling level.
其中，上述配置优先级和/或权重可以理解为，配置各调度单元的优先级，和/或，配置各调度单元的权重，另外，这里的配置可以根据业务需求进行配置。上述配置流量整形可以理解为，配置各调度级别的流量整形，或者可以是配置各调度单元的流量整形，另外，这里的配置可以是根据业务需求或者网络情况进行配置的。通过上述配置优先级和/或权重可以实现根据优先级和/或权重进行调度，通过上述配置流量整形可以实现对调度单元的流量进行管控。The configuration priorities and/or weights may be defined as the priority of each scheduling unit, and/or the weight of each scheduling unit. In addition, the configuration herein may be configured according to service requirements. The traffic shaping can be configured as the traffic shaping of each scheduling level, or the traffic shaping of each scheduling unit can be configured. In addition, the configuration here can be configured according to service requirements or network conditions. The priority of the configuration and/or the weight can be used to perform the scheduling according to the priority and/or the weight. The traffic shaping of the scheduling unit can be controlled and controlled by the traffic shaping.
需要说明的是，步骤S204和S205之前的执行顺序不作限定，另外，步骤S204和S205可以是在步骤S202之后执行的，例如：S204和S205可以和步骤S203同时执行，或者在其之前或者之后执行，对此本实施例不作限定。It should be noted that the execution order before steps S204 and S205 is not limited. In addition, steps S204 and S205 may be performed after step S202. For example, S204 and S205 may be executed simultaneously with step S203, or may be performed before or after it. This embodiment is not limited in this embodiment.
作为一种可选的实施方式，上述HQOS调度模型的调度级别包括队列级，所述队列级包括T个调度单元，其中，所述T大于或者等于所述业务场景中每个用户的业务数；As an optional implementation manner, the scheduling level of the HQOS scheduling model includes a queue level, where the queue level includes T scheduling units, where the T is greater than or equal to the number of services of each user in the service scenario;
上述设置每个调度级别的调度单元在各自级别内的链接关系，可以包括：The foregoing setting the link relationship of the scheduling units of each scheduling level in the respective levels may include:
为所述队列级的每个调度单元设置承诺信息速率(Committed Information Rate，CIR)链和额外信息速率(Excess Information Rate，EIR)链。A Committed Information Rate (CIR) chain and an Excess Information Rate (EIR) chain are set for each scheduling unit of the queue level.
该实施方式中，可以实现根据上述当前业务场景中每个用户的业务数确定队列级的调度单元的数量。另外，还可以实现为每个调度单元设置CIR链和EIR链，以保证每个调度单元的报文的调度传输。另外，该实施方式中，上述T可以是根据上述业务场景中每个用户的业务数设定的，例如：每个用户包括5种业务，那么，上述T可以是5、6、7或者8等等。In this implementation manner, the number of scheduling units that determine the queue level according to the number of services of each user in the current service scenario may be implemented. In addition, it is also possible to set a CIR chain and an EIR chain for each scheduling unit to ensure scheduled transmission of packets of each scheduling unit. In addition, in this implementation manner, the foregoing T may be set according to the number of services of each user in the foregoing service scenario. For example, each user includes five types of services, and then the T may be 5, 6, 7, or 8, and the like. Wait.
可选的，该实施方式中，上述HQOS调度模型的调度级别还可以包括用户级，上述设置每个调度级别的调度单元在级别之间的链接关系，包括：Optionally, in this implementation manner, the scheduling level of the HQOS scheduling model may further include a user level, where the foregoing setting a link relationship between the scheduling units of each scheduling level includes:
设置所述队列组的T个调度单元的CIR链与所述用户级的一部分调度单元的链接关系，以及设置所述队列组的T个调度单元的EIR链与所述用户级的另一部分调度单元的链接关系。Setting a link relationship between a CIR chain of T scheduling units of the queue group and a part of scheduling units of the user level, and setting an EIR chain of T scheduling units of the queue group and another part scheduling unit of the user level Link relationship.
其中，上述一部分调度单元可以是用户级中某一调度单元内的部分子调度 单元，例如：上述用户级中包括FQ8调度单元，那么，上述一部分调度单元可以是FQ8调度单元中的一部分FQ调度单元，而另一部分调度单元可以是FQ8调度单元中的另一部分FQ调度单元。当然，上述一部分调度单元还可以理解为用户级的一组调度单元，例如：用户级包括两个FQ8调度单元和两个WFQ调度单元，即用户级包括两组调度单元，其中，每组调度单元至少包括一调度单元。那么，上述一部分调度单元可以是一个FQ8调度单元和一个WFQ调度单元，上述一部分调度单元可以是另一个FQ8调度单元和另一个WFQ调度单元。即上述步骤可以实现设置所述队列组的T个调度单元的CIR链与所述用户级的一组调度单元的链接关系，以及设置所述队列组的T个调度单元的EIR链与所述用户级的另一组调度单元的链接关系。The foregoing part of the scheduling unit may be a partial sub-scheduling within a scheduling unit of the user level. The unit, for example, the FQ8 scheduling unit is included in the foregoing user level. Then, the part of the scheduling unit may be a part of the FQ scheduling unit in the FQ8 scheduling unit, and the other part of the scheduling unit may be another part of the FQ8 scheduling unit. Certainly, the foregoing part of the scheduling unit may also be understood as a group of scheduling units at the user level. For example, the user level includes two FQ8 scheduling units and two WFQ scheduling units, that is, the user level includes two scheduling units, where each group of scheduling units At least one scheduling unit is included. Then, the foregoing part of the scheduling unit may be one FQ8 scheduling unit and one WFQ scheduling unit, and the foregoing part of the scheduling unit may be another FQ8 scheduling unit and another WFQ scheduling unit. That is, the foregoing steps may be implemented to set a link relationship between a CIR chain of T scheduling units of the queue group and a group of scheduling units of the user level, and set an EIR chain of the T scheduling units of the queue group and the user. The link relationship of another set of scheduling units of the level.
该实施方式中，可以实现设置队列级中的各调度单元的CIR链与所述用户级的一组调度单元的链接关系，以及设置各调度单元的EIR链与所述用户级的另一组调度单元的链接关系，从而可以实现队列级中的CIR链和EIR链中的报文调度至用户级中不同的调度单元。In this implementation manner, a link relationship between a CIR chain of each scheduling unit in the queue level and a group of scheduling units at the user level may be implemented, and an EIR chain of each scheduling unit and another group scheduling of the user level may be set. The link relationship of the units, so that the packets in the CIR chain and the EIR chain in the queue level can be scheduled to different scheduling units in the user level.
另外，在上述用户级中的两组调度单元中，每组调度单元可以包括一个或者多个调度单元，可以根据上述当前业务场景的业务需求进行设置，例如：用户的业务中存在业务SP时，那么，在各组调度单元至少包括SP调度单元，以实现SP的调度。或者用户的业务包括中存在WRR时，那么，在各组调度单元至少包括WFQ调度单元，以实现WRR的调度。另外，需要说明的是，本实施例中，一个用户级可以对应多个队列级。In addition, in the two sets of scheduling units in the user level, each group of scheduling units may include one or more scheduling units, which may be set according to the service requirements of the current service scenario, for example, when there is a service SP in the service of the user. Then, at each group scheduling unit, at least an SP scheduling unit is included to implement scheduling of the SP. Or, when the user's service includes WRR in the presence, the scheduling unit in each group includes at least a WFQ scheduling unit to implement scheduling of the WRR. In addition, it should be noted that, in this embodiment, one user level may correspond to multiple queue levels.
可选的，在上述实施方式中，上述HQOS调度模型的调度级别包括端口级，所述设置每个调度级别的调度单元在级别之间的链接关系，还包括：Optionally, in the foregoing implementation manner, the scheduling level of the HQOS scheduling model includes a port level, and the setting a link relationship between the levels of the scheduling unit of each scheduling level further includes:
设置所述用户级的一部分调度单元与所述端口级的一部分调度单元的链接关系，以及设置所述用户级的另一部分调度单元与所述端口级的另一部分调度单元的链接关系。And setting a link relationship between a part of the scheduling unit of the user level and a part of the scheduling unit of the port level, and setting a link relationship between another part of the scheduling unit of the user level and another part of the scheduling unit of the port level.
同理，上述一部分调度单元可以是用户级中某一调度单元内的部分子调度单元，例如：上述用户级中包括FQ8调度单元，那么，上述用户级一部分调度单元可以是FQ8调度单元中的一部分FQ调度单元，而端口级一部分调度单元可以是某一调度单元的一部分子调度单元，例如：端口级的调度单元为WFQ2 调度单元，那么，端口级一部分调度单元可以是WFQ2调度单元中的一WFQ调度单元，而另一部分调度单元是WFQ2调度单元中的另一个WFQ调度单元，这两部分调度单元，再通过FQ调度单元进行调度。For the same reason, a part of the scheduling unit may be a partial sub-scheduling unit in a scheduling unit of the user level. For example, the user level includes an FQ8 scheduling unit, and then part of the scheduling unit of the user level may be part of the FQ8 scheduling unit. The FQ scheduling unit, and a part of the scheduling unit at the port level may be a part of the scheduling unit of a certain scheduling unit, for example, the scheduling unit at the port level is WFQ2. The scheduling unit, then, the port level part of the scheduling unit may be a WFQ scheduling unit in the WFQ2 scheduling unit, and the other part of the scheduling unit is another WFQ scheduling unit in the WFQ2 scheduling unit, the two parts of the scheduling unit, and then the FQ scheduling unit Schedule.
或者所述端口级可以包括两组调度单元和一端口调度单元，其中，每组调度单元至少包括一调度单元，那么，所述设置每个调度级别的调度单元在级别之间的链接关系，可以包括：Or the port level may include two sets of scheduling units and one port scheduling unit, where each group of scheduling units includes at least one scheduling unit, then the setting of the scheduling relationship between the scheduling units of each scheduling level may be include:
设置所述用户级的一组调度单元与所述端口级的一组调度单元的链接关系，以及设置所述用户级的另一组调度单元与所述端口级的另一组调度单元的链接关系；Setting a link relationship between a group of scheduling units of the user level and a group of scheduling units at the port level, and setting a link relationship between another group of scheduling units of the user level and another group of scheduling units at the port level ;
所述设置每个调度级别的调度单元在各自级别内的链接关系，包括：The setting a link relationship of the scheduling units of each scheduling level in respective levels includes:
设置所述端口级的两组调度单元与所述端口调度单元的链接关系。And setting a link relationship between the two groups of scheduling units at the port level and the port scheduling unit.
该实施方式中，可以实现端口级的一组调度单元对应用户级的一组调度单元，以及端口级的另一组调度单元对应用户级的另一组调度单元。另外，需要说明的是，本实施例中，一个端口级可以对应多个用户级。In this implementation manner, a group of scheduling units at the port level may be implemented corresponding to a group of scheduling units at the user level, and another group of scheduling units at the port level may be corresponding to another group of scheduling units at the user level. In addition, it should be noted that, in this embodiment, one port level may correspond to multiple user levels.
另外，上述端口调度单元可以是包括SP、WFQ或者FQ调度单元的调度单元，以实现保证用户业务的CIR和EIR。In addition, the foregoing port scheduling unit may be a scheduling unit including an SP, WFQ, or FQ scheduling unit to implement CIR and EIR for guaranteeing user services.
下面以一个业务场景进行举例，该业务场景中每个用户有8种业务，支持SP+WRR调度，其中4种业务SP，且4种业务WRR。以及端口级下有多个用户，端口级CIR和EIR间是SP，CIR高于EIR。CIR或EIR内部，按照业务优先级高低进行调度，相同优先级内的不同用户的流按照权重比例分配流量。针对该业务场景的业务需求执行上述方法，其中，在存储资源总和一定情况下，划分各种调度单元的数量，且划分后，各种调度单元保证够各业务场景用就行。The following is an example of a service scenario. Each user in the service scenario has eight services and supports SP+WRR scheduling, among which four service SPs and four types of service WRR. There are multiple users at the port level. The port level CIR and EIR are SP, and the CIR is higher than EIR. Within the CIR or EIR, scheduling is performed according to the priority of the service, and flows of different users in the same priority are allocated according to the weight ratio. The foregoing method is implemented for the service requirement of the service scenario, where the number of the various scheduling units is divided in a certain case of the total storage resource, and after the division, the various scheduling units ensure that the service scenarios are sufficient.
步骤S201可以将根据上面业务需求，确定3个级别层次，分别为队列级、用户级和端口级。例如：如图3所示，步骤S202可以设定队列级可以包括8个队列，即可以选择8个调度单元，以及在步骤S203可以为队列级每个调度单元设置2个链，其中1个链是CIR链，另外1个链是EIR链。以及在步骤S205中可以对CIR链和EIR链配置流程整形。In step S201, three levels of levels may be determined according to the above service requirements, which are queue level, user level, and port level. For example, as shown in FIG. 3, step S202 can set that the queue level can include 8 queues, that is, 8 scheduling units can be selected, and in step S203, 2 chains can be set for each scheduling unit of the queue level, one of which is 1 chain. It is a CIR chain and the other chain is an EIR chain. And the flow shaping can be configured for the CIR chain and the EIR chain in step S205.
另外，在步骤S202中根据上述业务需要，如图3所示，设置用户级需要2 个WFQ调度单元和2个FQ8调度单元，其中第1个WFQ调度单元和FQ8调度单元可以理解为一组调度单元，且是通CIR流量，第2个WFQ调度单元和FQ8调度单元可以理解为另一组调度单元，且是通EIR流量。In addition, in step S202, according to the above service needs, as shown in FIG. 3, setting user level needs 2 WFQ scheduling unit and 2 FQ8 scheduling units, wherein the first WFQ scheduling unit and the FQ8 scheduling unit can be understood as a group of scheduling units, and the CIR traffic is passed, and the second WFQ scheduling unit and the FQ8 scheduling unit can be understood as another A set of scheduling units, and is through EIR traffic.
如图3所示，步骤S203可以设置第1个WFQ调度单元链接到第1个FQ8调度单元内的最低优先级的FQ调度单元，以及设置第2个WFQ调度单元链接到第2个FQ8调度单元内的最低优先级的FQ调度单元。另外，在步骤S205中可以配置FQ8调度单元的流程整形，以实现用户级的CIR和EIR的整形。As shown in FIG. 3, step S203 may set the first WFQ scheduling unit to link to the lowest priority FQ scheduling unit in the first FQ8 scheduling unit, and set the second WFQ scheduling unit to link to the second FQ8 scheduling unit. The lowest priority FQ scheduling unit within. In addition, the flow shaping of the FQ8 scheduling unit may be configured in step S205 to implement user-level CIR and EIR shaping.
步骤S204还可以配置用户级各调度单元的优先级和/或权重，另外，如图3所示，针对需要SP调度的几个队列，步骤S203还可以设置队列级的CIR链，按照优先级高低顺序，链接到FQ8调度单元内的高优先级的几个FQ调度单元内。需要说明的是，在图3中以RR0至RR7和SP表示FQ8调度单元的内部8个FQ调度单元。另外，针对需要WRR调度的几个队列，步骤S203还可以设置队列级的CIR链，链接到第1个WFQ，同时配置权重。In step S204, the priority and/or the weight of each scheduling unit at the user level may be configured. In addition, as shown in FIG. 3, for several queues that require SP scheduling, step S203 may also set a CIR chain at the queue level, according to the priority level. The sequence is linked to several FQ scheduling units of high priority within the FQ8 scheduling unit. It should be noted that, in FIG. 3, the internal 8 FQ scheduling units of the FQ8 scheduling unit are represented by RR0 to RR7 and SP. In addition, for several queues that require WRR scheduling, step S203 may also set a CIR chain at the queue level, link to the first WFQ, and configure weights at the same time.
同理，步骤S203可以配置EIR部分的链接关系，即步骤S203针对需要SP调度的几个队列，设置队列出的CIR链，按照优先级高低顺序，链接到FQ8调度单元内的高优先级的几个FQ调度单元。以及针对需要WRR调度的几个队列，设置队列出的EIR链，链接到第2个WFQ调度单元，以及配置相应的权重。Similarly, step S203 can configure the link relationship of the EIR part, that is, step S203 sets the CIR chain out of the queue for several queues that need SP scheduling, and links to the high priority in the FQ8 scheduling unit according to the priority order. Several FQ scheduling units. And for several queues that need WRR scheduling, set the EIR chain out of the queue, link to the second WFQ scheduling unit, and configure the corresponding weights.
如图3所示，根据上述业务需求可以确定端口级的调度单元需要包括2个WFQ8调度单元和1个FQ2调度单元。其中，每个WFQ8调度单元可以理解为一组调度单元，而FQ2调度单元可以理解为端口调度单元。需要说明的是，在图3中以RR0至RR7和SP表示WFQ8调度单元的内部8个WFQ调度单元。另外，步骤S203可以将用户级的第1个WFQ8调度单元链接到FQ2调度单元内的第1个高优先级的FQ调度单元，以及用户级的第2个WFQ8调度单元链接到FQ2调度单元内的第2个高优先级的FQ调度单元。另外，步骤S205可以配置端口级的FQ2调度单元的流量整形，另外，这里整形还可以理解为设置峰值信息速率(Pulse Information Rate，PIR)。As shown in FIG. 3, according to the foregoing service requirement, it may be determined that the scheduling unit at the port level needs to include two WFQ8 scheduling units and one FQ2 scheduling unit. Each WFQ8 scheduling unit can be understood as a group of scheduling units, and the FQ2 scheduling unit can be understood as a port scheduling unit. It should be noted that, in FIG. 3, the internal 8 WFQ scheduling units of the WFQ8 scheduling unit are represented by RR0 to RR7 and SP. In addition, step S203 may link the first WFQ8 scheduling unit of the user level to the first high priority FQ scheduling unit in the FQ2 scheduling unit, and the second WFQ8 scheduling unit of the user level is linked to the FQ2 scheduling unit. The second high priority FQ scheduling unit. In addition, step S205 can be configured to configure traffic shaping of the FQ2 scheduling unit at the port level. In addition, shaping can also be understood as setting a peak information rate (PIR).
另外，如图3所示，步骤S203可以将用户级的通CIR流量的第1个FQ8调度单元的内部8个FQ调度单元，分别按照优先级高低顺序链接到端口级第1个WFQ8调度单元内的8个WFQ调度单元，且步骤S204可以为不同用户配置 不同权重。同理，步骤S203可以将用户级的通EIR流量的第2个FQ8调度单元的内部8个FQ调度单元，分别按照优先级高低顺序链接到端口级第2个WFQ8调度单元内的8个WFQ调度单元，且步骤S204可以为不同用户配置不同权重。In addition, as shown in FIG. 3, step S203 may link the internal 8 FQ scheduling units of the first FQ8 scheduling unit of the user-level CIR traffic to the first WFQ8 scheduling unit in the port level according to the priority level. 8 WFQ scheduling units, and step S204 can be configured for different users Different weights. Similarly, in step S203, the eight internal FQ scheduling units of the second FQ8 scheduling unit of the user-level EIR traffic may be linked to the eight WFQ schedulings in the second WFQ8 scheduling unit of the port level according to the priority level. Units, and step S204 can configure different weights for different users.
另外，需要说明的是，上述仅是一个针对业务场景的举例，本实施例中，可以实现根据不同的业务场景构建不同的HQOS调度模型。In addition, it should be noted that the foregoing is only an example of a service scenario. In this embodiment, different HQOS scheduling models may be constructed according to different service scenarios.
如图4所示，本公开实施例提供一种调度模型构建装置，调度模型构建装置400，包括：As shown in FIG. 4, an embodiment of the present disclosure provides a scheduling model construction apparatus, and a scheduling model construction apparatus 400, including:
确定模块401，被配置为根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数；The determining module 401 is configured to determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
选择模块402，被配置为从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元；The selecting module 402 is configured to select a scheduling unit of each of the N scheduling levels from the storage resource, where multiple types of scheduling units are preset in the storage resource;
第一配置模块403，被配置为配置每个调度级别的调度单元的链接关系。The first configuration module 403 is configured to configure a link relationship of the scheduling unit of each scheduling level.
作为一种可选的实施方式，选择模块402可以被配置为计算N个调度级别中每个调度级别的调度单元种类和调度单元个数，并在所述存储资源中选择每个调度级别的相应调度单元。As an optional implementation manner, the selecting module 402 may be configured to calculate a scheduling unit type and a scheduling unit number of each of the N scheduling levels, and select a corresponding one of each scheduling level in the storage resource. Scheduling unit.
作为一种可选的实施方式，如图5所示，第一配置模块403可以包括：As an optional implementation manner, as shown in FIG. 5, the first configuration module 403 may include:
第一设置单元4031，被配置为设置每个调度级别的调度单元在各自级别内的链接关系；The first setting unit 4031 is configured to set a link relationship of the scheduling units of each scheduling level within respective levels;
第二设置单元4032，被配置为设置每个调度级别的调度单元在级别之间的链接关系。The second setting unit 4032 is configured to set a link relationship between the levels of the scheduling units of each scheduling level.
作为一种可选的实施方式，如图5所示，上述装置还包括：As an optional implementation manner, as shown in FIG. 5, the foregoing apparatus further includes:
第二配置模块404，被配置为配置每个调度级别的调度单元的优先级和/或权重；和/或a second configuration module 404 configured to configure a priority and/or a weight of a scheduling unit of each scheduling level; and/or
第三配置模块405，被配置为配置每个调度级别的调度单元的流量整形。The third configuration module 405 is configured to configure traffic shaping of the scheduling unit of each scheduling level.
作为一种可选的实施方式，上述HQOS调度模型的调度级别包括队列级，所述队列级包括T个调度单元，其中，所述T大于或者等于所述业务场景中每个用户的业务数； As an optional implementation manner, the scheduling level of the HQOS scheduling model includes a queue level, where the queue level includes T scheduling units, where the T is greater than or equal to the number of services of each user in the service scenario;
第一设置单元4031可以被配置为为所述队列级的每个调度单元设置CIR链和EIR链。The first setting unit 4031 may be configured to set a CIR chain and an EIR chain for each scheduling unit of the queue level.
可选的，该实施方式中，所述HQOS调度模型的调度级别包括用户级，第二设置单元4032可以被配置为设置所述队列组的T个调度单元的CIR链与所述用户级的一部分调度单元的链接关系，以及设置所述队列组的T个调度单元的EIR链与所述用户级的另一部分调度单元的链接关系。Optionally, in this implementation manner, the scheduling level of the HQOS scheduling model includes a user level, and the second setting unit 4032 may be configured to set a CIR chain of the T scheduling units of the queue group and a part of the user level. a link relationship of the scheduling unit, and a link relationship between an EIR chain of T scheduling units of the queue group and another part of the scheduling unit of the user level.
可选的，该实施方式中，所述HQOS调度模型的调度级别包括端口级，所述第二设置单元4032可以被配置为设置所述用户级的一部分调度单元与所述端口级的一部分调度单元的链接关系，以及设置所述用户级的另一部分调度单元与所述端口级的另一部分调度单元的链接关系。Optionally, in this implementation manner, the scheduling level of the HQOS scheduling model includes a port level, and the second setting unit 4032 may be configured to set a part of the scheduling unit of the user level and a part of the scheduling unit of the port level. Linking relationship, and setting a link relationship between another part of the scheduling unit of the user level and another part of the scheduling unit of the port level.
需要说明的是，本实施例中，调度模型构建装置400可以是图1-图3中任一实施方式的装置，例如：该装置可以是上述实施例中描述的网络设备中的装置。It should be noted that, in this embodiment, the scheduling model construction device 400 may be the device in any of the embodiments of FIG. 1 to FIG. 3, for example, the device may be the device in the network device described in the foregoing embodiment.
本实施例中，通过根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级；从存储资源中选择N个调度级别中每个调度级别的调度单元；配置每个调度级别的调度单元的链接关系。可以实现根据不同业务场景的业务需求构建对应的HQOS调度模型，从而实现满足不同业务场景的业务需求。In this embodiment, the scheduling level of the current HQOS scheduling model to be built is determined to be N level according to the service requirement of the current service scenario; and the scheduling unit of each of the N scheduling levels is selected from the storage resources; The link relationship of the scheduling unit of the scheduling level. The HQOS scheduling model can be configured according to the service requirements of different service scenarios, so as to meet the service requirements of different service scenarios.
本领域普通技术人员可以理解实现上述实施例方法的全部或者部分步骤是可以通过程序指令相关的硬件来完成，所述的程序可以存储于一计算机可读取介质中，该程序在执行时，包括以下步骤：It will be understood by those skilled in the art that all or part of the steps of implementing the above embodiments may be performed by hardware associated with program instructions, which may be stored in a computer readable medium, including when executed, including The following steps:
根据当前业务场景的业务需求，确定当前待构建的层次化服务质量HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数；Determining, according to the service requirement of the current service scenario, that the scheduling level of the hierarchical service quality HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元；Selecting, by the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource;
配置每个调度级别的调度单元的链接关系。Configure the link relationship for the scheduling unit for each scheduling level.
可选的，所述从存储资源中选择N个调度级别中每个调度级别的调度单元，包括： Optionally, the scheduling unit that selects each of the N scheduling levels from the storage resource includes:
计算N个调度级别中每个调度级别的调度单元种类和调度单元个数，并在所述存储资源中选择每个调度级别的相应调度单元。Calculating a scheduling unit type and a scheduling unit number of each of the N scheduling levels, and selecting a corresponding scheduling unit of each scheduling level among the storage resources.
可选的，所述配置每个调度级别的调度单元的链接关系，包括：Optionally, the configuring a link relationship of the scheduling unit of each scheduling level includes:
设置每个调度级别的调度单元在各自级别内的链接关系；Setting a link relationship of scheduling units of each scheduling level within respective levels;
设置每个调度级别的调度单元在级别之间的链接关系。Set the link relationship between the scheduling units of each scheduling level.
可选的，所述方法还包括：Optionally, the method further includes:
配置每个调度级别的调度单元的优先级和/或权重；和/或Configuring the priority and/or weight of the scheduling unit for each scheduling level; and/or
配置每个调度级别的调度单元的流量整形。Configure traffic shaping for the scheduling unit for each scheduling level.
可选的，所述HQOS调度模型的调度级别包括队列级，所述队列级包括T个调度单元，其中，所述T大于或者等于所述业务场景中每个用户的业务数；Optionally, the scheduling level of the HQOS scheduling model includes a queue level, where the queue level includes T scheduling units, where the T is greater than or equal to the number of services of each user in the service scenario;
所述设置每个调度级别的调度单元在各自级别内的链接关系，包括：The setting a link relationship of the scheduling units of each scheduling level in respective levels includes:
为所述队列级的每个调度单元设置承诺信息速率CIR链和额外信息速率EIR链。A committed information rate CIR chain and an additional information rate EIR chain are set for each scheduling unit of the queue level.
可选的，所述HQOS调度模型的调度级别包括用户级，所述设置每个调度级别的调度单元在级别之间的链接关系，包括：Optionally, the scheduling level of the HQOS scheduling model includes a user level, and the setting a link relationship between the levels of the scheduling unit of each scheduling level includes:
设置所述队列组的T个调度单元的CIR链与所述用户级的一部分调度单元的链接关系，以及设置所述队列组的T个调度单元的EIR链与所述用户级的另一部分调度单元的链接关系。Setting a link relationship between a CIR chain of T scheduling units of the queue group and a part of scheduling units of the user level, and setting an EIR chain of T scheduling units of the queue group and another part scheduling unit of the user level Link relationship.
可选的，所述HQOS调度模型的调度级别包括端口级，所述设置每个调度级别的调度单元在级别之间的链接关系，还包括：Optionally, the scheduling level of the HQOS scheduling model includes a port level, where the setting of a scheduling relationship between the scheduling units of each scheduling level includes:
设置所述用户级的一部分调度单元与所述端口级的一部分调度单元的链接关系，以及设置所述用户级的另一部分调度单元与所述端口级的另一部分调度单元的链接关系。And setting a link relationship between a part of the scheduling unit of the user level and a part of the scheduling unit of the port level, and setting a link relationship between another part of the scheduling unit of the user level and another part of the scheduling unit of the port level.
所述的存储介质，如只读存储器(Read-Only Memory，简称ROM)、随机存取存储器(Random Access Memory，简称RAM)、磁碟或者光盘等。The storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
本公开实施例还提供了一种电子设备的结构示意图。参见图7，该电子设备 包括：The embodiment of the present disclosure further provides a schematic structural diagram of an electronic device. Referring to Figure 7, the electronic device include:
至少一个处理器(processor)70，图7中以一个处理器70为例；和存储器(memory)71，还可以包括通信接口(Communications Interface)72和总线73。其中，处理器70、通信接口72、存储器71可以通过总线73完成相互间的通信。通信接口72可以用于信息传输。处理器70可以调用存储器71中的逻辑指令，以执行上述实施例的方法。At least one processor 70, which is exemplified by a processor 70 in FIG. 7; and a memory 71, may further include a communication interface 72 and a bus 73. The processor 70, the communication interface 72, and the memory 71 can complete communication with each other through the bus 73. Communication interface 72 can be used for information transfer. Processor 70 can invoke logic instructions in memory 71 to perform the methods of the above-described embodiments.
此外，上述的存储器71中的逻辑指令可以通过软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。In addition, the logic instructions in the memory 71 described above may be implemented in the form of a software functional unit and sold or used as a stand-alone product, and may be stored in a computer readable storage medium.
存储器71作为一种计算机可读存储介质，可用于存储软件程序、计算机可执行程序，如本公开实施例中的方法对应的程序指令/模块。处理器70通过运行存储在存储器71中的软件程序、指令以及模块，从而执行功能应用以及数据处理，即实现上述方法实施例中的调度模型构建方法。The memory 71 is a computer readable storage medium, and can be used to store a software program, a computer executable program, a program instruction/module corresponding to the method in the embodiment of the present disclosure. The processor 70 executes the function application and the data processing by executing software programs, instructions, and modules stored in the memory 71, that is, implementing the scheduling model construction method in the above method embodiments.
存储器71可包括存储程序区和存储数据区，其中，存储程序区可存储操作***、至少一个功能所需的应用程序；存储数据区可存储根据终端设备的使用所创建的数据等。此外，存储器71可以包括高速随机存取存储器，还可以包括非易失性存储器。The memory 71 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to usage of the terminal device, and the like. Further, the memory 71 may include a high speed random access memory, and may also include a nonvolatile memory.
本公开实施例的技术方案可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括一个或多个指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本公开实施例所述方法的全部或部分步骤。而前述的存储介质可以是非暂态存储介质，包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等多种可以存储程序代码的介质，也可以是暂态存储介质。The technical solution of the embodiments of the present disclosure may be embodied in the form of a software product stored in a storage medium, including one or more instructions for causing a computer device (which may be a personal computer, a server, or a network) The device or the like) performs all or part of the steps of the method described in the embodiments of the present disclosure. The foregoing storage medium may be a non-transitory storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like. A medium that can store program code, or a transitory storage medium.
以上所述是本公开的实施方式，应当指出，对于本技术领域的普通技术人 员来说，在不脱离本公开实施例的范围的前提下，还可以作出若干改进和润饰，这些改进和润饰也应视为本公开的保护范围。The above is an embodiment of the present disclosure, and it should be noted that those of ordinary skill in the art It is to be understood that a number of modifications and refinements may be made without departing from the scope of the embodiments of the present disclosure.
工业实用性Industrial applicability
本申请提供的调度模型构建方法和装置，可以满足不同业务场景的业务需求。 The scheduling model construction method and device provided by the present application can meet the business requirements of different business scenarios.

Claims (15)

  1. 一种调度模型构建方法，包括：A scheduling model construction method includes:
    根据当前业务场景的业务需求，确定当前待构建的层次化服务质量HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数；Determining, according to the service requirement of the current service scenario, that the scheduling level of the hierarchical service quality HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
    从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元；Selecting, by the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource;
    配置每个调度级别的调度单元的链接关系。Configure the link relationship for the scheduling unit for each scheduling level.
  2. 如权利要求1所述的方法，其中，所述从存储资源中选择N个调度级别中每个调度级别的调度单元，包括：The method of claim 1, wherein the selecting a scheduling unit of each of the N scheduling levels from the storage resources comprises:
    计算N个调度级别中每个调度级别的调度单元种类和调度单元个数，并在所述存储资源中选择每个调度级别的相应调度单元。Calculating a scheduling unit type and a scheduling unit number of each of the N scheduling levels, and selecting a corresponding scheduling unit of each scheduling level among the storage resources.
  3. 如权利要求1所述的方法，其中，所述配置每个调度级别的调度单元的链接关系，包括：The method of claim 1, wherein the configuring a link relationship of the scheduling unit of each scheduling level comprises:
    设置每个调度级别的调度单元在各自级别内的链接关系；Setting a link relationship of scheduling units of each scheduling level within respective levels;
    设置每个调度级别的调度单元在级别之间的链接关系。Set the link relationship between the scheduling units of each scheduling level.
  4. 如权利要求1-3中任一项所述的方法，还包括：The method of any of claims 1-3, further comprising:
    配置每个调度级别的调度单元的优先级和/或权重；和/或Configuring the priority and/or weight of the scheduling unit for each scheduling level; and/or
    配置每个调度级别的调度单元的流量整形。Configure traffic shaping for the scheduling unit for each scheduling level.
  5. 如权利要求3所述的方法，其中，所述HQOS调度模型的调度级别包括队列级，所述队列级包括T个调度单元，其中，所述T大于或者等于所述业务场景中每个用户的业务数；The method of claim 3, wherein the scheduling level of the HQOS scheduling model comprises a queue level, the queue level comprises T scheduling units, wherein the T is greater than or equal to each user of the service scenario. Number of business;
    所述设置每个调度级别的调度单元在各自级别内的链接关系，包括： The setting a link relationship of the scheduling units of each scheduling level in respective levels includes:
    为所述队列级的每个调度单元设置承诺信息速率CIR链和额外信息速率EIR链。A committed information rate CIR chain and an additional information rate EIR chain are set for each scheduling unit of the queue level.
  6. 如权利要求5所述的方法，其中，所述HQOS调度模型的调度级别包括用户级，所述设置每个调度级别的调度单元在级别之间的链接关系，包括：The method of claim 5, wherein the scheduling level of the HQOS scheduling model comprises a user level, and the setting of a linking relationship between the levels of the scheduling units of each scheduling level comprises:
    设置所述队列组的T个调度单元的CIR链与所述用户级的一部分调度单元的链接关系，以及设置所述队列组的T个调度单元的EIR链与所述用户级的另一部分调度单元的链接关系。Setting a link relationship between a CIR chain of T scheduling units of the queue group and a part of scheduling units of the user level, and setting an EIR chain of T scheduling units of the queue group and another part scheduling unit of the user level Link relationship.
  7. 如权利要求6所述的方法，其中，所述HQOS调度模型的调度级别包括端口级，所述设置每个调度级别的调度单元在级别之间的链接关系，还包括：The method of claim 6, wherein the scheduling level of the HQOS scheduling model comprises a port level, and the setting of a link relationship between the levels of the scheduling units of each scheduling level further comprises:
    设置所述用户级的一部分调度单元与所述端口级的一部分调度单元的链接关系，以及设置所述用户级的另一部分调度单元与所述端口级的另一部分调度单元的链接关系。And setting a link relationship between a part of the scheduling unit of the user level and a part of the scheduling unit of the port level, and setting a link relationship between another part of the scheduling unit of the user level and another part of the scheduling unit of the port level.
  8. 一种调度模型构建装置，包括：A scheduling model building device includes:
    确定模块，被配置为根据当前业务场景的业务需求，确定当前待构建的HQOS调度模型的调度级别为N级，其中，所述N为大于或者等于2的正整数；The determining module is configured to determine, according to the service requirement of the current service scenario, that the scheduling level of the current HQOS scheduling model to be built is N, wherein the N is a positive integer greater than or equal to 2;
    选择模块，被配置为从存储资源中选择N个调度级别中每个调度级别的调度单元，其中，所述存储资源中预先设置有多种类型的调度单元；a selection module, configured to select, from the storage resource, a scheduling unit of each of the N scheduling levels, where multiple types of scheduling units are preset in the storage resource;
    第一配置模块，被配置为配置每个调度级别的调度单元的链接关系。The first configuration module is configured to configure a link relationship of the scheduling unit of each scheduling level.
  9. 如权利要求8所述的装置，其中，所述选择模块被配置为计算N个调度级别中每个调度级别的调度单元种类和调度单元个数，并在所述存储资源中选择每个调度级别的相应调度单元。The apparatus of claim 8, wherein the selection module is configured to calculate a scheduling unit category and a scheduling unit number of each of the N scheduling levels, and select each scheduling level among the storage resources The corresponding scheduling unit.
  10. 如权利要求8所述的装置，其中，所述第一配置模块包括： The apparatus of claim 8 wherein said first configuration module comprises:
    第一设置单元，被配置为设置每个调度级别的调度单元在各自级别内的链接关系；a first setting unit configured to set a link relationship of the scheduling units of each scheduling level within respective levels;
    第二设置单元，被配置为设置每个调度级别的调度单元在级别之间的链接关系。And a second setting unit configured to set a link relationship between the levels of the scheduling units of each scheduling level.
  11. 如权利要求8-9中任一项所述的装置，还包括：The apparatus of any of claims 8-9, further comprising:
    第二配置模块，被配置为配置每个调度级别的调度单元的优先级和/或权重；和/或a second configuration module configured to configure a priority and/or a weight of a scheduling unit of each scheduling level; and/or
    第三配置模块，被配置为配置每个调度级别的调度单元的流量整形。The third configuration module is configured to configure traffic shaping of the scheduling unit of each scheduling level.
  12. 如权利要求10所述的装置，其中，所述HQOS调度模型的调度级别包括队列级，所述队列级包括T个调度单元，其中，所述T大于或者等于所述业务场景中每个用户的业务数；The apparatus of claim 10, wherein the scheduling level of the HQOS scheduling model comprises a queue level, the queue level comprises T scheduling units, wherein the T is greater than or equal to each user of the service scenario. Number of business;
    所述第一设置单元被配置为为所述队列级的每个调度单元设置CIR链和EIR链。The first setting unit is configured to set a CIR chain and an EIR chain for each scheduling unit of the queue level.
  13. 如权利要求12所述的装置，其中，所述HQOS调度模型的调度级别包括用户级，所述第二设置单元被配置为设置所述队列组的T个调度单元的CIR链与所述用户级的一部分调度单元的链接关系，以及设置所述队列组的T个调度单元的EIR链与所述用户级的另一部分调度单元的链接关系。The apparatus of claim 12, wherein the scheduling level of the HQOS scheduling model comprises a user level, the second setting unit is configured to set a CIR chain of the T scheduling units of the queue group and the user level A part of the scheduling unit's link relationship, and a link relationship between the EIR chain of the T scheduling units of the queue group and another part of the scheduling unit of the user level.
  14. 如权利要求13所述的装置，其中，所述HQOS调度模型的调度级别包括端口级，所述第二设置单元被配置为设置所述用户级的一部分调度单元与所述端口级的一部分调度单元的链接关系，以及设置所述用户级的另一部分调度单元与所述端口级的另一部分调度单元的链接关系。The apparatus of claim 13, wherein the scheduling level of the HQOS scheduling model comprises a port level, the second setting unit is configured to set a part of the scheduling unit of the user level and a part of the scheduling unit of the port level Linking relationship, and setting a link relationship between another part of the scheduling unit of the user level and another part of the scheduling unit of the port level.
  15. 一种非暂态计算机可读存储介质，存储有计算机可执行指令，所述计 算机可执行指令设置为执行权利要求1-7中任一项的方法。 A non-transitory computer readable storage medium storing computer executable instructions The computer executable instructions are arranged to perform the method of any of claims 1-7.
PCT/CN2017/087406 2016-06-08 2017-06-07 Method and device for constructing scheduling model WO2017211287A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610405506.X 2016-06-08
CN201610405506.XA CN107483361B (en) 2016-06-08 2016-06-08 Scheduling model construction method and device

Publications (1)

Publication Number Publication Date
WO2017211287A1 true WO2017211287A1 (en) 2017-12-14

Family

ID=60577581

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/087406 WO2017211287A1 (en) 2016-06-08 2017-06-07 Method and device for constructing scheduling model

Country Status (2)

Country Link
CN (1) CN107483361B (en)
WO (1) WO2017211287A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4072085A4 (en) * 2019-12-16 2023-06-28 ZTE Corporation Traffic scheduling method, electronic device and storage medium

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112804162B (en) * 2019-11-13 2024-04-09 深圳市中兴微电子技术有限公司 Scheduling method, scheduling device, terminal equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103067308A (en) * 2012-12-26 2013-04-24 中兴通讯股份有限公司 Method and system for bandwidth distribution
CN103595654A (en) * 2013-11-01 2014-02-19 福建星网锐捷网络有限公司 HQoS implementation method, device and network equipment based on multi-core CPUs
CN103997468A (en) * 2014-05-23 2014-08-20 华为技术有限公司 Data scheduling method and device
WO2015154525A1 (en) * 2014-09-05 2015-10-15 中兴通讯股份有限公司 Method and device for protecting hqos using multiple board cards

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050047425A1 (en) * 2003-09-03 2005-03-03 Yonghe Liu Hierarchical scheduling for communications systems
US7675926B2 (en) * 2004-05-05 2010-03-09 Cisco Technology, Inc. Hierarchical QoS behavioral model
US8693489B2 (en) * 2011-04-28 2014-04-08 Alcatel Lucent Hierarchical profiled scheduling and shaping
CN102769566B (en) * 2012-07-13 2016-04-13 中兴通讯股份有限公司 A kind of Multi-stage scheduling system configuration, variation and device
CN104348751B (en) * 2013-07-31 2019-03-12 中兴通讯股份有限公司 Virtual output queue authorization management method and device
CN104378309B (en) * 2013-08-16 2019-05-21 中兴通讯股份有限公司 Method, system and the relevant device of QoS are realized in OpenFlow network

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103067308A (en) * 2012-12-26 2013-04-24 中兴通讯股份有限公司 Method and system for bandwidth distribution
CN103595654A (en) * 2013-11-01 2014-02-19 福建星网锐捷网络有限公司 HQoS implementation method, device and network equipment based on multi-core CPUs
CN103997468A (en) * 2014-05-23 2014-08-20 华为技术有限公司 Data scheduling method and device
WO2015154525A1 (en) * 2014-09-05 2015-10-15 中兴通讯股份有限公司 Method and device for protecting hqos using multiple board cards

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4072085A4 (en) * 2019-12-16 2023-06-28 ZTE Corporation Traffic scheduling method, electronic device and storage medium

Also Published As

Publication number Publication date
CN107483361A (en) 2017-12-15
CN107483361B (en) 2022-04-15

Similar Documents

Publication Publication Date Title
EP3553657A1 (en) Method and device for allocating distributed system task
US10620993B2 (en) Automated generation of scheduling algorithms based on task relevance assessment
US9584430B2 (en) Traffic scheduling device
Zhang et al. Task-and network-level schedule co-synthesis of Ethernet-based time-triggered systems
CN111030835B (en) Task scheduling model of TTFC network and message scheduling table generation method
CN104580396A (en) Task scheduling method, node and system
Zhou et al. Toward blind scheduling in mobile media cloud: Fairness, simplicity, and asymptotic optimality
CN1989738A (en) Propagation of minimum guaranteed scheduling rates
RU2643666C2 (en) Method and device to control virtual output queue authorization and also computer storage media
Rath et al. Cross layer based QoS platform for multimedia transmission in MANET
CN112468412A (en) Method for generating schedules for mixed-critical computer networks
CN112805971B (en) Traffic shaping method and related equipment
WO2017211287A1 (en) Method and device for constructing scheduling model
Kumar et al. A delay-optimal packet scheduler for M2M uplink
Li et al. Endpoint-flexible coflow scheduling across geo-distributed datacenters
Ren et al. Joint optimization of VNF placement and flow scheduling in mobile core network
WO2022062648A1 (en) Automatic driving simulation task scheduling method and apparatus, device, and readable medium
CN100477630C (en) Queue dispatching method and apparatus in data network
Kogan et al. Online scheduling FIFO policies with admission and push-out
Antequera et al. ADON: Application-driven overlay network-as-a-service for data-intensive science
Min et al. Reinforcement learning based routing for time-aware shaper scheduling in time-sensitive networks
Tawk et al. Optimal scheduling and delay analysis for AFDX end-systems
Balogh et al. Average bandwidth allocation model of WFQ
WO2020194322A1 (en) Apparatus, program, and method, for resource control
WO2016000326A1 (en) Method and apparatus for scheduling user queue and computer storage medium

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: 17809726

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17809726

Country of ref document: EP

Kind code of ref document: A1