WO2016070559A1 - 一种实现vnf弹性伸缩的方法及装置 - Google Patents

一种实现vnf弹性伸缩的方法及装置 Download PDF

Info

Publication number
WO2016070559A1
WO2016070559A1 PCT/CN2015/075714 CN2015075714W WO2016070559A1 WO 2016070559 A1 WO2016070559 A1 WO 2016070559A1 CN 2015075714 W CN2015075714 W CN 2015075714W WO 2016070559 A1 WO2016070559 A1 WO 2016070559A1
Authority
WO
WIPO (PCT)
Prior art keywords
vnf
vnfc
vnfm
migration
service data
Prior art date
Application number
PCT/CN2015/075714
Other languages
English (en)
French (fr)
Inventor
陶运锋
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 WO2016070559A1 publication Critical patent/WO2016070559A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/40—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using virtualisation of network functions or resources, e.g. SDN or NFV entities

Definitions

  • NFV Network Function Virtualization
  • NFV technology refers to the traditional network functions, such as: Proxy call session control function (P-CSCF), Interrogating call session control function (I-CSCF), service
  • P-CSCF Proxy call session control function
  • I-CSCF Interrogating call session control function
  • S-CSCF Serving call session control function
  • HSS Home Subscriber Server
  • AS Application Server
  • VNF Virtualized Network Function Component
  • the benefits of virtualizing the network function are: network function and hardware decoupling.
  • the heterogeneous network function is realized, and the network functions of different manufacturers are running on the same hardware.
  • the operator When the operator deploys the new network function, the operator can conveniently benefit the old and protect the investment of the operator; since the network function can run on the general hardware. Speeding up the speed at which operators can start new business; Using common hardware, large-scale procurement can reduce costs; because the network function has elastic scaling function, the network function needs to occupy resources, improve resource utilization, and at the same time, when the load is low, some general-purpose servers will be shut down, so it is more green and energy-saving. .
  • the European Telecommunications Standards Institute has developed the telecommunication NFV protocol standard, defines an end-to-end reference architecture in the ETSI NFV technical specification, and introduces a network function virtualization orchestrator (NFVO) in the end-to-end reference architecture.
  • NFVO network function virtualization orchestrator
  • Network Functions Virtualization Orchestrator also known as Orchestrator
  • VNFM Virtual Network Function Manager
  • VIM Virtual Infrastructure Manager
  • NFVI Network Functions Virtualization Infrastructure
  • VNF Functional units such as Virtualized Network Function
  • NFVO is mainly responsible for the network service, virtual resources and physical resource scheduling and related management functions of the whole network
  • VNFM is mainly responsible for the life cycle management of the network element VNF and the allocation and management of virtual resources related to VNF
  • VIM is mainly responsible for the virtualization foundation.
  • Facility management the main function is to realize the allocation and management of the entire infrastructure layer resources including computing, storage and network resources
  • NFVI as the virtualization resource layer, including the newly added virtualization platform and original hardware resources after virtualization
  • VNF The virtualized network element is deployed on the NFVI and performs the network element functions defined by the 3GPP. The functions are consistent with those of the non-virtualization.
  • the elastic expansion method is divided into horizontal out scale inversion and scale up/scale down.
  • trigger modes for elastic scaling one is automatic triggering, that is, the VNF dynamically adjusts its own resource usage according to its own load condition. This method is also called automatic bombing; the other is manual triggering, that is, the user bases Management needs to manually adjust the resource usage of the VNF.
  • the VNF elastic scaling process is defined in Appendix B.5 of the ETSI specification NFV-MAN001v039-clean.doc, where B.5.3 describes the elastic scaling of resource allocation performed by the Network Functions Virtualization Orchestrator (NFVO).
  • NFVO Network Functions Virtualization Orchestrator
  • B.5.4 describes the elastic scaling process for resource allocation performed by the VNF Manager (VNFM, VNF Manager), B.5.4.1 describes the process of VNF automatic bombing, and B.5.4.2 describes the management by network element ( EM, Element management)
  • VNFM VNF Manager
  • B.5.4.1 describes the process of VNF automatic bombing
  • B.5.4.2 describes the management by network element ( EM, Element management)
  • EM Element management
  • the process of VNF automatic bombing, B.5.4.3 describes the process of VNF auto-shrinking, and B.5.4.4 describes the VNF auto-shrinking process initiated by EM. It should be specially noted that the processes described in B5.3 and B5.4 do not distinguish between scale out/scale in or scale up/scale down, that is, for both lateral elastic expansion and longitudinal elastic expansion.
  • FIG. 1 is a schematic diagram of an elastic scaling process performed by NFVO to perform resource allocation in a related specification, as shown in FIG. 1 , including the following steps:
  • Step 100 A sender (Sender), such as an Operations Support System (OSS), sends an elastic scaling request to the NFVO.
  • Sender such as an Operations Support System (OSS)
  • OSS Operations Support System
  • Step 101 The NFVO checks the validity of the elastic scaling request to check whether the requirement meets the policy configuration.
  • Step 102 The NFVO finds the VNFM corresponding to the VNF currently requesting the elastic extension. If the VNF does not have a corresponding VNFM, the NFVO creates a new VNFM.
  • Step 103 Before performing the elastic scaling, the NFVO performs feasibility check on the elastic scaling request and reserves related resources. It should be noted that the implementation of the entire step 103 is optional, but if the step 103 is selected to be performed, it must be all executed;
  • step 103 includes:
  • Step 103a The NFVO requests the VNFM to perform a feasibility check on the elastic scalability request.
  • Step 103b The VNFM performs any necessary preparations, which may include modifying or supplementing the input parameters of the elastic scaling request according to the specific restrictions of the VNF, such as checking whether the input parameters are within the legal range, and if not within the legal range, the parameters may be performed. Modify, etc.
  • Step 103c The VNFM returns an updated elastic scaling request to the NFVO.
  • Step 103d The NFVO requests the VIM to check the availability of resources such as computing, storage, and network resources, and reserve these resources;
  • Step 103e The VIM checks the availability of resources such as computing, storage, and network resources, and reserves these resources;
  • Step 103f The result that the VIM returns the resource reservation to the NFVO;
  • Step 104 The NFVO sends an elastic scaling request to the VNFM, where the elastic scaling data is carried. It should be noted that if step 103 has been performed, resource reservation information is also carried;
  • Step 105 The VNFM performs any required preparations, such as verifying the elastic scalability request, parameter verification, etc., and may also include modifying or supplementing the input of the elastic scaling request according to the specific limitation of the VNF;
  • step 103 is consistent with step 103b. If step 103 has been performed, this step is not performed;
  • Step 106 The VNFM replies to the NFVO with a resource change response.
  • Step 107 NFVO sends a resource change request such as calculation, storage, network, etc. to the VIM;
  • Step 108 The VIM modifies the VNF internal network as needed
  • Step 109 The VIM instantiates the new virtual machine according to the need and connects to the internal network.
  • Step 110 The VIM returns a resource change completion response to the NFVO.
  • Step 111 NFVO notifies that the VNFM resource change is completed
  • Step 112 The VNFM responds to the NFVO response
  • Step 113 NFVO responds to the request to respond to the flexible application successfully.
  • FIG 2 is a schematic flow chart of the VNF automatic bomb in the relevant specification, as shown in Figure 2, including:
  • Step 200 The VNFM collects measurement data from the VNF (application related);
  • Step 201 The VNFM detects that the system load exceeds the elastic extension threshold, triggering the elastic extension request;
  • Step 202 The VNFM requests the NFVO to verify the elastic extension request.
  • Step 203 The NFVO decides to perform the elastic extension. According to the capacity database of the VNF and the VNF template, the resources required for the elastic extension such as calculation, storage, network, and the like are calculated, and whether there are sufficient resources currently checked;
  • Step 204 The NFVO authorizes the VNFM to flexibly extend the VNF;
  • Step 205 The VNFM sends a request to the VIM to create and start a virtual machine (VM, Virtual Machine) according to the indication of the NFVO.
  • VM Virtual Machine
  • Step 206 The VIM instantiates and starts a new VM, and returns a successful response to the VNFM.
  • Step 207 The VNFM configures related data on the VNF.
  • Step 208 The VNFM returns a resilient response to the NFVO.
  • Step 209 The VNFM notifies the EM that the VNF capacity has changed
  • Step 210 The NFVO saves a mapping relationship between the VNF and the VIM resource pool.
  • Step 211 EM and VNFM update VNF related data
  • Step 212 The EM configures VNF related application data.
  • FIG. 3 is a schematic flowchart of an EM-initiated VNF automatic bomb in a related specification, as shown in FIG. 3, including the following steps:
  • Step 300 The manual operator manually performs the elastic extension request on the EM, or the EM is based on the inspection.
  • the test result triggers an automatic elastic stretch request.
  • Step 301 The EM sends an elastic extension request to the VNFM.
  • Step 302 The VNFM requests the NFVO to verify the elastic extension request.
  • Step 303 The NFVO determines to perform the elastic extension. According to the capacity database of the VNF and the VNF template, the resources required for the elastic extension such as calculation, storage, network, and the like are calculated, and whether there are sufficient resources at present;
  • Step 304 The NFVO authorizes the VNFM to flexibly extend the VNF;
  • Step 305 The VNFM sends a request to the VIM to create and start the VM according to the indication of the NFVO.
  • Step 306 The VIM instantiates and starts a new VM, and returns a successful response to the VNFM.
  • Step 307 The VNFM configures related data on the VNF.
  • Step 308 The VNFM returns a resilient response to the NFVO.
  • Step 309 The VNFM notifies the EM that the VNF capacity of the elastic request is currently changed.
  • Step 310 The NFVO saves a mapping relationship between the VNF and the VIM resource pool.
  • Step 311 EM and VNFM update VNF related data
  • Step 312 The EM configures VNF related application data.
  • FIG. 4 is a schematic diagram of a VNF auto-shrinking process in a related specification, as shown in FIG. 4, including:
  • Step 400 The VNFM collects measurement data from the VNF (application related);
  • Step 401 The VNFM detects that the system load is lower than the elastic contraction threshold, and triggers the elastic contraction request;
  • Step 402 The VNFM requests the NFVO to check the elastic contraction request.
  • Step 403 The NFVO determines to perform the elastic contraction according to the configuration policy.
  • Step 404 NFVO authorizes the VNFM to elastically shrink the VNF
  • Step 405 The VNF gracefully closes the VNFC, and these VNFCAs have no effect on the ongoing business
  • Step 406 When there is no load on the VNFC (VM) to be deleted, the VNFM requests the VIM to delete the VM;
  • Step 407 The VIM deletes the VM and notifies the VNFM.
  • Step 408 The VNFM notifies the NFVO that the elastic contraction is successful
  • Step 409 The VNFM notifies the EM that the VNF capacity is elastically contracted successfully
  • Step 410 NFVO updates the mapping relationship between the VNF and the VIM resource pool.
  • Step 411 The EM and VNFM update the VNF related data.
  • FIG. 5 is a schematic flowchart of an EM-initiated VNF auto-reduction in a related specification, as shown in FIG. 5, including the following steps:
  • Step 500 The manual operator manually performs an elastic contraction request on the EM, or the EM triggers an automatic elastic contraction request according to the detection result;
  • Step 501 The EM sends a trigger elastic contract request to the VNFM.
  • Step 502 The VNFM requests the NFVO to check the elastic contraction request.
  • Step 503 NFVO decides to perform elastic contraction according to the configuration policy.
  • Step 504 The NFVO authorizes the VNFM to elastically shrink the VNF;
  • Step 505 The VNF gracefully closes the VNFC, and these VNFCAs have no effect on the ongoing business
  • Step 506 When there is no load on the VNFC (VM) to be deleted, the VNFM requests the VIM to delete the VM;
  • Step 507 The VIM deletes the VM and notifies the VNFM.
  • Step 508 The VNFM notifies the NFVO of the elastic contraction success
  • Step 509 The VNFM notifies the EM that the VNF capacity is elastically contracted successfully
  • Step 510 The NFVO updates the mapping relationship between the VNF and the VIM resource pool.
  • Step 511 EM and VNFM update VNF related data.
  • This paper provides a method and device for implementing VNF elastic expansion, which enables VNF to flexibly scale out and scale in to avoid the impact on ongoing business.
  • a method for realizing virtual network function VNF elastic expansion comprising:
  • the VNF migrates the stateful service data on the other VNFC under its own jurisdiction to the newly created VNFC according to the preset balancing policy;
  • the VNF migrates the stateful service data on the VNFC to be logged out to other VNFCs under its jurisdiction according to a preset balancing policy.
  • the migrating the stateful service data on the other VNFC to the newly created VNFC includes:
  • the VNFM After the network function virtualization orchestrator NFVO or the virtual network function management VNFM creates a new VM corresponding to the VNFC through the virtual infrastructure management VIM, the VNFM sends a migration stateful service data request to the VNF;
  • the VNF migrates stateful service data on other VNFCs under its own jurisdiction to the newly entered VNFC according to a preset balancing policy.
  • the method further includes: the VNF returns a migration request success response to the VNFM.
  • the method further includes: the VNF returns a migration request success response to the VNFM.
  • the migrating the stateful service data on the VNFC to be withdrawn from the service to the other VNFC includes:
  • the VNFM Before the NFVO or VNFM deletes the VM corresponding to the VNFC to be withdrawn from the service through the VIM, the VNFM sends a migration stateful service data request to the VNF;
  • the VNF migrates the stateful service data on the VNFC to be withdrawn from the service to other VNFCs under the jurisdiction of the VNF according to a preset balancing policy.
  • the method further includes: the VNF returns a migration request success response to the VNFM.
  • the method further includes: the NFVO or the VNFM deleting the VM corresponding to the VNFC of the service to be exited by the virtual network function management VIM.
  • the equalization policy is used to ensure load balancing between multiple VNFCs, including one or more of the following: a network association IP traffic balance between multiple VNFCs, an online session balance between multiple VNFCs, The balance of the number of registered users between multiple VNFCs.
  • a device for realizing VNF elastic expansion and contracting comprising a setting module and a processing module; wherein
  • the setting module is configured with an equalization strategy
  • the processing module is configured to: when the VM corresponding to the newly created VNFC enters the service, migrate the stateful service data on the other VNFCs to the newly created VNFC according to the preset balancing policy; and delete the VNFC corresponding to the service to be exited. Before the VM, the stateful service data on the VNFC to be logged out is migrated to other VNFCs according to a preset balancing policy.
  • the processing module is configured to: receive the migrated stateful service data request, and migrate the stateful service data on the other VNFC to the newly entered VNFC according to the preset balancing policy;
  • the processing module is configured to: receive the migrated stateful service data request, and migrate the stateful service data on the VNFC to be withdrawn from the service to another VNFC according to the preset balancing policy; when the VNF completes the migration After the work, the migration request is successfully returned to the VNFM.
  • the device is disposed in the VNF; the other VNFC is a VNFC under the jurisdiction of the VNF.
  • the embodiment of the invention further provides a computer readable storage medium storing program instructions, which can be implemented when the program instructions are executed.
  • the technical solution of the present application includes: when the VM corresponding to the newly created VNFC enters the service, the VNF migrates the stateful service data on other VNFCs under its jurisdiction to the newly created VNFC according to the preset balancing policy. Before deleting the VM corresponding to the VNFC to be withdrawn from the service, the VNF migrates the stateful service data on the VNFC to be withdrawn from the service to other VNFCs under the jurisdiction of the VNF according to a preset balancing policy.
  • the elastic expansion implementation scheme in the relevant specification is improved, and the elegant scale out and scale in elastic expansion are realized, thereby avoiding the impact on the ongoing service and improving the availability of the VNF.
  • FIG. 1 is a schematic diagram of an elastic scaling process performed by NFVO to perform resource allocation in a related specification
  • FIG. 2 is a schematic flow chart of a VNF automatic bomb in a related specification
  • FIG. 3 is a schematic flow chart of an EM-initiated VNF automatic bomb in a related specification
  • FIG. 5 is a schematic flowchart of an EM-initiated VNF auto-reduction in a related specification
  • FIG. 6 is a flowchart of a method for implementing VNF elastic expansion according to an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a device for realizing VNF elastic expansion and contraction according to an embodiment of the present invention.
  • the VNF needs to be gracefully scaled out and scaled in order to avoid impact on the ongoing services.
  • the P-CSCF, I-CSCF, and S-CSCF in the telecom field are elastically stretched, the telecommunications Operators generally do not have an impact on ongoing calls, registrations, etc.
  • the traffic on the other VNFCs needs to be gradually migrated to the new VNFC to ensure that the load on multiple VNFCs in the VNF is balanced. If the balance is unbalanced, the service on the high-load VNFC may be lost.
  • the VNFC may have the user registration data or the stateful service information such as the online session. If the virtual machine corresponding to the VNFC is forcibly deleted through the VIM, the virtual machine may be deleted. Stateful business data such as registration data or online sessions is lost, affecting user experience and reducing system availability.
  • FIG. 6 is a flowchart of a method for implementing VNF elastic expansion according to an embodiment of the present invention. As shown in FIG. 6, the method includes:
  • Step 600 When the VM corresponding to the newly created VNFC enters the service, the VNF migrates the stateful service data on the other VNFC under its jurisdiction to the newly created VNFC according to the preset balancing policy.
  • the migration of stateful service data on other VNFCs to the newly created VNFC in this step includes: VNF gradually migrates traffic on other VNFCs with higher load, such as registration data and online session data, to the newly added VNFC. In order to achieve the purpose of load balancing, so as to avoid excessive load on other VNFCs and cause business losses.
  • this step includes:
  • the VNFM After NFVO or VNFM creates a VM corresponding to the new VNFC through VIM, that is, during the scale out process, the VNFM sends a migration stateful service data request to the VNF, requesting the VNF to perform its own under the pre-set equilibrium policy.
  • the stateful business data on other VNFCs is balanced and migrated to the newly entered VNFC.
  • the migration stateful service data request carries an identifier for indicating that the current elastic extension process.
  • the VNF After the VNF completes the migration work, the VNF returns a migration request to the VNFM to successfully respond;
  • the migration time is relatively long, and the VNF can choose to reply to the VNFM in time, that is, when the VNF starts the migration work, the VNF returns a migration request to the VNFM to successfully respond.
  • the equalization policy in this step is used to ensure load balancing between multiple VNFCs, including but not limited to one or more of the following: a protocol for interconnecting networks between multiple VNFCs (IP, referred to as For network associations, traffic balancing, online session balancing between multiple VNFCs, and equalization of the number of registered users between multiple VNFCs.
  • IP protocol for interconnecting networks between multiple VNFCs
  • Step 601 Before deleting the VM corresponding to the VNFC to be withdrawn from the service, the VNF migrates the stateful service data on the VNFC to be withdrawn from the service to other VNFCs under its own jurisdiction according to the preset balancing policy.
  • the migration of stateful service data on the VNFC that exits the service to other VNFCs in this step includes: balancing the traffic on the VNFC to be withdrawn from the service, such as registration data and online session data, to the VNF itself.
  • the VM corresponding to the service VNFC is deleted by the VIM, ensuring the elegant exit of the VNFC.
  • this step includes:
  • the VNFM Before the NFVO or VNFM is about to delete the VM corresponding to the VNFC to be withdrawn from the service through the VIM, that is, during the scale in process, the VNFM sends a migration stateful service data request to the VNF, requesting the VNF to leave the service according to the preset balancing policy.
  • the stateful business data on the VNFC is balancedly migrated to other VNFCs under the VNF's own jurisdiction.
  • the migration stateful service data request carries an identifier indicating that the current elastic contraction process is performed.
  • the VNF After the VNF completes the migration, the VNF returns a migration request to the VNFM to successfully respond, and the NFVO or VNFM can delete the VM corresponding to the VNFC to be withdrawn from the service through the VIM.
  • steps 600 and 601 are not strictly sequential.
  • the elastic expansion implementation scheme in the relevant specification is improved, and the elegant scale out and scale in elastic expansion are realized, thereby avoiding the impact on the ongoing service and improving the availability of the VNF. .
  • the present example improves the scale out elastic extension, and in step 110, the VIM returns a resource change completion response to the NFVO, and step 111 That is, NFVO notifies VNFM resource change completion between Also includes:
  • the VNFM sends a migrated stateful service data request to the VNF, requesting the VNF to balance the stateful service data on the other VNFCs to the newly entered VNFC;
  • the VNF balances the stateful service data of other VNFCs, such as session or registered user data, to the newly created VNFC according to a preset balancing policy.
  • the VNF returns a migration request to the VNFM for a successful response.
  • the VNF can return a successful response to the migration request after the migration is complete.
  • the VNF can also return a successful migration request to the VNFM immediately after starting the migration work, because the data to be migrated may be many.
  • this example improves the scale in elasticity reduction.
  • the virtual machine needs to be deleted, so it is not necessary.
  • the resource reservation process is performed, so that the step 103d to the step 103f are deleted in FIG. 1 , and the NFVO sends the elastic scaling request to the VNFM in step 104, and the step 105, that is, the VNFM performs any required preparation work, the method further includes:
  • the VNFM sends the migrated stateful service data request to the VNF, and requests the VNF to balance the stateful service data on the VNFC to be withdrawn from the service to other VNFCs;
  • the VNF migrates the stateful service data on the VNFC to be withdrawn from the service to other VNFCs according to the preset balancing policy.
  • the VNF returns a migration request to the VNFM for a successful response.
  • the example improves the scale out elastic extension therein, and replaces step 207 in FIG. 2 with:
  • the VNFM configures the relevant data on the VNF.
  • the VNFM sends the migrated stateful service data request to the VNF, requesting the VNF to migrate the stateful service data on the other VNFCs to the new access service.
  • VNFC
  • the VNF balances the stateful service data of other VNFCs, such as session or registered user data, to the newly created VNFC according to a preset balancing policy.
  • the VNF returns a successful response to the migration request to the VNFM.
  • the VNF can return a successful response to the migration request after the migration is completed.
  • the VNF can also return a successful migration request to the VNFM immediately after starting the migration work, because the data to be migrated may be many.
  • the example improves the scale out elastic extension therein, and the step 307 in FIG. 3 is modified to:
  • the VNFM configures the relevant data on the VNF, and the VNFM sends a migration stateful service data request to the VNF, requesting the VNF to balance the stateful service data on the other VNFCs to the newly entered VNFC;
  • the VNF balances the stateful service data on other VNFCs, such as session or registered user data, to the new VNFC according to a preset balancing policy.
  • the VNF returns a migration request to the VNFM for a successful response.
  • the VNF can return a successful response to the migration request after the migration is complete.
  • the VNF can also return a successful migration request to the VNFM immediately after starting the migration work, because the data to be migrated may be many.
  • step 404 shown in FIG. 4 the NFVO authorizes the VNFM to elastically shrink the VNF, and step 405 That is, VNF gracefully closes between VNFC and also includes:
  • the VNFM sends the migrated stateful service data request to the VNF, and requests the VNF to balance the stateful service data on the VNFC to be withdrawn from the service to other VNFCs;
  • the VNF migrates the stateful service data on the VNFC to be withdrawn from the service to other VNFCs according to the preset balancing policy.
  • the VNF After the migration of the data is completed, the VNF returns a migration request to the VNFM for a successful response.
  • step 504 shown in FIG. 5 the NFVO authorizes the VNFM to elastically shrink the VNF.
  • step 505 that is, VNF gracefully closes VNFC, it also includes:
  • the VNFM sends the migrated stateful service data request to the VNF, and requests the VNF to balance the stateful service data on the VNFC to be withdrawn from the service to other VNFCs;
  • the VNF migrates the stateful service data on the VNFC to be withdrawn from the service to other VNFCs according to the preset balancing policy.
  • the VNF After the migration of the data is completed, the VNF returns a migration request to the VNFM for a successful response.
  • FIG. 7 is a schematic structural diagram of a device for implementing VNF elastic expansion and contraction according to an embodiment of the present invention. As shown in FIG. 7 , at least a setting module and a processing module are included, where:
  • the setting module is configured with an equalization strategy
  • the processing module is configured to: when the VM corresponding to the newly created VNFC enters the service, migrate the stateful service data on the other VNFCs to the newly created VNFC according to the preset balancing policy; and delete the VNFC corresponding to the service to be exited. Before the VM, the stateful service data on the VNFC to be logged out is migrated to other VNFCs according to a preset balancing policy.
  • the processing module is set to:
  • the migration request is successfully returned to the VNFM.
  • the processing module is configured to: receive the migrated stateful service data request, and balance the stateful service data on the VNFC to be withdrawn from the service to other VNFCs according to the preset balancing policy; when the VNF completes the migration work, VNFM returns a successful response to the migration request.
  • the above device can be placed in the VNF.
  • the elastic expansion implementation scheme in the relevant specification is improved, and the elegant scale out and scale in elastic expansion are realized, thereby avoiding the impact on the ongoing service and improving the availability of the VNF.

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

一种实现VNF弹性伸缩的方法及装置，所述方法包括：当新创建的VNFC对应的VM进入服务时，VNF按照预先设置的均衡策略将自身管辖下的其他VNFC上的有状态的业务数据迁移到新创建的VNFC；当删除待退出服务的VNFC对应的VM前，VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到自身管辖下的其他VNFC。所述装置包括设置模块和处理模块。

Description

一种实现VNF弹性伸缩的方法及装置 技术领域
本文涉及网络功能虚拟化(NFV，Network Functions Virtualization)技术，尤指一种实现虚拟网络功能(VNF，Virtualized Network Function)弹性伸缩的方法及装置。
背景技术
NFV技术指的是，将传统网络功能，比如：代理呼叫会话控制功能实体(P-CSCF，Proxy call session control Function)，查询呼叫会话控制功能实体(I-CSCF，Interrogating call session control Function)，服务呼叫会话控制功能实体(S-CSCF，Serving call session control Function)，归属用户服务器(HSS，Home subscriber Server)，以及应用服务器(AS，Application Server)，等进行虚拟化，并将这些虚拟网络功能组件(VNFC，Virtualized Network Function Component)部署在云虚拟机上，从而实现网络功能的虚拟化，虚拟化后的网络功能称为VNF，网络功能虚拟化后带来的好处是：网络功能和硬件解耦，实现了异种网络功能、异厂家的网络功能运行在相同的硬件上，运营商部署新网络功能时，能够方便的利旧，保护了运营商投资；由于网络功能都能运行在通用的硬件上，加快了运营商开展新业务的速度；由于采用通用硬件，大规模采购能够降低成本；由于网络功能具有弹性伸缩功能，网络功能因需占用资源，提高了资源利用率，同时，在负荷较低时，会关闭部分通用服务器，所以更加绿色节能。
欧洲电信标准化协会(ETSI，European Telecommunications Standards Institute)制定了电信NFV协议标准，在ETSI NFV技术规范中定义了端到端参考架构，在端到端参考架构中引入了网络功能虚拟化编排器(NFVO，Network Functions Virtualization Orchestrator，也称Orchestrator)、虚拟网络功能管理(VNFM，Virtualized Network Function Manager)、虚拟基础设施管理(VIM，Virtualized Infrastructure Manager)、网络功能虚拟化基础设施(NFVI，Network Functions Virtualization Infrastructure)，以及虚拟网络功能(VNF， Virtualized Network Function)等功能单元。其中，NFVO主要负责全网的网络服务、虚拟资源以及物理资源的编排和相关管理功能；VNFM主要负责网元VNF的生命周期管理和VNF相关的虚拟资源的分配和管理；VIM主要负责虚拟化基础设施管理，主要功能是实现对整个基础设施层资源包括计算、存储、网络资源的分配和管理；NFVI作为虚拟化资源层，包括虚拟化后新增的虚拟化平台和原有的硬件资源；VNF指虚拟化后的网元，部署在NFVI上，执行着3GPP定义的网元功能，功能与非虚拟化时保持一致。
NFV技术的一个关键功能是虚拟网络功能弹性伸缩技术，弹性伸缩方式又分为横向弹性伸缩(scale out/scale in)和纵向弹性伸缩(scale up/scale down)，其中，scale out指的是横向方式的弹性伸，scale in指的是横向方式的弹性缩；scale up指的是纵向方式的弹性伸，scale down指的是纵向方式的弹性缩。目前，弹性伸缩的触发方式有两种：一种是自动触发，即VNF根据自身的负荷情况动态调整自身的资源占用，这种方式也称为自动弹；另一种是手动触发，即用户根据管理需要手动调整VNF的资源占用。弹性伸缩的调整方式也有两种：一种是横向调整，即通过赠/删虚拟机的方式来调整，称之为scale out/scale in；另一种是纵向调整，即通过调整虚拟机占用资源的方式来调整，称之为scale up/scale down。ETSI规范《NFV-MAN001v039-clean.doc》中附录B.5定义了VNF弹性伸缩流程，其中B.5.3描述了由网络功能虚拟化编辑器(NFVO，Network Functions Virtualization Orchestrator)执行资源分配的弹性伸缩流程，B.5.4描述了由VNF管理器(VNFM，VNF Manager)执行资源分配的弹性伸缩流程，B.5.4.1描述了VNF自动弹的过程，B.5.4.2描述了由网元管理(EM，Element management)发起的VNF自动弹的过程，B.5.4.3描述了VNF自动缩的过程，B.5.4.4描述了由EM发起的VNF自动缩的过程。需要特别说明的是，B5.3和B5.4描述的流程都不区分是scale out/scale in，还是scale up/scale down，也就是说，对于横向弹性伸缩和纵向弹性伸缩均适用。
图1为相关规范中的由NFVO执行资源分配的弹性伸缩流程示意图，如图1所示，包括以下步骤：
步骤100：发送者(Sender)，比如运营支撑***(OSS，Operations Support System)向NFVO发送弹性伸缩请求；
步骤101：NFVO对弹性伸缩请求进行合法性校验，检查此需求是否符合策略配置；
步骤102：NFVO找到当前请求弹性伸缩的VNF对应的VNFM，如果该VNF没有对应的VNFM，NFVO创建新VNFM；
步骤103：NFVO在执行弹性伸缩之前，对弹性伸缩请求进行可行性校验，并预留相关资源。需要说明的是，整个步骤103的实现是可选的，但是如果步骤103被选择执行，则必须全部执行；
可选地，步骤103包括：
步骤103a：NFVO请求VNFM对弹性伸缩请求进行可行性检查；
步骤103b：VNFM执行任何必须的准备工作，这些工作可能包括根据VNF特定限制修改或补充弹性伸缩请求的输入参数，比如检查输入参数是否在合法范围内，如果不在合法范围内，可能会对参数进行修改等；
步骤103c：VNFM向NFVO返回更新后的弹性伸缩请求；
步骤103d：NFVO请求VIM检查资源如计算、存储、网络等资源的可用性并预留这些资源；
步骤103e：VIM检查资源如计算、存储、网络等资源的可用性并预留这些资源；
步骤103f：VIM向NFVO返回资源预留的结果；
步骤104：NFVO向VNFM发送弹性伸缩请求，其中携带有弹性伸缩数据。需要说明的是，如果步骤103已经执行，则还携带有资源预留信息；
步骤105：VNFM执行任何需要的准备工作：如对弹性伸缩请求进行校验，参数校验等，可能还包括根据VNF特定限制修改或补充弹性伸缩请求的入参；
这里，本步骤和步骤103b一致，如果步骤103已经执行，则不再执行本步骤；
步骤106：VNFM向NFVO回复资源变更响应；
步骤107：NFVO向VIM发送资源如计算、存储、网络等资源变更请求；
步骤108：VIM根据需要修改VNF内部网络；
步骤109：VIM根据需要实例化新虚拟机并连到内部网络；
步骤110：VIM向NFVO回复资源变更完成响应；
步骤111：NFVO通知VNFM资源变更完成；
步骤112：VNFM向NFVO回复响应；
步骤113：NFVO向请求回复弹性申请成功响应。
图2为相关规范中的VNF自动弹的流程示意图，如图2所示，包括：
步骤200：VNFM从VNF收集测量数据(应用相关)；
步骤201：VNFM检测到***负荷超过弹性伸阈值，触发弹性伸请求；
步骤202：VNFM请求NFVO对弹性伸请求进行校验；
步骤203：NFVO决定执行弹性伸，根据VNF的容量数据库及VNF模版，计算弹性伸需要的资源如计算、存储、网络等资源，检查当前是否有足够的资源；
步骤204：NFVO授权VNFM对VNF进行弹性伸；
步骤205：VNFM根据NFVO的指示，向VIM发送请求创建并启动虚拟机(VM，Virtual Machine)；
步骤206：VIM实例化并启动新VM，并给VNFM回复成功响应；
步骤207：VNFM配置VNF上相关数据；
步骤208：VNFM向NFVO回复弹性伸成功响应；
步骤209：VNFM通知EM此VNF容量发生变更；
步骤210：NFVO保存VNF和VIM资源池之间的映射关系；
步骤211：EM和VNFM更新VNF相关数据；
步骤212：EM配置VNF相关应用数据。
图3为相关规范中的EM发起的VNF自动弹的流程示意图，如图3所示，包括如下步骤：
步骤300：手工操作员在EM上手动执行弹性伸请求，或者EM根据检 测结果触发自动弹性伸请求。
步骤301：EM向VNFM发送弹性伸请求；
步骤302：VNFM请求NFVO对弹性伸请求进行校验；
步骤303：NFVO决定执行弹性伸，根据VNF的容量数据库及VNF模版，计算弹性伸需要的资源如计算、存储、网络等资源，检查当前是否有足够的资源；
步骤304：NFVO授权VNFM对VNF进行弹性伸；
步骤305：VNFM根据NFVO的指示，向VIM发送请求创建并启动VM；
步骤306：VIM实例化并启动新VM，并给VNFM回复成功响应；
步骤307：VNFM配置VNF上相关数据；
步骤308：VNFM向NFVO回复弹性伸成功响应；
步骤309：VNFM通知EM当前请求弹性伸缩的VNF容量发生变更；
步骤310：NFVO保存VNF和VIM资源池之间的映射关系；
步骤311：EM和VNFM更新VNF相关数据；
步骤312：EM配置VNF相关应用数据。
图4为相关规范中的VNF自动缩的流程示意图，如图4所示，包括：
步骤400：VNFM从VNF收集测量数据(应用相关)；
步骤401：VNFM检测到***负荷低于弹性缩阈值，触发弹性缩请求；
步骤402：VNFM请求NFVO对弹性缩请求进行校验；
步骤403：NFVO根据配置策略决定执行弹性缩；
步骤404：NFVO授权VNFM对VNF进行弹性缩；
步骤405：VNF优雅关闭VNFC，这些VNFCA是对正在进行的业务无影响的；
步骤406：当即将删除的VNFC(VM)上无负荷时，VNFM请求VIM删除VM；
步骤407：VIM删除VM并通知VNFM；
步骤408：VNFM通知NFVO弹性缩成功；
步骤409：VNFM通知EM，VNF容量弹性缩成功；
步骤410：NFVO更新VNF和VIM资源池之间的映射关系；
步骤411：EM和VNFM更新VNF相关数据。
图5为相关规范中的EM发起的VNF自动缩的流程示意图，如图5所示，包括如下步骤：
步骤500：手工操作员在EM上手动执行弹性缩请求，或者EM根据检测结果触发自动弹性缩请求；
步骤501：EM向VNFM发送触发弹性缩请求；
步骤502：VNFM请求NFVO对弹性缩请求进行校验；
步骤503：NFVO根据配置策略决定执行弹性缩；
步骤504：NFVO授权VNFM对VNF进行弹性缩；
步骤505：VNF优雅关闭VNFC，这些VNFCA是对正在进行的业务无影响的；
步骤506：当即将删除的VNFC(VM)上无负荷时，VNFM请求VIM删除VM；
步骤507：VIM删除VM并通知VNFM；
步骤508：VNFM通知NFVO弹性缩成功；
步骤509：VNFM通知EM，VNF容量弹性缩成功；
步骤510：NFVO更新VNF和VIM资源池之间的映射关系；
步骤511：EM和VNFM更新VNF相关数据。
上述相关规范规定的弹性伸缩的通用流程，适用于scale up/scale down和scale out/scale in两种方式的弹性伸缩，但是，当VNF发生scale out/scale in方式的弹性伸缩时，图1所示的流程中并没有描述是否需要优雅弹性伸缩及如何弹性伸缩；图2所示的流程中并没有描述是否需要优雅弹性伸及如何弹性伸；图3所示的流程中并描述是否需要优雅弹性伸及如何弹性伸；图4所示的流程中虽然述了需要优雅弹性缩，但没有描述如何弹性缩；图5所示的 流程中虽然描述了需要优雅弹性伸，但没有描述如何弹性缩。
也就是说，相关技术提供的实现方案中，没有针对优雅scale out和scale in弹性伸缩的详细解决方案。
发明内容
本文提供一种实现VNF弹性伸缩的方法及装置，使得VNF能够优雅scale out和scale in弹性伸缩，从而避免对正在进行的业务的影响。
一种实现虚拟网络功能VNF弹性伸缩的方法，包括：
当新创建的虚拟网络功能组件VNFC对应的虚拟机VM进入服务时，VNF按照预先设置的均衡策略将自身管辖下的其他VNFC上的有状态的业务数据迁移到新创建的VNFC；
当删除待退出服务的VNFC对应的VM前，VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到自身管辖下的其他VNFC。
可选地，所述将其他VNFC上的有状态的业务数据迁移到新创建的VNFC，包括：
当网络功能虚拟化编排器NFVO或虚拟网络功能管理VNFM通过虚拟基础设施管理VIM创建新的VNFC对应的VM之后，VNFM向所述VNF发送迁移有状态业务数据请求；
所述VNF按照预先设置的均衡策略，将自身管辖下的其他VNFC上的有状态的业务数据迁移到新进入服务的VNFC上。
可选地，所述VNF完成迁移后，该方法还包括：所述VNF向VNFM返回迁移请求成功响应。
可选地，所述VNF启动迁移后，该方法还包括：所述VNF向VNFM返回迁移请求成功响应。
可选地，所述将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC包括：
在NFVO或VNFM通过VIM删除待退出服务的VNFC对应的VM之前，VNFM向VNF发送迁移有状态业务数据请求；
VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到该VNF自身管辖下的其他VNFC上。
可选地，所述VNF完成迁移工作后，该方法还包括：所述VNF向VNFM返回迁移请求成功响应。
可选地，该方法还包括：所述NFVO或者VNFM通过虚拟网络功能管理VIM删除所述待退出服务的VNFC对应的VM。
可选地，所述均衡策略用于保证多个VNFC之间的负荷均衡，包括以下一种或多种：多个VNFC之间的网协IP流量均衡、多个VNFC之间的在线会话均衡、多个VNFC之间的注册用户数的均衡。
一种实现VNF弹性伸缩的装置，包括设置模块和处理模块；其中，
所述设置模块，其中设置有均衡策略；
所述处理模块，设置为当新创建的VNFC对应的VM进入服务时，按照预先设置的均衡策略将其他VNFC上的有状态的业务数据迁移到新创建的VNFC；当删除待退出服务的VNFC对应的VM前，按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC。
可选地，所述处理模块是设置为：接收到迁移有状态业务数据请求，按照预先设置的均衡策略将其他VNFC上的有状态的业务数据迁移到新进入服务的VNFC上；
当完成迁移后或当VNF启动迁移后，向VNFM返回迁移请求成功响应。
可选地，所述处理模块是设置为：接收到迁移有状态业务数据请求，按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC上；当VNF完成迁移工作后，向VNFM返回迁移请求成功响应。
可选地，所述装置设置在VNF中；所述其他VNFC为该VNF管辖下的VNFC。
本发明实施例还提供一种计算机可读存储介质，存储有程序指令，当该程序指令被执行时可实现上述方法。
与相关技术相比，本申请技术方案包括当新创建的VNFC对应的VM进入服务时，VNF按照预先设置的均衡策略将其管辖下的其他VNFC上的有状态的业务数据迁移到新创建的VNFC；当删除待退出服务的VNFC对应的VM前，VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到该VNF自身管辖下的其他VNFC。通过本发明实施例提供的技术方案，完善了相关规范中的弹性伸缩实现方案，实现了优雅scale out和scale in弹性伸缩，从而避免了对正在进行的业务的影响，提高了VNF的可用性。
附图概述
图1为相关规范中的由NFVO执行资源分配的弹性伸缩流程示意图；
图2为相关规范中的VNF自动弹的流程示意图；
图3为相关规范中的EM发起的VNF自动弹的流程示意图；
图4为相关规范中的VNF自动缩的流程示意图；
图5为相关规范中的EM发起的VNF自动缩的流程示意图；
图6为本发明实施例实现VNF弹性伸缩的方法的流程图；
图7为本发明实施例实现VNF弹性伸缩的装置的组成结构示意图。
本发明的实施方式
需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。
在实际的应用中，需要VNF能够优雅scale out和scale in弹性伸缩，以免对正在进行的业务进行影响，比如：当电信领域的P-CSCF、I-CSCF、S-CSCF发生弹性伸缩时，电信运营商一般要求不能对正在进行的呼叫、注册等产生影响，对于scale out方式的弹性伸而言，在VNF弹性伸的过程中，当新的 VNFC进入服务后，需要将其他VNFC上的话务逐步迁移到新的VNFC上，以确保VNF内的多个VNFC上的负荷均衡，如果不均衡，可能会导致高负荷的VNFC上业务产生损失；对于scale in方式的弹性缩而言，当决策任一个VNFC退出服务时，由于VNFC上可能有用户注册数据或者在线会话等有状态的业务信息，如果强制通过VIM删除VNFC对应的虚拟机，会导致注册数据或者在线会话等有状态的业务数据丢失，影响用户感受，降低***的可用性。
图6为本发明实施例实现VNF弹性伸缩的方法的流程图，如图6所示，包括：
步骤600：当新创建的VNFC对应的VM进入服务时，VNF按照预先设置的均衡策略将自身管辖下的其他VNFC上的有状态的业务数据迁移到新创建的VNFC；
本步骤中的将其他VNFC上的有状态的业务数据迁移到新创建的VNFC包括：VNF将其他负荷比较高的VNFC上的话务如注册数据、在线会话数据等逐步迁移到新增的VNFC上，以达到负荷均衡的目的，以免其他VNFC上的负荷过高而产生业务损失。
可选地，本步骤包括：
当NFVO或者VNFM通过VIM创建了新VNFC对应的VM之后，也就是说，在scale out的过程中，VNFM向VNF发送迁移有状态业务数据请求，请求VNF按照预先设置的均衡策略，将自身管辖下的其他VNFC上的有状态的业务数据均衡的迁移到新进入服务的VNFC上。此时，在迁移有状态业务数据请求中携带有用于表示当前为弹性伸过程的标识。
当VNF完成迁移工作后，VNF向VNFM返回迁移请求成功响应；
或者，对于数据量比较大，为了达到平稳迁移的效果，迁移时间比较长，而VNF可以选择及时给VNFM回复，即当VNF启动迁移工作后，VNF就向VNFM返回迁移请求成功响应。
本步骤中的均衡策略用于保证多个VNFC之间的负荷均衡，包括但不限于以下中的一种或多种：多个VNFC之间的网络之间互联的协议(IP，简称 为网协)流量均衡、多个VNFC之间的在线会话均衡、多个VNFC之间的注册用户数的均衡等。
步骤601：当删除待退出服务的VNFC对应的VM前，VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到自身管辖下的其他VNFC；
本步骤中的将退出服务的VNFC上的有状态的业务数据迁移到其他VNFC包括：将待退出服务的VNFC上的话务如注册数据、在线会话数据等均衡的迁移到该VNF自身管辖下的其他VNFC上，当迁移完毕后，再由VIM删除待退出服务VNFC对应的VM，确保了VNFC的优雅退出服务。
可选地，本步骤包括：
在NFVO或者VNFM即将通过VIM删除待退出服务的VNFC对应的VM之前，即在scale in的过程中，VNFM向VNF发送迁移有状态业务数据请求，请求VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据均衡的迁移到该VNF自身管辖下的其他VNFC上。此时，在迁移有状态业务数据请求中携带有用于表示当前为弹性缩过程的标识。
当VNF完成迁移工作后，VNF向VNFM返回迁移请求成功响应，NFVO或者VNFM即可通过VIM删除待退出服务的VNFC对应的VM。
需要说明的是，步骤600与步骤601没有严格的先后顺序。
本实施例方法中，如果注册用户数据迁移失败，退出本流程。
通过本实施例图6所示的技术方案，完善了相关规范中的弹性伸缩实现方案，实现了优雅scale out和scale in弹性伸缩，从而避免了对正在进行的业务的影响，提高了VNF的可用性。
下面结合一些应用示例进行说明。
应用示例1
针对图1所示的相关规范中的由NFVO执行资源分配的弹性伸缩的流程，本示例对其中的scale out弹性伸进行了改进，在步骤110即VIM向NFVO回复资源变更完成响应，与步骤111即NFVO通知VNFM资源变更完成之间， 还包括：
VNFM向VNF发送迁移有状态业务数据请求，请求VNF将其他VNFC上的有状态的业务数据均衡的迁移到新进入服务的VNFC；
VNF按照预先设置的均衡策略，将其他VNFC上的有状态业务数据如会话或者注册用户数据等均衡迁移到新创建的VNFC上；
VNF向VNFM返回迁移请求成功响应。这里，VNF可以在迁移完毕后返回迁移请求成功响应。当然，为了避免迁移过程时间过长，VNF也可以在启动迁移工作后，立刻向VNFM返回迁移请求成功响应，因为待迁移数据可能很多。
应用示例2
针对图1所示的相关规范中的由NFVO执行资源分配的弹性伸缩的流程，本示例对其中的scale in弹性缩进行了改进，首先，因为是scale in流程，需要删除虚拟机，所以不需要执行资源预留过程，因此图1中的将步骤103d～步骤103f删除，同时，在步骤104即NFVO向VNFM发送弹性伸缩请求，与步骤105即VNFM执行任何需要的准备工作之间，还包括：
VNFM向VNF发送迁移有状态业务数据请求，请求VNF将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；
VNF按照预先设置的均衡策略，将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；
VNF向VNFM返回迁移请求成功响应。
应用示例3
针对图2所示的相关规范中的VNF自动弹的流程，本示例对其中的scale out弹性伸进行了改进，将图2中步骤207替换为：
VNFM配置VNF上相关数据，VNFM向VNF发送迁移有状态业务数据请求，请求VNF将其他VNFC上的有状态的业务数据均衡的迁移到新进入服 务的VNFC；
VNF按照预先设置的均衡策略，将其他VNFC上的有状态业务数据如会话或者注册用户数据等均衡迁移到新创建的VNFC上；
VNF向VNFM返回迁移请求成功响应，这里，VNF可以在迁移完毕后返回迁移请求成功响应。当然，为了避免迁移过程时间过长，VNF也可以在启动迁移工作后，立刻向VNFM返回迁移请求成功响应，因为待迁移数据可能很多。
应用示例4
针对图3所示的相关规范中的EM发起的VNF自动弹的流程，本示例对其中的scale out弹性伸进行了改进，将图3中步骤307修改为：
VNFM配置VNF上相关数据，VNFM向VNF发送迁移有状态业务数据请求，请求VNF将其他VNFC上的有状态的业务数据均衡的迁移到新进入服务的VNFC；
VNF按照预先设置的均衡策略，将其他VNFC上的有状态业务数据如会话或者注册用户数据等均衡迁移到新的VNFC上；
VNF向VNFM返回迁移请求成功响应。这里，VNF可以在迁移完毕后返回迁移请求成功响应。当然，为了避免迁移过程时间过长，VNF也可以在启动迁移工作后，立刻向VNFM返回迁移请求成功响应，因为待迁移数据可能很多。
应用示例5
针对图4所示的相关规范中的VNF自动缩的流程，本示例对其中的scale in弹性缩进行了改进，在图4所示的步骤404即NFVO授权VNFM对VNF进行弹性缩，与步骤405即VNF优雅关闭VNFC之间，还包括：
VNFM向VNF发送迁移有状态业务数据请求，请求VNF将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；
VNF按照预先设置的均衡策略，将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；
当迁移数据完毕后，VNF向VNFM返回迁移请求成功响应。
应用示例6
针对图5所示的相关规范中的EM发起的VNF自动缩的流程，本示例对其中的scale in弹性缩进行了改进，在图5所示的步骤504即NFVO授权VNFM对VNF进行弹性缩，与步骤505即VNF优雅关闭VNFC之间，还包括：
VNFM向VNF发送迁移有状态业务数据请求，请求VNF将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；
VNF按照预先设置的均衡策略，将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；
当迁移数据完毕后，VNF向VNFM返回迁移请求成功响应。
图7为本发明实施例实现VNF弹性伸缩的装置的组成结构示意图，如图7所示，至少包括设置模块和处理模块，其中：
所述设置模块，其中设置有均衡策略；
所述处理模块，设置为当新创建的VNFC对应的VM进入服务时，按照预先设置的均衡策略将其他VNFC上的有状态的业务数据迁移到新创建的VNFC；当删除待退出服务的VNFC对应的VM前，按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC。
其中，处理模块是设置为：
接收到迁移有状态业务数据请求，按照预先设置的均衡策略将其他VNFC上的有状态的业务数据均衡的迁移到新进入服务的VNFC上；
当完成迁移工作后或当VNF启动迁移工作后，向VNFM返回迁移请求成功响应。
或者，
处理模块是设置为：接收到迁移有状态业务数据请求，按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据均衡的迁移到其他VNFC上；当VNF完成迁移工作后，向VNFM返回迁移请求成功响应。
上述装置，可以设置在VNF中。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序来指令相关硬件完成，所述程序可以存储于计算机可读存储介质中，如只读存储器、磁盘或光盘等。可选地，上述实施例的全部或部分步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。
工业实用性
通过本发明实施例提供的技术方案，完善了相关规范中的弹性伸缩实现方案，实现了优雅scale out和scale in弹性伸缩，从而避免了对正在进行的业务的影响，提高了VNF的可用性。

Claims (13)

  1. 一种实现虚拟网络功能VNF弹性伸缩的方法，包括：
    当新创建的虚拟网络功能组件VNFC对应的虚拟机VM进入服务时，VNF按照预先设置的均衡策略将自身管辖下的其他VNFC上的有状态的业务数据迁移到新创建的VNFC；
    当删除待退出服务的VNFC对应的VM前，VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到自身管辖下的其他VNFC。
  2. 根据权利要求1所述的方法，其中，所述将其他VNFC上的有状态的业务数据迁移到新创建的VNFC，包括：
    当网络功能虚拟化编排器NFVO或虚拟网络功能管理VNFM通过虚拟基础设施管理VIM创建新的VNFC对应的VM之后，VNFM向所述VNF发送迁移有状态业务数据请求；
    所述VNF按照预先设置的均衡策略，将自身管辖下的其他VNFC上的有状态的业务数据迁移到新进入服务的VNFC上。
  3. 根据权利要求2所述的方法，所述VNF完成迁移后，该方法还包括：所述VNF向VNFM返回迁移请求成功响应。
  4. 根据权利要求2所述的方法，所述VNF启动迁移后，该方法还包括：所述VNF向VNFM返回迁移请求成功响应。
  5. 根据权利要求1所述的方法，其中，所述将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC，包括：
    在NFVO或VNFM通过VIM删除待退出服务的VNFC对应的VM之前，VNFM向VNF发送迁移有状态业务数据请求；
    VNF按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到该VNF自身管辖下的其他VNFC上。
  6. 根据权利要求5所述的方法，所述VNF完成迁移后，该方法还包括：所述VNF向VNFM返回迁移请求成功响应。
  7. 根据权利要求6所述的方法，该方法还包括：所述NFVO或者VNFM通过虚拟网络功能管理VIM删除所述待退出服务的VNFC对应的VM。
  8. 根据权利要求1～7任一项所述的方法，其中，所述均衡策略用于保证多个VNFC之间的负荷均衡，包括以下一种或多种：多个VNFC之间的网协IP流量均衡、多个VNFC之间的在线会话均衡、多个VNFC之间的注册用户数的均衡。
  9. 一种实现VNF弹性伸缩的装置，包括设置模块和处理模块；其中，
    所述设置模块，其中设置有均衡策略；
    所述处理模块，设置为当新创建的VNFC对应的VM进入服务时，按照预先设置的均衡策略将其他VNFC上的有状态的业务数据迁移到新创建的VNFC；当删除待退出服务的VNFC对应的VM前，按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC。
  10. 根据权利要求9所述的装置，其中，所述处理模块是设置为：接收到迁移有状态业务数据请求，按照预先设置的均衡策略将其他VNFC上的有状态的业务数据迁移到新进入服务的VNFC上；当完成迁移后或当VNF启动迁移后，向VNFM返回迁移请求成功响应。
  11. 根据权利要求9所述的装置，其中，所述处理模块是设置为：接收到迁移有状态业务数据请求，按照预先设置的均衡策略将待退出服务的VNFC上的有状态的业务数据迁移到其他VNFC上；当VNF完成迁移后，向VNFM返回迁移请求成功响应。
  12. 根据权利要求9～11任一项所述的装置，其中，所述装置设置在VNF中；所述其他VNFC为该VNF管辖下的VNFC。
  13. 一种计算机可读存储介质，存储有程序指令，当该程序指令被执行时可实现权利要求1-8任一项所述的方法。
PCT/CN2015/075714 2014-11-04 2015-04-01 一种实现vnf弹性伸缩的方法及装置 WO2016070559A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410614514.6A CN105634780B (zh) 2014-11-04 2014-11-04 一种实现vnf弹性伸缩的方法及装置
CN201410614514.6 2014-11-04

Publications (1)

Publication Number Publication Date
WO2016070559A1 true WO2016070559A1 (zh) 2016-05-12

Family

ID=55908473

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/075714 WO2016070559A1 (zh) 2014-11-04 2015-04-01 一种实现vnf弹性伸缩的方法及装置

Country Status (2)

Country Link
CN (1) CN105634780B (zh)
WO (1) WO2016070559A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10764132B2 (en) * 2015-09-30 2020-09-01 Huawei Technologies Co., Ltd. Scale-out association method and apparatus, and system
US11640313B2 (en) * 2017-11-07 2023-05-02 Huawei Technologies Co., Ltd. Device upgrade method and apparatus

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105975330B (zh) 2016-06-27 2019-06-18 华为技术有限公司 一种网络边缘计算的虚拟网络功能部署方法、装置和***
US20170373938A1 (en) * 2016-06-27 2017-12-28 Alcatel-Lucent Usa Inc. Predictive auto-scaling of virtualized network functions for a network
CN109417488B (zh) * 2016-06-28 2020-10-23 华为技术有限公司 虚拟网络功能资源管理的方法和设备
WO2018018459A1 (zh) * 2016-07-27 2018-02-01 华为技术有限公司 一种策略处理的方法和设备
CN107770067B (zh) * 2016-08-23 2021-05-11 中兴通讯股份有限公司 消息发送方法和装置
CN108023749A (zh) * 2016-11-03 2018-05-11 华为技术有限公司 管理网络切片的方法、设备和***
CN108123822B (zh) * 2016-11-30 2020-02-14 华为技术有限公司 一种链路处理方法以及链路处理设备
CN108132827B (zh) * 2016-11-30 2021-12-10 华为技术有限公司 一种网络切片资源映射方法、相关设备及***
CN108762928B (zh) * 2018-05-29 2022-03-22 郑州云海信息技术有限公司 一种弹性伸缩服务执行操作时的处理方法和***
WO2020076140A1 (en) 2018-10-12 2020-04-16 Samsung Electronics Co., Ltd. System and method for call selection and migration in ng-cu over ngran
CN111935244B (zh) * 2020-07-20 2022-11-29 江苏安超云软件有限公司 一种业务请求处理***及超融合一体机

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103257683A (zh) * 2013-05-07 2013-08-21 华为技术有限公司 一种云计算业务伸缩的方法和装置
US8631458B1 (en) * 2011-09-29 2014-01-14 Symantec Corporation Method and apparatus for elastic (re)allocation of enterprise workloads on clouds while minimizing compliance costs
CN103810020A (zh) * 2014-02-14 2014-05-21 华为技术有限公司 虚拟机弹性伸缩方法及装置
CN103984602A (zh) * 2014-05-20 2014-08-13 华为技术有限公司 一种vm资源调度方法、装置及***

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9098200B2 (en) * 2012-02-10 2015-08-04 Hitachi, Ltd. Storage system with virtual volume having data arranged astride storage devices, and volume management method
CN103455512A (zh) * 2012-05-31 2013-12-18 上海博腾信息科技有限公司 Saas平台多租户数据管理模型

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8631458B1 (en) * 2011-09-29 2014-01-14 Symantec Corporation Method and apparatus for elastic (re)allocation of enterprise workloads on clouds while minimizing compliance costs
CN103257683A (zh) * 2013-05-07 2013-08-21 华为技术有限公司 一种云计算业务伸缩的方法和装置
CN103810020A (zh) * 2014-02-14 2014-05-21 华为技术有限公司 虚拟机弹性伸缩方法及装置
CN103984602A (zh) * 2014-05-20 2014-08-13 华为技术有限公司 一种vm资源调度方法、装置及***

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10764132B2 (en) * 2015-09-30 2020-09-01 Huawei Technologies Co., Ltd. Scale-out association method and apparatus, and system
US11640313B2 (en) * 2017-11-07 2023-05-02 Huawei Technologies Co., Ltd. Device upgrade method and apparatus

Also Published As

Publication number Publication date
CN105634780A (zh) 2016-06-01
CN105634780B (zh) 2020-04-07

Similar Documents

Publication Publication Date Title
WO2016070559A1 (zh) 一种实现vnf弹性伸缩的方法及装置
EP3804282B1 (en) Native blockchain platform for improving workload mobility in telecommunication networks
US10057127B2 (en) Processing method for service allocation and related apparatus
RU2683629C2 (ru) Способ и устройство изменения ресурса виртуальной вычислительной машины и устройство для функционирования виртуальной сети передачи данных
US10019281B2 (en) Handoff of virtual machines based on security requirements
US10701139B2 (en) Life cycle management method and apparatus
EP3457733B1 (en) Mobile edge orchestrator and application migration system
US20190149408A1 (en) Method and apparatus for deploying service in virtualized network
CN105429776B (zh) 一种虚拟化网络功能管理的方法和***
KR20180112825A (ko) Mec 애플리케이션의 재배치를 제어하기 위한 방법 및 서버
CN105634778B (zh) 实现vnf实例化的方法、***及nfvo和vnfm
WO2019029327A1 (zh) 网络功能服务的发现方法及设备
EP3823389A1 (en) User plane resource management method, user plane network element, and control plane network element
WO2015131671A1 (zh) 目标资源占用情况的处理方法及装置
WO2016155291A1 (zh) 一种虚拟化网络功能伸缩的管理方法及装置
CN106209402A (zh) 一种虚拟网络功能的伸缩方法和设备
RU2690198C1 (ru) Способ обработки ресурсов виртуализации сетевых функций и диспетчер виртуализованной сетевой функции
WO2018153354A1 (zh) 一种资源申请、vnf实例创建方法及装置
CN111726241B (zh) 网络资源管理方法、***、网络设备和可读存储介质
CN106230623A (zh) 一种vim站点选择方法及装置
CN106576068B (zh) 一种软件安全验证方法、设备及***
WO2017011938A1 (zh) 虚拟网络功能扩容的方法和装置
CN108768744B (zh) 一种对云平台创建网络的管理方法和装置
JP6591045B2 (ja) ネットワークサービスを移行するための方法及びネットワークサービス装置
EP3280091A1 (en) Method, apparatus and system for establishing interface between vnfms

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: 15856944

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: 15856944

Country of ref document: EP

Kind code of ref document: A1