CN108243205B

CN108243205B - Method, equipment and system for controlling resource allocation of cloud platform

Info

Publication number: CN108243205B
Application number: CN201611209229.1A
Authority: CN
Inventors: 黄小群; 王煜坚; 左轶
Original assignee: Nokia Shanghai Bell Co Ltd
Current assignee: Nokia Shanghai Bell Co Ltd
Priority date: 2016-12-23
Filing date: 2016-12-23
Publication date: 2021-06-08
Anticipated expiration: 2036-12-23
Also published as: CN108243205A

Abstract

The invention aims to provide a method, equipment and a system for controlling resource allocation of a cloud platform. The invention adds a timing protection function, associates virtual network functions with virtual resources in advance in the virtual network function manager, sends event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation in real time, and monitors each scheduling operation by the network function virtual orchestrator so as to determine whether the scheduling operation is normally performed. Compared with the prior art, the method improves the high availability and reliability during the resource allocation of the cloud platform, and can automatically detect and evaluate isolated resources; meanwhile, the performance of each element can be enhanced to schedule resource allocation and feed back various problems during resource allocation in real time, so that the purposes of ensuring stability and simplifying operation are achieved, the visibility of resource allocation operation is favorably improved, and the implementation of network function virtualization is ensured and accelerated.

Description

Method, equipment and system for controlling resource allocation of cloud platform

Technical Field

The invention relates to the field of cloud computing, in particular to a technology for controlling resource allocation of a cloud platform.

Background

The cloud fusion network can achieve intellectualization, effectiveness, automatic implementation, dynamic adjustment and sensitivity connection optimization. Nfv (network Function virtualization), namely, network Function virtualization, can implement a cloud-end fusion network on a cloud platform level.

NFV introduces and defines MANO (Network Functions Virtualization Management and organization) architecture, including NFVO (NFV Resource organization ), VNFM (Virtualized Network Function Lifecycle Management), Infrastructure (NFVI + VIM: NFV Infrastructure, NFV Infrastructure; Virtualized Infrastructure Manager).

NFVI sharing capabilities after software hardware decoupling are included in MANO management and orchestration. That is, each VIM manages a virtual resource pool of the NFVI, and implements, together with the NFVO and the VNFM, a Virtual Network Function (VNF) lifecycle and resource management Function.

During VNF resource allocation and termination, VNFM is used for task creation and resource allocation. The NFVO is only responsible for resource authorization and the full NFVI resource pool topology view, which includes VNF and VIM resource pool mappings. The time consumption of resource allocation/termination depends on VNF type and VNFC (virtualized network function module Component) number, configuration specification (loader) of the associated VM, Image (Image) size and VM number, involving steps in the flow, spanning 3 standard NFV MANO elements.

The advent of this MANO architecture has changed the resource allocation model for VNF (virtual Network Function) deployment and VNF lifecycle management. The architecture enables the topology view optimization, the automatic arrangement of Virtual network functions, the strategic automatic operation and the like of basic architecture resources under a virtualized network environment by introducing a virtualization technology, an automation technology and an intelligence technology, but a method for ensuring the high reliability of the resource arrangement among systems does not bring enough attention, which is restrictively embodied in the Virtual resource management during the VNF allocation period, the processing process (including the instantiation, elastic expansion and automatic recovery and termination of VNF instances) in the VNF lifecycle, the overload of components, the abnormal state of a transceiver unit and the like, which cause the abnormal resource allocation process, cause the failure of NFVO to associate the VNF with the allocated Virtual resources, the failure of sensing that the resources allocated by the VIM (or the failure of complete resource deletion) are in the unmanageable state, and cause the phenomenon of orphan Virtual Machine resources (orphaned VM, orphaned Virtual Machine), this results in that although there are enough resources in the NFVI, these resources are not available for VNF deployment in the topological view of the NFVO. In an extreme case, the continued presence of an orphaned virtual machine in the NFVI will result in the exhaustion of the NFVI's resource pool.

Fig. 1 shows an example of a basic call scenario for VNF resource allocation and release in the prior art, and from the following analysis of this example, it can be seen that since NFVO, VNFM and VIM do not provide additional protection, failover (failover) can only be recovered through HA (High Availability) architecture, and there is a limit to MANO in core NFVI resource allocation and termination function processing.

In step S10 of fig. 1, a VNF operation is triggered between NFVO and NVFM; the VNFM parses a virtual resource VNFD (Virtualized Network Function Descriptor, Virtualized Network Function module Descriptor); in step S11, the VNFM sends a lifecycle authorization request to the NFVO; and according to the response information of the NFVO to the lifecycle authorization request, in step S12, a resource allocation/release command is sent to the VIM, thereby requiring resource allocation; when the resource to be allocated is determined, in step S13, a request and a response for resource change are made between the VIM and the NFVO; in step S14 and step S15, the VNFM performs initial configuration on the allocated resources; in step S16, since the resource has been created, a service is activated to the EMS by the VNFM; in step S17, the VNFM notifies the NFVO, the NVF lifecycle changes, and the NFVO obtains the information that the VM is instantiated, and maps the VNF to the VIM and the resource pool.

As shown in fig. 1, the VNFM is a core element in the VNF resource allocation. Any abnormal situation can lead to message sending errors. The failover recovery procedure through the VNFM HA system is a limitation on the request message protection.

Without intermediate task state information, the NVFO cannot know the information of the resource allocation process unless it is able to receive VNF lifecycle change messages. If the NFVO periodically requests the task state for a long time, it will degrade the system performance, which will result in poor visibility and operational experience of the control process.

In summary, the MANO resource allocation flow is designed to be defective in high reliability such that isolated VMs are generated:

the time of resource allocation varies based on VNF type at task implementation. Such as data corruption and system overload control, will cause an anomaly in the VNFM. While the HA is only used for failover, stability is difficult to guarantee.

VIM allocates NFVI resources for VMs. If NFVO is unable to map VNF to VIM resource pool and update VFVI resource topology, an isolated VM will be generated. This results in a loss of cloud pool resources and a reduction in system stability. The reason for the failure to map successfully is as follows:

1: and a resource allocation stage: the VM was successfully created in VIM, but there may be some problems when the creation of the VM is reported to the NFVO, such as:

the VIM cannot send a message to the VNFM or lose a portion of the resource information due to sending multiple resource responses in a task by only a single request or response;

the VNFM cannot fully accept the already allocated resources;

the VNFM system cannot send a life cycle message to the NFVO;

the NFVO cannot receive VNF lifecycle messages, although it cannot receive notification resource change messages from the VIM, it can update the NFVI resource state, but cannot map to VNF;

NFVO error exception, etc.

2: and a resource termination stage: VM resources are successfully created, then the NFVO issues a request to terminate one or more VIMs in the VNF, the NFVO discovery operation is terminated, the relationship between VNF and NFVI resources (VNF, virtual machine and hardware) is released, but the termination operation may fail for the following reasons:

NFVO waits long without receiving a reply;

when forwarding from VNFM to VIM, termination operation messages are lost;

the VIM makes an error in VM resource release, in which case the relevant VM will remain non-terminated and the VM will remain in the NFVI.

In summary, how to improve the stability and reliability of the MANO, prevent isolated VMs, and ensure stable and convenient operation and maintenance becomes a problem to be solved by those skilled in the art.

Disclosure of Invention

The invention aims to provide a method, equipment and a system for controlling resource allocation of a cloud platform.

According to an aspect of the present invention, there is provided a method in a virtual network function manager for controlling cloud platform resource allocation, wherein the method comprises the steps of:

x, according to the description information of the virtual network function, estimating the maximum time spent by each scheduling operation when the virtual network function resource allocation operation is executed;

y sending the estimated maximum time to a network function virtual orchestrator corresponding to the virtual network function manager;

wherein, the method also comprises:

a associating a virtual network function with a virtual resource after the virtual resource is allocated;

b, according to each operation of the virtual network function manager in virtual resource allocation, sending an event notification to the network function virtual orchestrator, wherein the event notification comprises an operation action of the operation and a resource corresponding to the operation.

Optionally, the virtualized network function description information includes at least any one of:

-a virtualized network function descriptor;

-a mirror image and a size of the mirror image;

-network load information.

Optionally, the method further comprises:

-retrieving and executing resource handling instructions transmitted by the network function virtualizer, wherein the resource handling instructions are determined based on the event notification.

According to another aspect of the present invention, there is also provided a method in a network function virtual orchestrator for controlling cloud platform resource allocation, wherein the method comprises the steps of:

x, acquiring the estimated maximum time spent by each scheduling operation by a virtual network function manager, wherein the virtual network function manager corresponds to the network function virtual orchestrator;

y, acquiring one or more event notifications sent by the virtual network function manager, wherein the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications comprise operation actions of the operation and resources corresponding to the operation;

and Z, monitoring each scheduling operation by combining the event notification according to the estimated maximum time so as to determine whether the scheduling operation is normally performed.

Optionally, the method further comprises:

-adjusting the estimated maximum time spent for each scheduling operation in dependence on the monitored monitoring result.

Optionally, the method further comprises:

-if the scheduling operation is not normally performed, determining an operation state of the scheduling operation, wherein the operation state includes a resource state corresponding to the scheduling operation;

-determining, from the operating state, a resource handling instruction corresponding to the scheduling operation;

-sending the resource handling instructions to the virtual network function manager.

Optionally, the resource processing instructions comprise at least any one of:

-declaring the resources corresponding to the scheduling operation as orphaned resources and reallocating resources for the scheduling operation;

-stopping the scheduling operation and releasing all resources.

According to another aspect of the present invention, there is also provided a virtual network function manager for controlling resource allocation of a cloud platform, wherein the virtual network function manager includes:

the pre-estimation device is used for pre-estimating the maximum time spent by each scheduling operation when the virtual network function resource allocation operation is executed according to the virtual network function description information;

the estimated sending device is used for sending the estimated maximum time to a network function virtual orchestrator corresponding to the virtual network function manager;

wherein the virtual network function manager further comprises:

associating means for associating a virtual network function with a virtual resource after the virtual resource is allocated;

and a notification sending device, configured to send an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation.

-a virtualized network function descriptor;

-a mirror image and a size of the mirror image;

-network load information.

Optionally, the virtual network function manager further includes:

and the instruction acquisition device is used for acquiring and executing the resource processing instruction sent by the network function virtual orchestrator, wherein the resource processing instruction is determined based on the event notification.

According to another aspect of the present invention, there is also provided a network function virtual orchestrator for controlling cloud platform resource allocation, wherein the network function virtual orchestrator comprises:

the estimation acquisition device is used for acquiring the estimated maximum time spent by each scheduling operation by a virtual network function manager, wherein the virtual network function manager corresponds to the network function virtual orchestrator;

a notification obtaining device, configured to obtain one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation;

and the monitoring device is used for monitoring each scheduling operation according to the estimated maximum time and by combining the event notification so as to determine whether the scheduling operation is normally carried out.

Optionally, the network function virtual orchestrator further comprises:

and the adjusting device is used for adjusting the estimated maximum time spent by each scheduling operation according to the monitored monitoring result.

Optionally, the network function virtual orchestrator further comprises:

a state determining device, configured to determine an operation state of the scheduling operation if the scheduling operation is not performed normally, where the operation state includes a resource state corresponding to the scheduling operation;

the instruction determining device is used for determining a resource processing instruction corresponding to the scheduling operation according to the operation state;

and the instruction sending device is used for sending the resource processing instruction to the virtual network function manager.

Optionally, the resource processing instructions comprise at least any one of:

-stopping the scheduling operation and releasing all resources.

According to another aspect of the present invention, there is also provided a system for controlling resource allocation of a cloud platform, including the virtual network function manager as described above, and the network function virtual orchestrator as described above.

Compared with the prior art, the method adds a timing protection function in a virtual network function manager and a network function virtual orchestrator, associates a virtual network function with a virtual resource in the virtual network function manager in advance, sends an event notification to the network function virtual orchestrator in real time according to each operation of the virtual network function manager during virtual resource allocation, and monitors each scheduling operation according to the estimated maximum time and the event notification to determine whether the scheduling operation is normally performed. Therefore, the resource allocation scene is optimized, the high availability and the reliability during the resource allocation of the cloud platform are improved, and the isolated resources can be automatically detected and evaluated; meanwhile, the performance of each element can be enhanced to schedule resource allocation, various problems in resource allocation can be fed back immediately, the operation experience in resource allocation is improved, the implementation of network function virtualization is guaranteed and accelerated, the purposes of guaranteeing stability and simplifying operation are achieved, and the visibility of resource allocation operation is favorably improved; compared with the prior art, the method specifically comprises the following advantages:

1) the method is simple: the additional timing protection function is simple and convenient to realize and is easy to implement in a resource allocation scene;

2) the method has the following advantages: the adjustment of the resource allocation operation flow does not increase the load of each element, but reduces the burden of each element;

3) controllable: the invention can send event notice to the network function virtual orchestrator in real time, thereby facilitating the operation and maintenance of resource allocation.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:

figure 1 shows a schematic diagram of a basic call scenario for VNF resource allocation and release in the prior art;

FIG. 2 illustrates an apparatus diagram of a virtual network function manager and network function virtual orchestrator for controlling cloud platform resource allocation, according to an aspect of the invention;

FIG. 3 is a schematic diagram of a virtual network function manager and network function virtual orchestrator for controlling cloud platform resource allocation according to another preferred embodiment of the present invention;

FIG. 4 illustrates a flow diagram of a method implemented by a virtual network function manager in cooperation with a network function virtualization orchestrator for controlling cloud platform resource allocation, according to another aspect of the present invention;

FIG. 5 illustrates a flowchart of a method implemented by a virtual network function manager in cooperation with a network function virtualization orchestrator for controlling cloud platform resource allocation, according to a preferred embodiment of the present invention;

figure 6 shows a flow chart of a method for VNF resource allocation according to another preferred embodiment of the present invention;

fig. 7 shows a flowchart of a method for VNF resource allocation and release according to another preferred embodiment of the present invention.

The same or similar reference numbers in the drawings identify the same or similar elements.

Detailed Description

Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel, concurrently, or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.

The "virtual network function manager" or "network function virtual orchestrator" in this context is a "computer device," also referred to as a "computer," and refers to an intelligent electronic device that can execute predetermined processing procedures, such as numerical calculation and/or logic calculation, by running a predetermined program or instruction, and may include a processor and a memory, where the processor executes a pre-stored instruction stored in the memory to execute the predetermined processing procedure, or the hardware, such as ASIC, FPGA, DSP, executes the predetermined processing procedure, or a combination thereof.

The computer device comprises a network device. The network device includes, but is not limited to, a single network server, a server group consisting of a plurality of network servers, or a Cloud Computing (Cloud Computing) based Cloud consisting of a large number of computers or network servers, wherein Cloud Computing is one of distributed Computing, a super virtual computer consisting of a collection of loosely coupled computers. Wherein the computer device can be operated alone to implement the invention, or can be accessed to a network and implement the invention through interoperation with other computer devices in the network. The network in which the computer device is located includes, but is not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a VPN network, and the like.

It should be noted that the network devices and networks are only examples, and other existing or future computer devices or networks may be suitable for the present invention, and are included within the scope of the present invention.

Here, those skilled in the art should understand that the present invention can be applied to resource allocation and management of various cloud platforms, such as resource allocation in telecommunication NFV environments of various operators, resource allocation in what private cloud environment is in a public cloud environment, and the like.

Specific structural and functional details disclosed herein are merely representative and are provided for purposes of describing example embodiments of the present invention. The present invention may, however, be embodied in many alternate forms and should not be construed as limited to only the embodiments set forth herein.

It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element may be termed a second element, and, similarly, a second element may be termed a first element, without departing from the scope of example embodiments. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used herein, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It should also be noted that, in some alternative implementations, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed substantially concurrently, or the figures may sometimes be executed in the reverse order, depending upon the functionality/acts involved.

The present invention is described in further detail below with reference to the attached drawing figures.

FIG. 2 illustrates an apparatus diagram of a virtual network function manager 1 and a network function virtual orchestrator 2 for controlling cloud platform resource allocation, according to an aspect of the present invention; the virtual network function manager 1 includes a forecast device 11, a forecast sending device 12, a correlation device 13 and a notification sending device 14, and the network function virtual orchestrator 2 includes a forecast obtaining device 21, a notification obtaining device 22 and a monitoring device 23.

Specifically, the estimation device 11 of the virtual network function manager 1 estimates the maximum time spent by each scheduling operation when executing the virtual network function resource allocation operation according to the virtualized network function description information; the estimated sending device 12 of the virtual network function manager 1 sends the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager; accordingly, the estimation obtaining means 21 of the network function virtual orchestrator 2 obtains the maximum time that each scheduling operation takes estimated by a virtual network function manager, wherein the virtual network function manager corresponds to the network function virtual orchestrator; the association means 13 of the virtual network function manager 1 associates a virtual network function with a virtual resource after the virtual resource has been allocated; the notification sending device 14 of the virtual network function manager 1 sends an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation; accordingly, the notification obtaining device 22 of the network function virtual orchestrator 2 obtains one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation; the monitoring device 23 of the network function virtual orchestrator 2 monitors each scheduling operation according to the estimated maximum time in combination with the event notification to determine whether the scheduling operation is performed normally.

The estimation device 11 of the virtual network function manager 1 estimates the maximum time spent by each scheduling operation when performing the virtual network function resource allocation operation according to the virtualized network function description information.

Specifically, the pre-estimation device 11 may determine, according to the system configuration, each scheduling operation required when performing the virtual network function configuration; then, determining the virtualized network function description information based on historical statistical data of the whole system or based on data such as the size of data packets, the number of data packets, system capacity, operation flow, resource conditions and the like in the system, or interacting with other devices capable of providing the virtualized network function description information to directly obtain the virtualized network function description information; the virtualized network function description information is then parsed to determine the maximum time spent for each scheduling operation. Here, the maximum time is an estimated value.

For example, virtual network function configuration requires several scheduling operations: the method comprises the steps of life cycle request authorization, resource allocation/release, initial configuration, network virtual function parameter configuration, network virtual function registration and life cycle change request/response; the estimation device 11 may analyze the obtained virtualized network function description information to perform time estimation on each scheduling operation, or divide the obtained virtualized network function description information into N stages, perform time estimation on each stage for scheduling operation, and the like.

Preferably, the virtualized network function description information includes at least any one of:

-a virtualized network function descriptor: i.e., VNFD, i.e., Virtualized Network Function Descriptor, Virtualized Network Function module Descriptor;

-mirror image and size of the mirror image: a mirror and a mirror size involved in the virtualized network function;

-network load information: i.e. network load information of the virtual network.

The virtualized network function description information may be acquired in real time or preset by the system.

For example, the virtual network function manager 1 parses the VNFD, and then the predicting means 11 predicts the maximum time spent for each scheduling operation based on the virtualized network function description information.

The estimated sending device 12 of the virtual network function manager 1 sends the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager; accordingly, the estimate obtaining means 21 of the network function virtual orchestrator 2 obtains the maximum time that each scheduling operation takes, which is estimated by the virtual network function manager corresponding to the network function virtual orchestrator.

Specifically, the estimated sending device 12 may send the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager through an independent message; the estimated maximum time may also be sent to the network function virtualizer along with other requests (e.g., allowed lifecycle requests, etc.).

Accordingly, the estimated acquisition means 21 of the network function virtualizer 2 acquires the estimated maximum time for time control of each scheduling operation.

The association means 13 of the virtual network function manager 1 associates the virtual network function with the virtual resource after the virtual resource has been allocated.

In particular, the virtual network function manager 1 may interact with a virtual infrastructure manager (e.g. VIM) to inform the virtual infrastructure manager of resource allocation; after acquiring the resources allocated by the virtual infrastructure manager, associating the virtual network function with the allocated virtual resources, i.e. before performing the initial configuration, so as to subsequently send an event notification to the network function virtual orchestrator.

The notification sending device 14 of the virtual network function manager 1 sends an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation; accordingly, the notification obtaining device 22 of the network function virtual orchestrator 2 obtains one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation.

Specifically, the notification sending device 14 monitors each operation of the virtual network function manager 1 in virtual resource allocation in real time, and sends information such as an operation action of the operation and a resource related to the operation to the network function virtual orchestrator 2 for the notification obtaining device 22 to obtain each operation, each time the virtual network function manager executes one operation.

For example, when the virtual network function manager 1 performs resource allocation and associates the virtual network function with a virtual resource, an event notification is sent to the network function virtual orchestrator 2, which may include the following: "operation; completing distribution; a virtual network function identifier; the allocated resources are 2 virtual machines and one virtual network "; when the virtual network function manager 1 performs initial configuration on the allocated resources, the virtual network function manager 1 sends an event notification to the network function virtual orchestrator 2, which may include the following: "operation; completing initialization configuration; a virtual network function identifier; the allocated resources are 2 virtual machines and one virtual network "; when the virtual network function manager 1 configures a virtual network function, the virtual network function manager 1 sends an event notification to the network function virtual orchestrator 2, which may include the following: "operation; completing the configuration of the scheduling parameters; a virtual network function identifier; the allocated resources are 2 virtual machines and one virtual network ", and so on.

It will be appreciated by those skilled in the art that during one virtual resource allocation operation, the virtual network function manager 1 may send event notifications to the network function virtual orchestrator 2 multiple times, rather than only once, based on the operation; the event notifications sent are not completely consistent according to different operation flows. Furthermore, the virtual network function manager 1 may send an event notification regardless of whether the operation is completed, and inform the network function virtual orchestrator 2 of "operation not completed" in the event notification.

The monitoring device 23 of the network function virtual orchestrator 2 monitors each scheduling operation according to the estimated maximum time in combination with the event notification to determine whether the scheduling operation is performed normally.

Specifically, the monitoring device 23 compares the sending time of the event notification, the completion condition of the operation action in the event notification, and other information with the estimated maximum time to determine whether the scheduling operation is completed within the estimated maximum time, so that the network function virtual orchestrator 2 maintains time control and avoids long-term request operation. If the event is finished within the estimated time, waiting for the next event notification and monitoring; if not, then the subsequent operation steps are performed, such as adjusting the estimated maximum time spent by each scheduling operation and determining resource processing instructions.

Here, it should be understood by those skilled in the art that the monitoring of the monitoring device 23 is performed for each scheduling operation, that is, each time the notification obtaining device 22 obtains an event notification, the monitoring device 23 monitors the event notification in combination with the estimated maximum time; in addition, if no corresponding event notification is received within the estimated maximum time, the monitoring device 23 also monitors this situation. Based on the monitoring, the monitoring means 23 may determine, for example, when the allocated resources are suitable for registration, when suitable for virtual network function scheduling parameter configuration, etc. Since the virtual resource is already associated with a virtual network function, this equates to a hint that the network function virtualizer 2 is able to perform the virtual network function of the virtual resource.

Preferably, the network function virtualizer 2 further comprises an adjusting means (not shown), wherein the adjusting means adjusts the estimated maximum time spent for each scheduling operation according to the monitored monitoring result.

Specifically, the adjusting device may adjust the estimated maximum time spent by each scheduling operation according to the monitoring result of the monitoring, such as corresponding information about whether the scheduling operation is overtime or not, whether the scheduling operation is completed in advance, and the like, so as to be used for monitoring the next task by the monitoring device 23.

For example, if the scheduling operation frequently times out, the estimated maximum time is extended; if the scheduling operation is frequently or substantially completed in advance, the estimated maximum time is reduced.

Fig. 3 is a schematic device diagram of a virtual network function manager 1 and a network function virtual orchestrator 2 for controlling cloud platform resource allocation according to another preferred embodiment of the present invention; the virtual network function manager 1 includes an estimation device 11 ', an estimation sending device 12 ', a correlation device 13 ', a notification sending device 14 ' and an instruction obtaining device 15 ', and the network function virtual orchestrator 2 includes an estimation obtaining device 21 ', a notification obtaining device 22 ', a monitoring device 23 ', a state determining device 24 ', an instruction determining device 25 ' and an instruction sending device 26 '.

Specifically, the estimation device 11' of the virtual network function manager 1 estimates the maximum time spent by each scheduling operation when executing the virtual network function resource allocation operation according to the virtualized network function description information; the estimated sending device 12' of the virtual network function manager 1 sends the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager; correspondingly, the estimation obtaining means 21' of the network function virtual orchestrator 2 obtains the maximum time that each scheduling operation takes, which is estimated by a virtual network function manager corresponding to the network function virtual orchestrator; the associating means 13' of the virtual network function manager 1 associates a virtual network function with a virtual resource after the virtual resource has been allocated; the notification sending device 14' of the virtual network function manager 1 sends an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation; accordingly, the notification obtaining device 22' of the network function virtual orchestrator 2 obtains one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation; the monitoring device 23' of the network function virtual orchestrator 2 monitors each scheduling operation according to the estimated maximum time in combination with the event notification to determine whether the scheduling operation is normally performed;

if the scheduling operation is not performed normally, the state determining device 24' of the network function virtual orchestrator 2 determines an operation state of the scheduling operation, where the operation state includes a resource state corresponding to the scheduling operation; the instruction determining device 25' determines a resource processing instruction corresponding to the scheduling operation according to the operation state; the instruction sending device 26' sends the resource processing instruction to the virtual network function manager; accordingly, the instruction obtaining means 14' of the virtual network function manager 1 obtains and executes the resource processing instruction sent by the network function virtual orchestrator, wherein the resource processing instruction is determined based on the event notification.

The estimation device 11 ', the estimation sending device 12 ', the association device 13 ', and the notification sending device 14 ' of the virtual network function manager 1 are the same as or similar to the estimation obtaining device 21 ', the notification obtaining device 22 ', and the monitoring device 23 ' of the network function virtual orchestrator 2, and therefore are not described herein again and are included herein by way of reference.

If the scheduling operation is not performed normally, the state determining device 24' of the network function virtual orchestrator 2 determines an operation state of the scheduling operation, where the operation state includes a resource state corresponding to the scheduling operation.

Specifically, if the monitoring device 23 'determines that the scheduling operation is not performed normally based on a timeout or failure feedback of event notification, the state determining device 24' may determine the operation state of the scheduling operation, where the operation state includes, but is not limited to, a timeout, a partial resource scheduling failure, a full resource scheduling success, and the like.

For example, the state determination device 24' may determine the resource state corresponding to the scheduling operation according to information such as allocated resources included in the event notification.

The instruction determining means 25' determines a resource processing instruction corresponding to the scheduling operation according to the operation state.

Specifically, the instruction determining device 25' determines the resource processing instruction corresponding to the scheduling operation according to different operation states, such as when the resource states are different, according to a predetermined processing rule.

For example, if all resources are successfully scheduled in the operating state, the monitoring result of the next scheduling operation may be directly waited for; alternatively, for example, if only a timeout is displayed in the operation state, a retry operation may be determined as a resource processing instruction, and if the retry operation fails after several times, all resources may be released.

Further, if some or all of the resource scheduling is not successful in the operating state, preferably, the resource processing instruction includes at least any one of the following:

-declaring the resources corresponding to the scheduling operation as orphaned resources and reallocating resources for the scheduling operation: if part of the resources are successfully allocated in the operation state, declaring the resources which are not successfully allocated as isolated resources, and attempting to use another resource for replacement based on the type of the isolated resources so as to reallocate the resources for the scheduling operation.

-stopping the scheduling operation and releasing all resources: and if all the resources are failed to be allocated in the operating state, stopping the tasks of the virtual network function and releasing all the allocated virtual resources.

The instruction sending device 26' sends the resource processing instruction to the virtual network function manager; accordingly, the instruction obtaining means 14' of the virtual network function manager 1 obtains and executes the resource processing instruction sent by the network function virtual orchestrator, wherein the resource processing instruction is determined based on the event notification.

Specifically, the instruction obtaining device 14' performs operations such as reallocation or resource release based on the resource processing instruction after obtaining the resource processing instruction.

FIG. 4 illustrates a flow diagram of a method implemented by a virtual network function manager in cooperation with a network function virtualization orchestrator for controlling cloud platform resource allocation, according to another aspect of the invention.

Specifically, in step S41, the virtual network function manager 1 estimates the maximum time spent by each scheduling operation when performing the virtual network function resource allocation operation according to the virtualized network function description information; in step S42, the virtual network function manager 1 sends the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager; accordingly, in step S42, the network function virtual orchestrator 2 obtains the maximum time spent by each scheduling operation, which is estimated by a virtual network function manager corresponding to the network function virtual orchestrator; in step S43, the virtual network function manager 1 associates the virtual network function with the virtual resource after the virtual resource is allocated; in step S44, the virtual network function manager 1 sends an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation; accordingly, in step S44, the network function virtual orchestrator 2 acquires one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation; in step S45, the network function virtualizer 2 monitors each scheduling operation according to the estimated maximum time and in combination with the event notification, to determine whether the scheduling operation is performed normally.

In step S41, the virtual network function manager 1 estimates the maximum time spent by each scheduling operation when performing the virtual network function resource allocation operation, based on the virtualized network function description information.

Specifically, in step S41, the virtual network function manager 1 may determine, according to the system configuration, each scheduling operation required when executing the virtual network function configuration; then, determining the virtualized network function description information based on historical statistical data of the whole system or based on data such as the size of data packets, the number of data packets, system capacity, operation flow, resource conditions and the like in the system, or interacting with other devices capable of providing the virtualized network function description information to directly obtain the virtualized network function description information; the virtualized network function description information is then parsed to determine the maximum time spent for each scheduling operation. Here, the maximum time is an estimated value.

For example, virtual network function configuration requires several scheduling operations: the method comprises the steps of life cycle request authorization, resource allocation/release, initial configuration, network virtual function parameter configuration, network virtual function registration and life cycle change request/response; the virtual network function manager 1 may parse the obtained virtualized network function description information to perform time estimation for each scheduling operation, or divide the virtualized network function description information into N stages, perform time estimation for each stage for scheduling operation, and so on in step S41.

For example, the virtual network function manager 1 parses the VNFD, and then each scheduling operation predicts the maximum time spent based on the virtualized network function description information.

In step S42, the virtual network function manager 1 sends the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager; accordingly, in step S42, the network function virtualizer 2 acquires the maximum time spent by each scheduling operation, which is estimated by the virtual network function manager corresponding to the network function virtualizer.

Specifically, in step S42, the virtual network function manager 1 may send the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager through an independent message; the estimated maximum time may also be sent to the network function virtualizer along with other requests (e.g., allowed lifecycle requests, etc.).

Accordingly, in step S42, the network function virtualizer 2 acquires the estimated maximum time for time control of the respective scheduling operations.

In step S43, the virtual network function manager 1 associates the virtual network function with the virtual resource after the virtual resource is allocated.

Specifically, in step S43, the virtual network function manager 1 may interact with a virtual infrastructure manager (e.g., VIM) to inform the virtual infrastructure manager of resource allocation; after acquiring the resources allocated by the virtual infrastructure manager, associating the virtual network function with the allocated virtual resources, i.e. before performing the initial configuration, so as to subsequently send an event notification to the network function virtual orchestrator.

In step S44, the virtual network function manager 1 sends an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation; accordingly, in step S44, the network function virtual orchestrator 2 obtains one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation.

Specifically, in step S44, the virtual network function manager 1 monitors each operation of the virtual network function manager 1 in virtual resource allocation in real time, and transmits information such as an operation action of the operation and a resource related to the operation to the network function virtual orchestrator 2 every time the virtual network function manager executes one operation.

In step S45, the network function virtualizer 2 monitors each scheduling operation according to the estimated maximum time and in combination with the event notification, to determine whether the scheduling operation is performed normally.

Specifically, in step S45, the network function virtual organizer 2 compares the sending time of the event notification, the completion of the operation action in the event notification, and other information with the estimated maximum time to determine whether the scheduling operation is completed within the estimated maximum time, so that the network function virtual organizer 2 maintains time control to avoid long-term request operation. If the event is finished within the estimated time, waiting for the next event notification and monitoring; if not, then the subsequent operation steps are performed, such as adjusting the estimated maximum time spent by each scheduling operation and determining resource processing instructions.

Here, it should be understood by those skilled in the art that the monitoring of the network function virtualizer 2 is performed for each scheduling operation, that is, each time an event notification is obtained, the network function virtualizer 2 monitors the event notification in combination with the estimated maximum time; in addition, if no corresponding event notification is received within the estimated maximum time, the network function virtualizer 2 may monitor this situation. Based on the monitoring, the network function virtual orchestrator 2 may determine, for example, when the allocated resources are suitable for registration, when the virtual network function scheduling parameter configuration is suitable, etc. Since the virtual resource is already associated with a virtual network function, this equates to a hint that the network function virtualizer 2 is able to perform the virtual network function of the virtual resource.

Preferably, the network function virtual orchestrator 2 is capable of adjusting the estimated maximum time spent for each scheduling operation according to the monitored monitoring result.

Specifically, the network function virtual orchestrator 2 may adjust the estimated maximum time spent by each scheduling operation according to the monitoring result, such as corresponding information about whether the monitoring result is overtime or not, whether the monitoring result is completed in advance, and the like, so as to monitor the next task.

Fig. 5 is a flowchart of a method implemented by a virtual network function manager in cooperation with a network function virtual orchestrator for controlling cloud platform resource allocation, according to a preferred embodiment of the present invention.

Specifically, in step S51', the virtual network function manager 1 estimates the maximum time spent by each scheduling operation when performing the virtual network function resource allocation operation according to the virtualized network function description information; in step S52', the virtual network function manager 1 sends the estimated maximum time to the network function virtual orchestrator corresponding to the virtual network function manager; accordingly, in step S52', the network function virtualizer 2 obtains the maximum time spent by each scheduling operation estimated by the virtual network function manager corresponding to the network function virtualizer; in step S53', the virtual network function manager 1 associates the virtual network function with the virtual resource after the virtual resource is allocated; in step S54', the virtual network function manager 1 sends an event notification to the network function virtual orchestrator according to each operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the operation and a resource corresponding to the operation; accordingly, in step S54', the network function virtual orchestrator 2 obtains one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the operation and resources corresponding to the operation; in step S55', the network function virtualization orchestrator 2 monitors each scheduling operation according to the estimated maximum time, in combination with the event notification, to determine whether the scheduling operation is performed normally; if the scheduling operation is not performed normally, in step S56', the network function virtualization orchestrator 2 determines an operation state of the scheduling operation, where the operation state includes a resource state corresponding to the scheduling operation; in step S57', the network function virtualizer 2 determines a resource processing instruction corresponding to the scheduling operation according to the operation state; in step S58', the network function virtual orchestrator 2 sends the resource processing instruction to the virtual network function manager; accordingly, in step S58', the virtual network function manager 1 acquires and executes the resource processing instruction sent by the network function virtual orchestrator, wherein the resource processing instruction is determined based on the event notification.

The steps S51 ', S52 ', S53 ', S54 ' and S55 ' are respectively the same as or similar to the corresponding steps S41, S42, S43, S44 and S45 in fig. 4, and therefore are not described herein again and are included herein by reference.

If the scheduling operation is not performed normally, in step S56', the network function virtualization orchestrator 2 determines an operation state of the scheduling operation, where the operation state includes a resource state corresponding to the scheduling operation.

Specifically, if the network function virtual orchestrator 2 determines that the scheduling operation is not performed normally based on a timeout or failure feedback of event notification, in step S56', the network function virtual orchestrator 2 may determine an operation state of the scheduling operation, where the operation state includes, but is not limited to, a timeout, a partial resource scheduling failure, a full resource scheduling success, and the like.

For example, the network function virtualizer 2 may determine the resource status corresponding to the scheduling operation according to information such as allocated resources included in the event notification.

In step S57', the network function virtualizer 2 determines a resource processing instruction corresponding to the scheduling operation according to the operation state.

Specifically, in step S57', the network function virtual orchestrator 2 determines a resource processing instruction corresponding to the scheduling operation according to different operation states, such as according to a predetermined processing rule, when the resource states are different.

In step S58', the network function virtual orchestrator 2 sends the resource processing instruction to the virtual network function manager; accordingly, in step S58', the virtual network function manager 1 acquires and executes the resource processing instruction sent by the network function virtual orchestrator, wherein the resource processing instruction is determined based on the event notification.

Specifically, in step S58', the virtual network function manager 1, after acquiring the resource processing instruction, performs operations such as reallocation or resource release based on the resource processing instruction.

Fig. 6 shows a flow chart of a method for VNF resource allocation according to another preferred embodiment of the present invention. The flow chart shows an example where no orphaned resources are present at VNF resource allocation.

In step S60, the NVFM is triggered to perform VNF operations, such as VNF, (dynamic) measure VNF, terminate VNF, and the like.

The NVFM creates the task and parses the VNFD and estimates the maximum time for each step of the VNF deployment based on the VNFD.

A timing analysis engine is deployed in the NFVO to monitor the various operations. In this example, the timing analysis engine includes 4 timers, timer 1, timer 2, timer 3, and timer 4. Each timer is used for timing respectively so as to judge whether the corresponding operation is normally finished.

In step S61, the VNFM sends a permitted lifecycle request to the NFVO, where the permitted lifecycle request includes the VNF, and the event: example/scale and duration of each step, etc.

Accordingly, the NFVO sends an allowed lifecycle response to the VNFM, which includes the VIM, the VNF, the principals, etc.

In step S62, the VNFM sends an allocate/release resource command to the VIM and obtains the resources of 1 virtual network and 2 VMs.

The VNFM then associates the VNF with the VM.

In step S63, the VNFM sends a task event status notification to the NVFO, including the operation, assignment completion, VNF id, and assigned 2 VMs and 1 virtual network.

At this time, in NFVO, mapping is performed on VNF, VIM, and resource pool, that is, VNF: 2 VMs (VM1 and VM2) +1 virtual network.

In step S64, the VIM sends a notification resource change request to the NFVO; accordingly, the NFVO sends a notify resource change response to the VIM.

In step S65, the VNFM performs initialization configuration on the allocated resources, such as detecting whether the resources are valid or not.

In step S66, the VNFM sends a task event status notification to the NFVO, which informs the current operation status, including, for example, operation, initial configuration completion, VNF id, configured resources of 2 VMs and 1 virtual network, etc.

In step S67, the VNFM configures NVF scheduling parameters.

In step S68, the VNFM sends a task event status notification to the NFVO, which informs the current operation status, including, for example, operation, scheduling configuration completion, VNF id, configured resources of 2 VMs and 1 virtual network, etc.

In step S69, the VNFM registers the VNF to an EMS (Enhanced Management System).

In step S610, the VNFM sends a VNF lifecycle change request to the NFVO, where the VNF lifecycle change request includes operations, VNF completion, resources of 2 VMs, and one virtual network.

And finally, the NFVO sends a VNF lifecycle change response to the VNFM, and the whole VNF resource allocation process is completed.

Fig. 7 shows a flowchart of a method for VNF resource allocation and release according to another preferred embodiment of the present invention. The flow chart shows an example of isolated resources occurring at VNF resource allocation.

In step S70, the NVFM is triggered to perform VNF operations, such as VNF, (dynamic) measure VNF, terminate VNF, and the like.

In step S71, the VNFM sends a permitted lifecycle request to the NFVO, where the permitted lifecycle request includes the VNF, and the event: example/scale and duration of each step, etc.

In step S72, the VNFM sends an allocate/release resource command to the VIM, and obtains resources of 1 virtual network and 2 VMs (VM1 and VM 2).

The VNFM then associates the VNF with the VM.

In step S73, the VNFM sends a task event status notification to the NVFO, including the operation, assignment completion, VNF id, and assigned 2 VMs and 1 virtual network.

In step S74, the VIM sends a notification resource change request to the NFVO; accordingly, the NFVO sends a notify resource change response to the VIM.

In step S751, the VNFM performs initialization configuration on the allocated resources, such as detecting whether the resources are valid or not. The initialization configuration for the virtual network and VM1 is successful.

Whereas in step S752, the initialization configuration for the VM2 is unsuccessful.

In step S76, the VNFM sends a task event status notification to the NFVO informing of the current operating state, including, for example, operation, initial configuration completion, VNF id, configured resources of 1 VM (VM1) and 1 virtual network, and VM2 is not included in the current operating state.

At this time, in NFVO, mapping is performed on VNF, VIM, and resource pool, that is, VNF: 1 VM (VM1) +1 virtual network, VM2 is an isolated resource.

In step S77, the NFVO sends a terminate resource command to the VNFM, where the command includes information such as VIM, VNF, principle, VM2, and the like.

In step S78, the VNFM informs the VM2 to release the resource.

In step S79, NFVO scales the new resource, i.e., VM3, and sends this information to VNFM; the information includes VIM, VNF, principals, and VM 3.

In step S710, the VNFM interacts with the VM3 to allocate resources.

In step S711, the VNFM sends a task event status notification to the NFVO, which informs the current operation status, including, for example, operation, initial configuration completion, VNF id, and configured resource is VM 3.

In step S712, the VNFM performs initialization configuration on the VM 3.

In step S713, the VNFM configures NVF scheduling parameters.

In step S714, the VNFM sends a task event status notification to the NFVO, which informs the current operation status, including, for example, operation, scheduling configuration completion, VNF id, configured resources of 2 VMs (VM1 and VM3), 1 virtual network, and the like.

At this time, in NFVO, mapping is performed on VNF, VIM, and resource pool, that is, VNF: 2 VMs (VM1 and VM3) +1 virtual network.

In step S715, the VNFM registers the VNF to the EMS.

In step S716, the VNFM sends a VNF lifecycle change request to the NFVO, where the VNF lifecycle change request includes operations, VNF completion, resources of 2 VMs, and one virtual network.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims

1. A method in a virtual network function manager for controlling cloud platform resource allocation, wherein the method comprises the steps of:

according to the description information of the virtual network function, the maximum time spent by each scheduling operation when the virtual network function resource allocation operation is executed is estimated;

sending the estimated maximum time to a network function virtual orchestrator corresponding to the virtual network function manager;

wherein, the method also comprises:

associating the virtual network function with the virtual resource after acquiring the virtual resource allocated by the virtual infrastructure manager;

and sending an event notification to the network function virtual orchestrator according to each scheduling operation of the virtual network function manager during virtual resource allocation, wherein the event notification comprises an operation action of the scheduling operation and a resource corresponding to the scheduling operation, and the virtual network function manager sends the event notification corresponding to the operation to the network function virtual orchestrator every time one operation is executed.

2. The method of claim 1, wherein the virtualized network function description information comprises at least any of:

-a virtualized network function descriptor;

-a mirror image and a size of the mirror image;

-network load information.

3. The method of claim 1 or 2, wherein the method further comprises:

4. A method in a network functions virtualizer for controlling cloud platform resource allocation, wherein the method comprises the steps of:

acquiring the estimated maximum time spent by each scheduling operation by a virtual network function manager, wherein the virtual network function manager corresponds to the network function virtual orchestrator;

acquiring one or more event notifications sent by the virtual network function manager, wherein the event notifications are determined based on each scheduling operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the scheduling operation and resources corresponding to the scheduling operation, and each time an operation is executed by the virtual network function manager, the virtual network function manager sends an event notification corresponding to the operation to the network function virtual orchestrator;

and monitoring each scheduling operation by combining the event notification according to the estimated maximum time so as to determine whether the scheduling operation is normally performed.

5. The method of claim 4, wherein the method further comprises:

6. The method of claim 4 or 5, wherein the method further comprises:

7. The method of claim 6, wherein the resource processing instructions comprise at least any one of:

-stopping the scheduling operation and releasing all resources.

8. A virtual network function manager for controlling cloud platform resource allocation, wherein the virtual network function manager comprises:

wherein the virtual network function manager further comprises:

associating means for associating a virtual network function with a virtual resource after acquiring the virtual resource allocated by the virtual infrastructure manager;

and a notification sending device, configured to send an event notification to the network function virtual orchestrator according to each scheduling operation of the virtual network function manager during virtual resource allocation, where the event notification includes an operation action of the scheduling operation and a resource corresponding to the scheduling operation, and the virtual network function manager sends the event notification corresponding to the operation to the network function virtual orchestrator every time an operation is executed.

9. The virtual network function manager of claim 8, wherein the virtualized network function description information comprises at least any one of:

-a virtualized network function descriptor;

-a mirror image and a size of the mirror image;

-network load information.

10. The virtual network function manager of claim 8 or 9, wherein the virtual network function manager further comprises:

11. A network function virtualizer for controlling cloud platform resource allocation, wherein the network function virtualizer comprises:

a notification obtaining device, configured to obtain one or more event notifications sent by the virtual network function manager, where the event notifications are determined based on each scheduling operation of the virtual network function manager during virtual resource allocation, and the event notifications include operation actions of the scheduling operation and resources corresponding to the scheduling operation, and each time an operation is executed by the virtual network function manager, the virtual network function manager sends an event notification corresponding to the operation to the network function virtual orchestrator;

12. The network function virtualizer as recited in claim 11, wherein the network function virtualizer further comprises:

13. A network function virtualizer as claimed in claim 11 or 12, wherein the network function virtualizer further comprises:

14. The network function virtualization orchestrator of claim 13, wherein the resource processing instructions comprise at least any one of:

-stopping the scheduling operation and releasing all resources.

15. A system for controlling cloud platform resource allocation, comprising a virtual network function manager according to any of claims 8 to 10, a network function virtual orchestrator according to any of claims 11 to 14.