CN114095382B

CN114095382B - Method, system, device and equipment for scheduling virtual resources of network slice

Info

Publication number: CN114095382B
Application number: CN202010776695.8A
Authority: CN
Inventors: 傅友; 蔡锴; 李瑜
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Group Chongqing Co Ltd
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Group Chongqing Co Ltd
Priority date: 2020-08-05
Filing date: 2020-08-05
Publication date: 2023-11-10
Anticipated expiration: 2040-08-05
Also published as: CN114095382A

Abstract

The embodiment of the invention relates to the technical field of communication, and discloses a network slice virtual resource scheduling method, which comprises the following steps: receiving a virtual resource configuration update request, and determining at least one slice to be updated according to the virtual resource configuration update request; acquiring current network operation data, virtual resource allocation weights and the number of idle virtual resources of the at least one slice to be updated; respectively determining mapping relation functions corresponding to the at least one slice to be updated; acquiring the number of idle virtual resources, and determining target allocation virtual resource data which maximizes the sum of the service satisfaction degrees of the slices to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight; and carrying out configuration updating on the at least one slice to be updated according to the target allocation virtual resource data. By the mode, the scheduling efficiency of the virtual resources is improved when the plurality of network slices are updated.

Description

Method, system, device and equipment for scheduling virtual resources of network slice

Technical Field

The embodiment of the invention relates to the technical field of communication, in particular to a method, a system, a device and equipment for scheduling virtual resources of network slices.

Background

Network slicing is one of the key technical features of network function virtualization (network function virtualization, NFV) and software defined networking (software defined network, SDN) applications to fifth generation (5G) communication networks. A physical network may be logically divided into a plurality of network slices, each constituting a separate end-to-end network serving customers, logically isolated from each other. Thus, the network slice may serve as a type NaaS (Network as a Service) to provide customized services to vertical business users. The user can subscribe to the network slice at the operator to bear the own application service, and the network slice provided by the operator for the user can meet the service requirement of the application service, namely, meet the requirements of service level agreements (service level agreement, SLA).

In order to better meet the needs of slicing tenants, the network slicing can perform dynamic resource scheduling according to the meeting conditions of the tenant SLAs in the running process, and virtual resources (such as the number of virtual machines, CPU, memory, network bandwidth and the like) of the VNF (virtual network function) in the core network sub-slice are rescheduled and allocated, so that slice updating is realized.

The method for updating the slice in the prior art mainly comprises two methods: the system is configured for manual work, a slice tenant applies for modifying resource configuration information of a slice to a slice management system according to self service experience data and queried slice operation conditions, the slice management system applies for virtual resource configuration update to a MANO (Management and Orchestration, network management orchestrator) after receiving the application, and the system automatically collects service experience data (such as service satisfaction) of a slice KPI and the tenant and feeds the service experience data back to the slice management system, and the slice management system automatically applies for the slice virtual resource configuration update to the MANO when SLA is difficult to satisfy.

The problem of doing so is: in the prior art, the relation between the tenant SLA or service experience data and the sliced KPI and virtual resources is not given, only the fact that the sliced virtual resources need to be added when the SLA cannot be met is shown, and an explicit resource scheduling flow is not given. Meanwhile, in the existing scheme, only the situation that a single slice needs to be updated is considered, and when a plurality of tenants need to update the slice virtual resources at the same time under the condition of limited resources, a specific resource allocation scheduling principle and a specific resource allocation scheduling method are not considered, and the defects in the prior art cause lower scheduling efficiency and poor scheduling effect of the virtual resources for a plurality of network slices.

Disclosure of Invention

In view of the above problems, embodiments of the present invention provide a method, an apparatus, a device, and a readable medium for scheduling virtual resources of network slices, which are used to solve the problems in the prior art that the scheduling efficiency of network slice resources is low and the scheduling effect is poor.

According to an aspect of an embodiment of the present invention, there is provided a network slice virtual resource scheduling method, including:

receiving a virtual resource configuration update request, and determining at least one slice to be updated according to the virtual resource configuration update request;

acquiring current network operation data, virtual resource allocation weights and the number of idle virtual resources of the at least one slice to be updated;

respectively determining a mapping relation function corresponding to the at least one slice to be updated, wherein the mapping relation function is used for representing the functional relation among the current network operation data of the slice to be updated, the slice allocation virtual resource data and the service satisfaction, and the mapping relation function is obtained by inputting a preset regression model training according to the historical slice network operation data, the historical slice allocation virtual resource data and the corresponding historical service satisfaction;

determining target allocation virtual resource data which maximizes the sum of the service satisfaction degrees of the slices to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight;

And carrying out configuration updating on the at least one slice to be updated according to the target allocation virtual resource data.

In an optional manner, the current network operation data at least includes the number of load users, processing traffic, coverage area and network KPI of the current slice, and the slice allocation virtual resource data at least includes the number of virtual machines, the number of virtual CPUs, the virtual memory capacity and the network bandwidth corresponding to the slice.

In an optional manner, the number of free virtual resources includes a number of free virtual machines, and the determining the target allocation virtual resource data that maximizes a sum of the service satisfaction degrees of the slices to be updated further includes:

traversing each combination of the to-be-updated slices not exceeding the number of the idle virtual machines as an optional resource configuration combination,

determining the sum of service satisfaction of all the slices to be updated corresponding to each selectable resource configuration combination according to the current network operation data, the mapping relation function and the virtual resource allocation weight corresponding to each slice to be updated;

acquiring the optional resource configuration combination which enables the sum of the service satisfaction degree of the slice to be updated to be maximum as the target allocation virtual resource data;

And under the condition that the sum of service satisfaction corresponding to two or more selectable resource configuration combinations is the same, acquiring the selectable resource configuration combination which minimizes the sum of the numbers of the distributed virtual machines corresponding to all the slices to be updated as the target distributed virtual resource data.

In an optional manner, the number of idle virtual resources further includes a number of idle virtual CPUs, an idle virtual memory capacity, and an idle network bandwidth, and the determining the target allocation virtual resource data that maximizes the sum of the service satisfaction of the slice to be updated further includes:

traversing the combination of the number of the idle virtual CPUs, the idle virtual memory capacity and the idle network bandwidth of each slice to be updated as an optional resource configuration combination;

acquiring an optional resource configuration combination which enables the sum of service satisfaction of all slices to be updated to be maximum as the target allocation virtual resource data;

Under the condition that the sum of service satisfaction degrees corresponding to two or more than two optional resource allocation combinations is the same, respectively acquiring the number of idle virtual CPUs, the idle virtual memory capacity and the resource unit price corresponding to idle network bandwidth, and respectively determining the resource total price of the optional resource allocation combinations with the same sum of the service satisfaction degrees according to the resource unit price;

and acquiring the optional virtual resource configuration combination with the minimum total price of resources as the target allocation virtual resource data.

According to an aspect of an embodiment of the present invention, there is provided a network slice virtual resource scheduling system, including:

the CSMF module is used for receiving a network slice update request of a tenant, notifying the NSMF module that the network slice of the tenant needs to be configured and updated, and notifying the network slice ID;

the NSMF module is used for evaluating and decomposing a received slice update request according to slice network data, determining whether the network slice needs virtual resource configuration update, determining the network slice ID as a slice ID to be updated when the network slice needs virtual resource configuration update, and sending the virtual resource configuration update request of the slice to be updated to the NSSMF module;

The NSSMF module is used for sending a virtual resource configuration update request of the slice to be updated to the NRSDF module, wherein the virtual resource configuration update request of the slice to be updated carries the ID of the slice to be updated, applying for virtual resource configuration update to the MANO according to the received target allocation virtual resource quantity, and reporting the virtual resource configuration update condition of the network slice to the CSMF through the NSMF;

the NRSDF module is used for sending a mapping relation function of the service satisfaction degree of the sub-slice corresponding to the slice ID to be updated and the virtual resources to the NWAF module after receiving the virtual resource configuration update request, obtaining the number of idle virtual resources which can be allocated currently to the MANO, determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight, solving to obtain the number of target allocation virtual resources, and feeding back the number of target allocation virtual resources to the NSSMF;

the NWAF module is used for feeding back the mapping relation function of the sub-slice corresponding to the slice ID to be updated and the resource allocation weight of the tenant corresponding to the slice to be updated to the NRSDF;

And the MANO module is used for feeding back the number of the idle virtual resources.

According to another aspect of the embodiment of the present invention, there is provided a network slice virtual resource scheduling apparatus, including:

the request receiving module is used for receiving a virtual resource configuration update request and determining at least one slice to be updated according to the virtual resource configuration update request;

the model determining module is used for acquiring the current network operation data of the at least one slice to be updated, the virtual resource allocation weight and the mapping relation function, wherein the mapping relation function represents the functional relation among the slice network operation data, the slice allocation virtual resource data and the service satisfaction degree;

the scheme determining module is used for obtaining the number of idle virtual resources, and determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight;

and the configuration updating module is used for carrying out configuration updating on the at least one slice to be updated according to the target allocation virtual resource data.

In an alternative manner, the configuration scheme determining module is further configured to: traversing the combinations of which the number of the to-be-updated slices does not exceed that of the idle virtual machines as selectable resource configuration combinations;

In another alternative, the configuration scheme determination module may be further configured to:

traversing the combination of the number of idle virtual CPUs, the idle virtual memory capacity and the idle network bandwidth under each slice to be updated as an optional resource configuration combination;

Acquiring an optional resource allocation combination which enables the sum of the service satisfaction degree to be maximum as the target allocation virtual resource data;

under the condition that the sum of service satisfaction degrees corresponding to two or more than two optional resource allocation combinations is the same, respectively acquiring the number of virtual CPUs, the virtual memory capacity and the resource unit price corresponding to network bandwidth, and respectively determining the resource total price of the optional resource allocation combinations with the same sum of the service satisfaction degrees according to the resource unit price;

The embodiment of the invention receives a virtual resource configuration update request, and determines at least one slice to be updated according to the virtual resource configuration update request;

According to the scheme, the invention realizes the integer programming problem that the total number of the allocated resources of all the slices to be updated does not exceed the existing callable resources as constraint conditions and the sum of the service satisfaction of all the slices to be updated is maximized as an objective function, and the final overall resource allocation scheme is obtained by solving the integer programming problem, so that the problems that in the prior art, a clear resource scheduling flow is not provided, only the situation that a single slice needs to be updated is considered, and under the condition that resources are limited, the network slice virtual resource scheduling efficiency is low and the scheduling effect is poor due to the resource scheduling method that a plurality of tenants simultaneously need to update the slice virtual resources are not considered.

The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and may be implemented according to the content of the specification, so that the technical means of the embodiments of the present invention can be more clearly understood, and the following specific embodiments of the present invention are given for clarity and understanding.

Drawings

The drawings are only for purposes of illustrating embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:

FIG. 1 illustrates a flow chart of an embodiment of a network slice virtual resource scheduling method of the present invention;

FIG. 2 illustrates an interactive flow diagram for network slice virtual resource update in one embodiment;

FIG. 3 illustrates a flow diagram for determining target allocation virtual resource data in one embodiment;

FIG. 4 illustrates a flow chart of determining target allocation virtual resource data in another embodiment;

FIG. 5 is a schematic diagram illustrating the architecture of an embodiment of a network slice virtual resource scheduling apparatus according to the present invention;

fig. 6 shows a schematic structural diagram of an embodiment of the network slice virtual resource scheduling device of the present invention.

Detailed Description

Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.

Fig. 1 shows a flow chart of an embodiment of the network slice virtual resource scheduling method of the present invention, which method may be implemented by means of a computer program that is executable on a von neumann system-based computer system, in particular, on a computer processing device such as a cell phone, a notebook computer, etc. As shown in fig. 1, the method may include at least steps 110-150 as shown in fig. 1:

step 110: and receiving a virtual resource configuration update request, and determining at least one slice to be updated according to the virtual resource configuration update request.

Firstly, updating the network slice resources can be two cases, one is that a user can autonomously send a slice update request under the condition that the slice network experience is poor (such as poor network signals, slow network response speed and the like), and then the system analyzes and evaluates the slice update request sent by a specific user to judge whether the service satisfaction degree is insufficient caused by the mismatch of virtual resource allocation of core network sub-slices.

This is because in practice, there is no need to reschedule virtual resources of sub-slices of the core network in the slice, since some network performance metrics are independent of the core network, such as signal strength, etc.

However, at the same time, although the core network subslice need not be updated, it is possible to update the radio subslice or the transmission subslice (i.e., the subslice of the non-core network type), and evaluate and determine the specific subslice type to be updated.

Alternatively, the user may directly monitor in real time through a slice management function (NSMF) module in the system without manually sending a slice update request, and detect that the service satisfaction does not meet

And acquiring the service satisfaction degree of each tenant in the target network in real time through the slice management function module, and determining whether the tenant with the service satisfaction degree smaller than a preset satisfaction degree threshold exists.

In connection with the foregoing illustration, a specific analysis may be to determine, by the NSMF module, whether the current service satisfaction is caused by a virtual resource allocation mismatch of the core network sub-slices.

I.e. screening out the situation which is irrelevant to the core network and causes lower service satisfaction, such as weaker signal strength of the base station, etc., and the situation does not involve rescheduling the virtual resources of the core network subslice in the slice.

It should be noted that, the satisfaction threshold may be determined by historical user experience data within a certain period of time, or may be determined by acquiring an average service satisfaction of all network slices in the current network.

The service satisfaction degree can be obtained by automatically calculating according to a preset formula through the current network operation, or can be the score submitted by the user. In a specific embodiment, the service satisfaction may be a service MOS (Mean Opinion Score ) satisfaction submitted by the user.

Step 120: and acquiring the current network operation data, the virtual resource allocation weight and the number of idle virtual resources of the at least one slice to be updated.

Step 130: and respectively determining a mapping relation function corresponding to the at least one slice to be updated, wherein the mapping relation function is used for representing the functional relation among the current network operation data of the slice to be updated, the slice allocation virtual resource data and the service satisfaction, and the mapping relation function is obtained by inputting a preset regression model training according to the historical slice network operation data, the historical slice allocation virtual resource data and the corresponding historical service satisfaction.

Before a specific network slice resource scheduling method is described, the principle and characteristics of network slices are described first.

The 5G end-to-end network slicing refers to flexible allocation of network resources, virtual generation of a plurality of logic subnets with different characteristics based on a 5G network, and each end-to-end slicing is formed by combining core network, wireless network and transmission network sub-slices and unified management through a slice management system. Different parts of the Network Slice Instance (NSI) are grouped into network slice subnets (e.g., RAN, 5GC and transport) and the lifecycle of the network slice subnets instance (NSI) is allowed to be managed independent of the lifecycle of the NSI.

The network slice serves as a NaaS (Network as a Service) customized service for users in the vertical industry, the users can subscribe to the network slice at an operator to bear own application business, and the network slice provided by the operator for the users can meet the service requirement of the application business, namely, the service level agreement (service level agreement, SLA) requirement.

Meanwhile, in order to meet the needs of the slicing tenant, the network slice can perform dynamic resource scheduling according to the meeting condition of the tenant SLA in the running process, and virtual resources (such as the number of virtual machines, CPU, memory, network bandwidth and the like) of the VNF (virtual network function) in the core network sub-slice are rescheduled and allocated, so that slice updating is realized.

I.e. how much virtual resources are used by each slice has a direct impact on its traffic satisfaction. In general, the more virtual resources are owned, for example, the more virtual machines or the more virtual machine CPUs are owned, and at the same time, when the network running state of each slice is changed under the condition that the number of virtual resources used is unchanged, for example, the number of users born by one slice and the traffic volume to be processed are changed, the service satisfaction degree is also influenced, for example, when the traffic volume to be processed of one slice is increased, the service satisfaction degree is reduced because the virtual resources which can be used by one slice are fixed.

Therefore, in order to maximize the service satisfaction degree of the whole network for the virtual resources of the slices in the network, that is, maximize the proportion of users in the network slices that reach or exceed the service satisfaction degree requirement, it is first required to determine the current network running state of each slice, the current virtual resource condition used, and the final service satisfaction degree of each slice, that is, determine the mapping relation function between the service satisfaction degree corresponding to each slice and the running data of the slice network and the number of the allocated virtual resources, that is, step 130.

In order to schedule virtual resources of network slices with the goal of maximizing service satisfaction, firstly, the virtual resource quantity data currently used by each slice, the running state (bearing and load state) of the slice network in which the virtual resource quantity data is currently located, and the service satisfaction of each slice need to be obtained.

In a specific embodiment, the current network operation data of the slice in step 120 may at least include parameters such as the number of current users, processing traffic, coverage area, network KPI, etc. of each network slice, which are used to evaluate the operation status and load of the current network slice.

Also by way of example, in the case where one network slice uses 5 virtual machines (the virtual resource number configuration is not updated), the number of users it carries is 100 and 10000 in different network operation states, and the service satisfaction corresponding to the slice is different (generally, in the case where the number of virtual resources is unchanged, as the network load increases, the service satisfaction decreases). In addition, the coverage area may refer to a wireless coverage area, and may be understood as the number of base stations accessed by the current network slice, where the larger the number of base stations accessed, the larger the coverage area.

Correspondingly, in combination with the foregoing explanation of the network slicing principle, the allocated virtual resources corresponding to each slice may include at least the number of virtual machines used by the virtual resources, and further may further include parameter items related to virtual resources (mainly referred to as virtual machines) such as the number of virtual CPUs, the virtual memory capacity, the network bandwidth, and the like.

Correspondingly, considering that there may be a difference in importance degree of tenants corresponding to different slices, for example, there may be some tenants with higher priority that need to allocate virtual resources preferentially, so the virtual resource allocation weight in step 120 reflects the priority degree of virtual resource allocation of different slices.

Importantly, the mapping relation function among the network operation data, the service satisfaction degree and the virtual resource quantity of the slice is explained next.

In a specific embodiment, a mapping relationship function between a service satisfaction degree corresponding to a network slice and virtual resources used by the network slice and slice network operation data of the network slice may be recorded as:

MOS _i ＝f _i (x _i ,n _i ,t _i ,v _i )

wherein MOS is provided with _i Service satisfaction for a network slice with sequence number i (which may be determined based on collecting service satisfaction data from a network management orchestration module (i.e., MANO module) through a preset data analysis module during daily operation of the network slice), n _i ，t _i ，v _i Parameter items, x, included in network data such as number of users, traffic, coverage area, etc. respectively carried by clients of the network slice _i Refers to the number of virtual resources of the network slice, f _i Is a mapping relation function fitted by a trained machine learning model.

Specifically, the training process of the mapping relation function may be that a training set is input to train a preset machine learning model, where the training set includes a plurality of network operation data, samples formed by combining virtual resource data and service satisfaction degrees, and functional relations corresponding to the samples.

The training algorithm can be a regression algorithm such as polynomial regression, SVD algorithm and the like, and in an alternative implementation path, the training can also be performed by adopting a neural network algorithm. Therefore, the mapping relation function of each slice is determined according to the situation of the slice, and the mapping relation functions corresponding to the slices are different.

For the mapping relation function, it should be specifically noted that, first, the relevant parameter items included in the network operation data in the mapping relation function are not limited to the above three examples, that is, the mapping relation between the plurality of preset parameter items and the machine model in the overall network operation state can be obtained through training the training data set.

In addition, the unit of the number of virtual resources in the mapping relationship function may be a measurement of virtual resources in various different layers, for example, from the viewpoint of virtual machines, the number of virtual machines may be the number of virtual machines, or may be deeper, for example, the number of memory CPUs of the virtual machines, the memory capacity of the virtual machines, the network bandwidth, and the like.

Therefore, in order to collect the satisfaction degree of the history service experience and the history slice network operation data as a training set to train the mapping relation function and solve the target integer programming problem of virtual resource allocation of a plurality of slices, the invention adds a network resource scheduling decision function module, namely a NRSDF (Network resource scheduling decision function) module, on the basis of the network slice management system architecture in the prior art.

In an alternative embodiment, the invention further comprises a network slice virtual resource scheduling system comprising modules that interact with each other to update the slice virtual resource.

And the CSMF (Communication Service Management Function) module is used for receiving a network slice update request of a tenant, notifying the NSMF module that the network slice of the tenant needs to be configured for updating, and carrying the network slice ID in the notification.

And the NSMF (Network Slice Management Function, slice management function) module is used for evaluating and decomposing the received slice update request according to slice network data, determining whether the network slice needs virtual resource configuration update, determining the network slice ID as a slice ID to be updated when the network slice needs virtual resource configuration update, and sending the virtual resource configuration update request of the slice to be updated to the NSSMF module.

And the NSSMF (Network Slice Subnet Management Function, sub-slice management function) module is used for sending the virtual resource configuration update request of the slice to be updated to the NRSDF module, wherein the virtual resource configuration update request of the slice to be updated carries the slice ID to be updated, applying for virtual resource configuration update to the MANO according to the received target allocation virtual resource quantity, and reporting the virtual resource configuration update condition of the network slice to the CSMF through the NSMF.

And the NRSDF (Network resource scheduling decision function ) module is used for sending a mapping relation function of the service satisfaction degree of the sub-slice corresponding to the to-be-updated slice ID and the virtual resource to the NWAF module after receiving the virtual resource configuration update request, obtaining the currently allocated idle virtual resource quantity from the MANO, determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the to-be-updated slice according to the idle virtual resource quantity, the mapping relation function, the current network operation data and the virtual resource allocation weight, solving to obtain the target allocation virtual resource quantity, and feeding back the target allocation virtual resource quantity to the NSSMF.

And the NWAF (Network Data Analytics Function, network data analysis function) module is used for feeding back the mapping relation function of the sub-slice corresponding to the slice ID to be updated and the resource allocation weight of the tenant corresponding to the slice to be updated to the NRSDF.

MANO (Management and Orchestration, network management orchestration) module for feeding back the number of free virtual resources currently available.

In an alternative embodiment, the NSSMF module is further configured to apply for a virtual resource configuration update to a MANO according to the received target allocated virtual resource amount, and report the virtual resource configuration update of the network slice to the CSMF via the NSMF.

In a specific embodiment, a flowchart of interaction between various modules in the network slice virtual resource system of the present invention to complete an update of a network slice virtual resource may be illustrated with reference to fig. 2. FIG. 2 illustrates an interactive flow diagram for network slice virtual resource update in one embodiment.

In an optional embodiment, the NSMF module is further configured to obtain a service satisfaction of a tenant, determine whether the service satisfaction of the tenant is lower than a satisfaction threshold, and send a virtual resource configuration update request to the NSSMF module when the service satisfaction of the tenant is lower than the satisfaction threshold.

In an optional embodiment, the CSMF module is further configured to send the virtual resource configuration update condition of the network slice to a user corresponding to the network slice.

In a specific embodiment, the NRSDF module for virtual resource may be used as a new functional module to be combined with the NSWAF, or may be set separately.

Step 140: and acquiring the number of idle virtual resources, and determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight.

That is, in order to make the sum of the service satisfaction of all the slices needing to be updated in the whole network reach the maximum after the scheduling within the allowable range of the total amount of the existing schedulable idle virtual resources, that is, the average satisfaction level of the whole system is highest.

The following is an illustration of a target integer programming problem, which in one particular embodiment may be represented as follows:

x _i ≥0

x _i is an integer

Wherein M is a slice set for simultaneously applying for virtual resource update, X is the number of idle virtual resources (such as the number of virtual machines), and X is the number of virtual machines _i The number of virtual resources (such as the number of virtual machines used by the ith slice) of the corresponding type of the ith slice under the granularity of virtual resource allocation is also a solution to be obtained.

In particular, it can be seen from the above formula that when only one slice needs to be updated, there is an optimization target ω·f _i (x _i ,n _i ,t _i ,v _i ) The target integer programming problem is converted into a solution problem that maximizes the service MOS for a single slice.

In addition, it is particularly required to explain that the virtual resource allocation weights are different in grades or weights corresponding to different tenants, and when virtual resource scheduling is performed, users with larger virtual resource allocation weights generally need to be prioritized, that is, users with larger weights have a greater influence on the sum of service satisfaction degrees of the entire system in the service satisfaction degrees corresponding to the allocated virtual resources.

It should be specifically noted that, the first virtual resource allocation granularity may include the number of virtual machines, and the determining the optimal solution of the target integer programming problem may further include steps 1301-1304 shown in fig. 3. FIG. 3 illustrates a flow diagram for determining target allocation virtual resource data in one embodiment.

Step 1301: traversing the combination that each slice to be updated does not exceed the number of the idle virtual machines as an optional resource configuration combination.

Firstly, consider that in practical application, for each slice to be updated, the current state of the slice to be updated is n _i ,t _i ,v _i Parameters equal to the network operating state are known (and are integers), and the mapping relation f in the target integer programming problem is as described above _i Also known models derived by training, the integer programming problem can be solved by enumeration, i.e. by traversing each slice within the boundary conditionsX of (2) _i Combining to obtain the optimal solution.

Step 1302: and determining the sum of the service satisfaction degrees of all the slices to be updated corresponding to each selectable resource configuration combination according to the current network operation data, the mapping relation function and the virtual resource allocation weight corresponding to each slice to be updated.

Step 1303: and acquiring the optional resource configuration combination which enables the sum of the service satisfaction degree of the slice to be updated to be maximum as the target allocation virtual resource data.

Step 1304: and under the condition that the sum of service satisfaction corresponding to two or more selectable resource configuration combinations is the same, acquiring the selectable resource configuration combination which minimizes the sum of the numbers of the distributed virtual machines corresponding to all the slices to be updated as the target distributed virtual resource data.

The final objective of the present invention is to maximize the sum of the service satisfaction of the whole system, so that when the sum of the service satisfaction of several different xi-valued combinations is the same, the total number of virtual machines required for maximizing the sum of the service satisfactionThe minimum angle determines the optimal solution.

In addition, it is considered that in an alternative embodiment, besides the above-mentioned transverse scheduling manner of taking the number of virtual machines as the granularity of virtual resource allocation (i.e. the unit of resource allocation scheduling), there may be some resource management and scheduling capability of network slices that are stronger, so that scheduling on a smaller resource granularity may be implemented, for example, in addition to updating and allocating the number of virtual machines, for each virtual machine, the number of virtual CPUs, the virtual memory capacity and the network bandwidth used by the virtual machines may be further allocated for scheduling, i.e. the vertical scheduling manner of Scale up/down.

Thus, in an alternative embodiment, the virtual resource allocation granularity may further include a virtual CPU number, a virtual memory capacity, and a network bandwidth, and the determining a configuration scheme may further include steps 1311-1315 shown in fig. 4. FIG. 4 illustrates a flow chart of determining target allocation virtual resource data in another embodiment.

Step 1311: traversing the combination of the number of the idle virtual CPUs, the idle virtual memory capacity and the idle network bandwidth which are not exceeded by each slice to be updated as an optional resource configuration combination.

Step 1312: and determining the sum of the service satisfaction degrees of all the slices to be updated corresponding to each selectable resource configuration combination according to the current network operation data, the mapping relation function and the virtual resource allocation weight corresponding to each slice to be updated.

Step 1313: and acquiring an optional resource configuration combination which enables the sum of the service satisfaction degree of all the slices to be updated to be maximum as the target allocation virtual resource data.

Step 1314: and under the condition that the sum of service satisfaction degrees corresponding to two or more than two optional resource allocation combinations is the same, respectively acquiring the number of the virtual CPUs, the virtual memory capacity and the resource unit price corresponding to the network bandwidth, and respectively determining the resource total price of the optional resource allocation combinations with the same sum of the service satisfaction degrees according to the resource unit price.

Step 1315: and acquiring the optional virtual resource configuration combination with the minimum total price of resources as the target allocation virtual resource data.

Similar to the operation in step 1304, in the case where there are two or more selectable resource configuration combinations that correspond to the same sum of service satisfaction, the optimal solution may be further determined with the goal that the total price of resources corresponding to the virtual CPU number, the virtual memory capacity, and the network bandwidth is the lowest.

The total price of the resource may be a price of a resource per unit of economy, or may be another value reflecting the degree of scarcity and the degree of importance. In practical applications, compared with the virtual memory capacity, the scarcity degree of a unit of virtual CPU and the market trading value of the virtual CPU are higher than that of the virtual memory capacity, and the corresponding resource unit price is higher.

Step 150: and carrying out configuration updating on the at least one slice to be updated according to the target allocation virtual resource data.

The specific configuration process may be that the NRSDF (network resource scheduling decision) module based on the present invention sends a resource request of a corresponding configuration scheme to the MANO (network management orchestration) module, and the network management orchestration module performs configuration update according to the allocated virtual resource information.

Finally, it should be noted that the resource scheduling method of the present invention is not only suitable for scheduling virtual type resources, but also can be calculated and implemented on the allocation of other types of resources used. For example, the allocation of entity resources corresponding to different users, etc., in a specific embodiment, the data acquisition and processing process can be adaptively adjusted, and the mapping relation function and the solution of the target integer programming problem of the invention are not changed.

Fig. 5 shows a schematic structural diagram of an embodiment of the network slice virtual resource scheduling apparatus of the present invention. As shown in fig. 5, the apparatus 200 includes: a request receiving module 210, a data acquisition module 220, a function determination module 230, a scheme determination module 240, a configuration update module 250.

The request receiving module 210 is configured to receive a virtual resource configuration update request, and determine at least one slice to be updated according to the virtual resource configuration update request;

a data obtaining module 220, configured to obtain current network operation data, virtual resource allocation weights and number of idle virtual resources of the at least one slice to be updated;

the function determining module 230 is configured to determine mapping relation functions corresponding to the at least one slice to be updated, where the mapping relation functions are used to characterize a functional relation among current network operation data of the slice to be updated, slice allocation virtual resource data and service satisfaction, and the mapping relation functions are obtained by inputting preset regression model training according to historical slice network operation data, historical slice allocation virtual resource data and corresponding historical service satisfaction;

the scheme determining module 240 is configured to obtain the number of idle virtual resources, and determine target allocation virtual resource data that maximizes a sum of service satisfaction of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data, and the virtual resource allocation weight;

And the configuration updating module 250 is configured to perform configuration updating on the at least one slice to be updated according to the target allocation virtual resource data.

In an alternative manner, the scheme determination module 230 is further configured to:

traversing the combinations of which the number of the to-be-updated slices does not exceed that of the idle virtual machines as selectable resource configuration combinations;

In another alternative embodiment, the scheme determination module 230 is further configured to:

traversing the combination of the number of the idle virtual CPUs, the idle virtual memory capacity and the idle network bandwidth which are not exceeded by each slice to be updated as an optional resource configuration combination;

In an alternative manner, the invention further comprises a network slice virtual resource scheduling system, which comprises the following modules:

In an optional embodiment, the NSMF module is further configured to obtain a service satisfaction of the tenant, determine whether the service satisfaction of the tenant is lower than a satisfaction threshold, and send a virtual resource configuration update request to the NSSMF module when the service satisfaction of the tenant is lower than the satisfaction threshold.

In an alternative embodiment, the CSMF module is further configured to send the virtual resource configuration update condition of the network slice to the user corresponding to the network slice.

Fig. 6 shows a schematic structural diagram of an embodiment of the network slice virtual resource scheduling device of the present invention, and the specific embodiment of the present invention does not limit the specific implementation of the network slice virtual resource scheduling device.

As shown in fig. 6, the network slice virtual resource scheduling apparatus may include: a processor (processor) 302, a communication interface (Communications Interface) 304, a memory (memory) 306, and a communication bus 308.

Wherein: processor 302, communication interface 304, and memory 306 perform communication with each other via communication bus 408. A communication interface 304 for communicating with network elements of other devices, such as clients or other servers. The processor 302 is configured to execute the program 310, and may specifically perform the relevant steps in the embodiment of the network slice virtual resource scheduling method described above.

In particular, program 310 may include program code comprising computer-executable instructions.

The processor 302 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included in the network slice virtual resource scheduling device may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.

Memory 306 for storing programs 310. Memory 406 may comprise high-speed RAM memory or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.

Program 310 may be specifically invoked by processor 302 to cause the network slice virtual resource scheduling device to:

acquiring the number of idle virtual resources, and determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight;

In an alternative manner, program 310 may be specifically invoked by processor 302 to cause the network slice virtual resource scheduling device to:

The invention provides a structural schematic diagram of an embodiment of network slice virtual resource scheduling equipment. The apparatus includes: one or more processors and a communication interface;

the processor is configured to perform the steps in the network slice virtual resource method embodiment described above.

Embodiments of the present invention provide a computer readable storage medium storing at least one executable instruction that, when executed on a network slice virtual resource device/apparatus, causes the network slice virtual resource device/apparatus to perform the network slice virtual resource method in any of the above method embodiments.

The executable instructions may be specifically operable to cause a network slice virtual resource device/apparatus to:

In an alternative manner, the executable instructions may be specifically operable to cause a network slice virtual resource device/apparatus to:

The embodiment of the invention provides a network slice virtual resource scheduling device which is used for executing the network slice virtual resource scheduling method.

The embodiment of the invention provides a computer program which can be called by a processor to enable network slice virtual resource scheduling equipment to execute the network slice virtual resource scheduling method in any of the method embodiments.

An embodiment of the present invention provides a computer program product, where the computer program product includes a computer program stored on a computer readable storage medium, where the computer program includes program instructions, when the program instructions are executed on a computer, cause the computer to perform the network slice virtual resource scheduling method in any of the above method embodiments.

The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general-purpose systems may also be used with the teachings herein. The required structure for a construction of such a system is apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the invention as described herein, and any references to specific languages are provided for disclosure of enablement and best mode of the invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the above description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed invention requires more features than are expressly recited in each claim.

Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component, and they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specifically stated.

Claims

1. A network slice virtual resource scheduling method, the method comprising:

acquiring the number of idle virtual resources, and determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight; wherein the number of free virtual resources includes a number of free virtual machines, the determining the target allocation virtual resource data that maximizes a sum of service satisfaction of the slice to be updated further includes:

under the condition that the sum of service satisfaction corresponding to two or more selectable resource configuration combinations is the same, acquiring the selectable resource configuration combination with the minimum sum of the number of the distributed virtual machines corresponding to all the slices to be updated as the target distributed virtual resource data;

2. The method for scheduling virtual resources of a network slice according to claim 1, wherein the current network operation data at least includes a number of load users, a processing traffic volume, a coverage area, and a network KPI of the current slice, and the slice allocation virtual resource data at least includes a number of virtual machines, a number of virtual CPUs, a virtual memory capacity, and a network bandwidth corresponding to the slice.

3. The method for scheduling virtual resources of a network slice according to claim 1, wherein the number of idle virtual resources further includes an idle virtual CPU number, an idle virtual memory capacity, and an idle network bandwidth, and the determining the target allocation virtual resource data that maximizes a sum of service satisfaction degrees of the slice to be updated further includes:

4. A network slice virtual resource management system, the network slice virtual resource management system comprising:

the CSMF module is used for receiving a network slice update request of a tenant, notifying the NSMF module that the network slice of the tenant needs to be configured and updated, and notifying the network slice ID; the CSMF module is a communication service management module;

the NSMF module is used for evaluating and decomposing a received slice update request according to slice network data, determining whether the network slice needs virtual resource configuration update, determining the network slice ID as a slice ID to be updated when the network slice needs virtual resource configuration update, and sending the virtual resource configuration update request of the slice to be updated to the NSSMF module; the NSMF module is a slice management function management module; the NSSMF module is a sub-slice management function module;

the NSSMF module is used for sending a virtual resource configuration update request of the slice to be updated to the NRSDF module, wherein the virtual resource configuration update request of the slice to be updated carries the ID of the slice to be updated, applying for virtual resource configuration update to the MANO according to the received target allocation virtual resource quantity, and reporting the virtual resource configuration update condition of the network slice to the CSMF through the NSMF; the NRSDF module is a network resource scheduling decision function module; the MANO is a network management scheduling module;

The NRSDF module is used for sending a mapping relation function of the service satisfaction degree of the sub-slice corresponding to the slice ID to be updated and the virtual resource to the NWAF module after receiving the virtual resource configuration update request, obtaining the number of idle virtual resources which can be allocated currently to the MANO, determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data of the slice ID to be updated and the virtual resource allocation weight, solving to obtain the number of target allocation virtual resources, and feeding back the number of target allocation virtual resources to the NSSMF; wherein the number of free virtual resources includes a number of free virtual machines, the determining the target allocation virtual resource data that maximizes a sum of service satisfaction of the slice to be updated further includes: traversing the combinations of which the number of the to-be-updated slices does not exceed that of the idle virtual machines as selectable resource configuration combinations; determining the sum of service satisfaction of all the slices to be updated corresponding to each selectable resource configuration combination according to the current network operation data, the mapping relation function and the virtual resource allocation weight corresponding to each slice to be updated; acquiring the optional resource configuration combination which enables the sum of the service satisfaction degree of the slice to be updated to be maximum as the target allocation virtual resource data; under the condition that the sum of service satisfaction corresponding to two or more selectable resource configuration combinations is the same, acquiring the selectable resource configuration combination with the minimum sum of the number of the distributed virtual machines corresponding to all the slices to be updated as the target distributed virtual resource data; the NWAF module is a network data analysis function module;

5. The system of claim 4, wherein the NSMF module is further configured to

And acquiring the service satisfaction degree of each tenant in the target network, determining whether a tenant with the service satisfaction degree smaller than a preset satisfaction degree threshold exists, acquiring a slice identifier corresponding to the tenant with the service satisfaction degree smaller than the preset satisfaction degree threshold, and sending the slice identifier to an NSSMF module as an identifier to be updated.

6. The system of claim 4, wherein the CSMF module is further configured to send a virtual resource configuration update instance for the network slice to a user corresponding to the network slice.

7. A network slice virtual resource scheduling apparatus, the apparatus comprising:

the data acquisition module is used for acquiring the current network operation data of the at least one slice to be updated, the virtual resource allocation weight and the number of idle virtual resources;

The function determining module is used for determining mapping relation functions corresponding to the at least one slice to be updated respectively, wherein the mapping relation functions are used for representing the functional relation among the current network operation data of the slice to be updated, the slice allocation virtual resource data and the service satisfaction, and the mapping relation functions are obtained by inputting preset regression model training according to the historical slice network operation data, the historical slice allocation virtual resource data and the corresponding historical service satisfaction;

the scheme determining module is used for obtaining the number of idle virtual resources, and determining target allocation virtual resource data which maximizes the sum of the service satisfaction degree of the slice to be updated according to the number of idle virtual resources, the mapping relation function, the current network operation data and the virtual resource allocation weight; wherein the number of free virtual resources includes a number of free virtual machines, the determining the target allocation virtual resource data that maximizes a sum of service satisfaction of the slice to be updated further includes: traversing the combinations of which the number of the to-be-updated slices does not exceed that of the idle virtual machines as selectable resource configuration combinations; determining the sum of service satisfaction of all the slices to be updated corresponding to each selectable resource configuration combination according to the current network operation data, the mapping relation function and the virtual resource allocation weight corresponding to each slice to be updated; acquiring the optional resource configuration combination which enables the sum of the service satisfaction degree of the slice to be updated to be maximum as the target allocation virtual resource data; under the condition that the sum of service satisfaction corresponding to two or more selectable resource configuration combinations is the same, acquiring the selectable resource configuration combination with the minimum sum of the number of the distributed virtual machines corresponding to all the slices to be updated as the target distributed virtual resource data;

8. A network slice virtual resource scheduling apparatus, comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;

the memory is configured to store at least one executable instruction that causes the processor to perform the operations of the network slice virtual resource scheduling method of any one of claims 1-3.

9. A computer readable storage medium, wherein at least one executable instruction is stored in the storage medium, which when run on a network slice virtual resource scheduling device, causes the network slice virtual resource scheduling device to perform the operations of the network slice virtual resource scheduling method according to any one of claims 1-3.