CN110086675B

CN110086675B - Service chain construction method, device and computer readable storage medium

Info

Publication number: CN110086675B
Application number: CN201910374659.6A
Authority: CN
Inventors: 蔡君; 黄忠玮; 罗建桢; 魏文国; 刘燕
Original assignee: ZTE Corp; Guangdong Polytechnic Normal University
Current assignee: ZTE Corp; Guangdong Polytechnic Normal University
Priority date: 2019-05-05
Filing date: 2019-05-05
Publication date: 2022-03-11
Anticipated expiration: 2039-05-05
Also published as: CN110086675A

Abstract

The invention discloses a method, equipment and a computer readable storage medium for constructing a service chain, wherein the method comprises the following steps: determining a first VNF and a second VNF which respectively have serial function requirements and parallel function requirements in a virtual network function VNF corresponding to each NF according to the operation type of each preset network function NF; performing parallelization processing on each second VNF to generate a third VNF, and performing serialization processing on the first VNF and the third VNF according to a preset sequence relation to generate a service chain; and determining a physical node corresponding to the service chain according to the resource requirement of each VNF, deploying the service chain to the physical node, and completing construction of the service chain. The multiple VNFs subjected to parallelization processing in the service chain constructed by the scheme can process data at the same time, so that the processing time is saved, and the requirement of time-sensitive application for obtaining the processing result in real time is met.

Description

Service chain construction method, device and computer readable storage medium

Technical Field

The present invention relates to the field of internet technologies, and in particular, to a method and an apparatus for constructing a service chain, and a computer-readable storage medium.

Background

With the rapid development of communication network technology, more and more service chains are provided for time-sensitive applications of the industrial internet. The service chain is serially constructed by a plurality of virtual network function VNFs, and according to the arrangement sequence of each VNF in the service chain, after the data is processed by the VNF located before the arrangement sequence, the data is transmitted to the VNF located after the arrangement sequence, and the data is processed until each VNF in the service chain processes the data, and then the data processing process is completed.

However, in the method of serially constructing the service chain by using the VNFs, when the number of VNFs is large, the constructed service chain is also long, and thus the time length for data processing is too long, that is, the time delay for data processing through the whole service chain is too long, so that the requirement of obtaining the processing result in real time for the time-sensitive application cannot be met.

Disclosure of Invention

The invention mainly aims to provide a method and equipment for constructing a service chain and a computer readable storage medium, and aims to solve the problem that the delay of processing data is too long due to the service chain constructed in a serial mode in the prior art.

In order to achieve the above object, the present invention provides a method for constructing a service chain, including the following steps:

determining a first VNF and a second VNF which respectively have serial function requirements and parallel function requirements in a virtual network function VNF corresponding to each NF according to the operation type of each preset network function NF;

performing parallelization processing on each second VNF to generate a third VNF, and performing serialization processing on the first VNF and the third VNF according to a preset sequence relation to generate a service chain;

and determining a physical node corresponding to the service chain according to the resource requirement of each VNF, deploying the service chain to the physical node, and completing construction of the service chain.

Preferably, the step of determining a physical node corresponding to the service chain according to the resource requirement of each VNF includes:

reading the resource capacity value of each preset physical node, and judging whether the number of the VNFs is smaller than a preset number;

if the quantity is smaller than the preset quantity, reading the function resource requirement values of the VNFs in the service chain, and integrating the function resource requirement values to generate a total function resource requirement value of the service chain;

comparing each resource capacity value with the total function resource requirement value, and determining a function resource capacity value which is greater than the total function resource requirement value in each resource capacity value;

reading a function calculation capacity value corresponding to each function resource capacity value, determining a maximum function calculation capacity value in each function calculation capacity value, and determining a preset physical node corresponding to the maximum function calculation capacity value as a physical node corresponding to the service chain.

Preferably, the step of determining whether the number of VNFs is less than a preset number includes:

if the number of the VNFs is not less than the preset number, reading a first resource requirement value of each first VNF and a second resource requirement value of each second VNF;

determining a first physical node corresponding to a first VNF in the service chain according to each first resource demand value;

determining a third physical node corresponding to a third VNF in the service chain according to each second resource demand value;

and merging the first physical node and the third physical node to form a physical node corresponding to the service chain.

Preferably, the step of determining a first physical node corresponding to a first VNF in the service chain according to each first resource requirement value includes:

reading each first resource requirement value one by one, and executing the following steps aiming at each first resource requirement value:

comparing the first resource demand value with each resource capacity value, and determining a first resource capacity value which is greater than the first resource demand value in each resource capacity value;

forming preset physical nodes corresponding to the first resource capacity values into a first physical node set, and reading a first node distance between every two adjacent preset physical nodes in the first physical node set;

reading a first calculation capacity value of each preset physical node in the first physical node set, and generating a first capacity score and a first distance score of each preset physical node in the first physical node set according to each first calculation capacity value and each first node distance;

and determining a preset physical node with the highest score in the first physical node set according to each first capability score and each first distance score, and determining the preset physical node with the highest score in the first physical node set as a first physical node corresponding to a first VNF in the service chain.

Preferably, the step of generating a first capability score and a first distance score of each preset physical node in the first physical node set according to each first calculation capability value and each first node distance includes:

according to the sequence of the first computing capacity values from large to small, capacity sorting is carried out on all preset physical nodes in the first physical node set, and according to a first preset computing rule, a first capacity score of each preset physical node in the first physical node set after capacity sorting is determined;

and according to the sequence of the distances of the first nodes from small to large, distance sorting is carried out on all preset physical nodes in the first physical node set, and according to a second preset calculation rule, the first distance scores of all preset physical nodes in the first physical node set after distance sorting are determined.

Preferably, the step of determining a third physical node corresponding to a third VNF in the service chain according to each of the second resource requirement values includes:

integrating the second resource demand values to generate third resource demand values, determining the maximum capacity value of the resource capacity values, and judging whether the maximum capacity value is greater than the third resource demand value;

if the resource capacity value is larger than the third resource demand value, comparing the third resource demand value with each resource capacity value, and determining a third resource capacity value which is larger than the third resource demand value in each resource capacity value;

forming preset physical nodes corresponding to each third resource capacity value into a third physical node set, and reading a third node distance between every two adjacent preset physical nodes in the third physical node set;

reading a third calculation capacity value of each preset physical node in the third physical node set, and generating a third capacity score and a third distance score of each preset physical node in the third physical node set according to each third calculation capacity value and each third node distance;

and determining the preset physical node with the highest score in the third physical node set according to each third capability score and each third distance score, and determining the preset physical node with the highest score in the third physical node set as the third physical node corresponding to the third VNF in the service chain.

Preferably, the step of determining whether the maximum capacity value is greater than the third resource requirement value is followed by:

if the maximum capacity value is not larger than the third resource requirement value, generating a second physical node set according to each second resource requirement value and each resource capacity value;

calculating a target second resource requirement value with the minimum difference value between the target second resource requirement value and each resource capacity value according to a preset formula, and forming a VNF instance of a second VNF corresponding to each target second resource requirement value into an instance set;

reading preset physical nodes in the second physical node set as target physical nodes one by one, and VNF instances in the instance set as target instances, and establishing a matching relationship between a second VNF corresponding to the target instance and the target physical nodes when a second resource capacity value of the target physical nodes is greater than a second resource required value corresponding to the target instances;

and determining a third physical node corresponding to a third VNF in the service chain according to the matching relation.

Preferably, the step of generating a second physical node set according to each second resource requirement value and each resource capacity value includes:

comparing the second resource demand values, and determining a maximum demand value and a minimum demand value in the second resource demand values;

screening the resource capacity values, and determining a second resource capacity value with a numerical value between the minimum demand value and the maximum demand value;

and reading second computing capacity values corresponding to the second resource capacity values, and sequencing preset physical nodes corresponding to the second resource capacity values according to the second computing capacity values to generate a second physical node set.

In addition, to achieve the above object, the present invention further provides a service chain construction device, including: a memory, a processor, a communication bus, and a service chain builder stored on the memory;

the communication bus is used for realizing connection communication between the processor and the memory;

the processor is used for executing the service chain building program to realize the following steps:

Further, to achieve the above object, the present invention also provides a computer-readable storage medium storing one or more programs, the one or more programs being executable by one or more processors for:

In the method for constructing the service chain according to this embodiment, first, according to the operation type of each preset network function NF, a first VNF and a second VNF, which respectively have a serial function requirement and a parallel function requirement, in a virtual network function VNF corresponding to each NF are determined; performing parallelization processing on each second VNF to generate a third VNF, and performing serialization processing on the first VNF and the third VNF according to a preset sequence relation to generate a service chain; and then, according to the resource requirement of each VNF, determining a physical node corresponding to the service chain, deploying the service chain into the physical node, and completing construction of the service chain. In the scheme, the operation type corresponding to each preset network function NF determines the serial function requirement and the parallel function requirement of the VNF corresponding to each NF; performing parallelization processing on the VNF with the parallel function requirement, and performing serialization processing on the VNF subjected to the parallelization processing and the VNF with the serial function requirement to generate a service chain; and the generated service chain is deployed into the physical node so as to process the data transmitted into the service chain according to the resources of the physical node. Because a plurality of VNFs subjected to parallelization processing in the processing process can process data at the same time, the situation that the data are transmitted to another VNF for processing after the processing of one VNF is completed is avoided, the processing time is saved, and the requirement of time-sensitive application for obtaining the processing result in real time is met.

Drawings

FIG. 1 is a schematic flow chart diagram of a first embodiment of a method for constructing a service chain according to the present invention;

FIG. 2 is a schematic diagram of an apparatus architecture of a hardware operating environment to which a method according to an embodiment of the present invention relates;

fig. 3 is a schematic diagram of a service chain generation process of the service chain construction method of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The invention provides a method for constructing a service chain.

Referring to fig. 1, fig. 1 is a flowchart illustrating a method for constructing a service chain according to a first embodiment of the present invention. In this embodiment, the method for constructing the service chain includes:

step S10, determining, according to the operation type of each preset network function NF, a first VNF and a second VNF (identity determination) that respectively have a serial function requirement and a parallel function requirement in the virtual network function VNFs corresponding to each NF;

the method for constructing the service chain is applied to the server and is suitable for constructing the service function chain facing the time-sensitive application, namely the service chain, through the server. The service chain is constructed by a plurality of VNFs (virtual Network functions), the VNFs and the NFs (Network functions) have corresponding relations, one VNF corresponds to one NF, and certain operation of data processing is achieved. At the beginning of building the service chain, developers preset a plurality of NFs and operation types of the NFs, such as read operation, write operation, etc., according to operations required by the service chain. The NF with different operation types has different dependency relationships, and the dependency relationship represents whether the previous operation is required to be depended on when the data is operated; if so, it is described that the previous and subsequent VNFs are in a serial relationship, and the operation of the subsequent VNF on data needs to depend on the operation of the previous VNF; if not, the previous and subsequent VNFs are in a parallel relationship, and the operation of the subsequent VNF on the data does not need to depend on the operation of the previous VNF. And uploading each set NF serving as a preset network function NF to a server by a developer, reading an identifier representing the operation type of each NF by the server when a service chain is constructed, and screening a first VNF with a serial function requirement and a second VNF with a parallel function requirement in VNFs corresponding to each NF according to the difference of the identifiers.

Step S20, performing parallelization processing on each second VNF to generate a third VNF, and performing serialization processing on the first VNF and the third VNF according to a preset sequential relationship to generate a service chain;

furthermore, because each second VNF does not have a dependency relationship, data can be processed at the same time; and performing parallel processing on each second VNF, and performing parallel group chain operation to form a third VNF. The first VNF cannot process data at the same time due to the fact that the first VNF has a dependency relationship, and the data need to be processed according to a sequence which is preset by a developer and uploaded to a server according to a preset sequence relationship; and after the server generates a third VNF in the parallel group chain, reading the preset sequence relation, serializing each first VNF and each third VNF according to the preset sequence relation, and performing serial group chain. Wherein, the serial group chain between every first VNF is directly serialized; and for the serial group chain between the first VNF and the second VNF, adding a copy node and a merge node according to the precedence relationship between the first VNF and the third VNF. If the first VNF is located before the third VNF, a replication node is added between the first VNF and the third VNF to replicate data processed by the first VNF according to the number of second VNFs forming the third VNF, the number of replicated data copies is consistent with the number of second VNFs, and each replicated data is conveniently transmitted to each second VNF to be processed simultaneously. If the first VNF is located after the third VNF, a merge node is added between the first VNF and the third VNF to merge data processed by each second VNF in the third VNF, so that the merged data is transmitted to the subsequent first VNF for processing.

And after the group chain of each first VNF and each third VNF is completed, generating a service chain capable of completely processing the data. For example, in an embodiment, VNFs corresponding to NFs include VNF1, VNF2, VNF3, and VNF4, and referring to fig. 3, VNF1 and VNF2, VNF3 and VNF4 have a dependency relationship therebetween, and are not capable of performing parallelization, and VNF2 and VNF3 have no dependency relationship therebetween, and are capable of performing parallelization; therefore, when a service chain is constructed, parallelization processing is performed between the VNF2 and the VNF3 to generate a third VNF, a merge node is added between the VNF1 and the VNF2 and between the VNF3, and a merge node is added between the VNF2, the VNF3 and the VNF4 to form the service chain. It should be noted that, in fig. 3, No pall (No parallelism) represents that parallelization cannot be performed, pall (parallelizable) represents that parallelization can be performed, and PNs (parallelizable VNFs) represents the number of VNFs to be parallelized, and in the case where PNs is equal to 1, this represents a special case where VNFs required by a single serial function are handled as parallel.

Step S30, determining a physical node corresponding to the service chain according to the resource requirement of each VNF, and deploying the service chain to the physical node to complete the construction of the service chain.

Understandably, the service chain needs to be mapped into the physical node, and the data can be processed only by depending on the resources in the physical node; the required resources of each VNF have differences according to the difference of the operation functions supported by the VNF, and the resources provided by different physical nodes are also different; therefore, when the service chain is deployed in the physical node, the physical node corresponding to the service chain is determined according to the resource requirement of each VNF in the service chain. The resource requirement of each VNF represents the amount of resources required to implement the operation function of each VNF in the service chain, and the determined physical node corresponding to the service chain may be the amount of resources required to satisfy the operation function of each VNF. And then deploying the service chain to the determined physical node, namely processing the transmitted data by depending on the resources of the physical node, thereby completing the construction operation of the service chain.

Further, in another embodiment of the method for constructing a service chain of the present invention, the step of determining a physical node corresponding to the service chain according to the resource requirement of each VNF includes:

step S31, reading the resource capacity value of each preset physical node, and judging whether the quantity of the VNFs is less than a preset quantity;

step S32, if the quantity is less than the preset quantity, reading the function resource requirement value of each VNF in the service chain, and integrating the function resource requirement values to generate a total function resource requirement value of the service chain;

step S33, comparing each resource capacity value with the total function resource requirement value, and determining the function resource capacity value which is larger than the total function resource requirement value in each resource capacity value;

step S34, reading function calculation capability values corresponding to the capacity values of the function resources, determining a maximum function calculation capability value among the function calculation capability values, and determining a preset physical node corresponding to the maximum function calculation capability value as a physical node corresponding to the service chain.

Understandably, the larger the number of physical nodes is, the more time is consumed for the transmission of the processed data among the physical nodes, which results in the increase of the processing time; therefore, in order to further reduce the data processing delay, when the service chain is deployed on the physical node, the service chain is deployed in the same physical node as much as possible; therefore, when the physical node corresponding to the service chain is determined, the matching is performed according to the matching degree between the resources required by each VNF and the resources provided by each physical node. Specifically, a plurality of preset physical nodes which can be used for deploying the VNFs are preset, and the VNF number representing less required resources is preset according to historical experience and is used as the preset number. If the preset number is set to be 3, representing that when the number of the VNFs in the service chain is less than 3, resources required by each VNF in the service chain are less, and the VNFs can be deployed in the same physical node; when the number of VNFs in the service chain is greater than or equal to 3, it indicates that resources required by each VNF in the service chain are large and cannot be deployed in the same physical node.

Further, a resource capacity value of each preset physical node is read, where the resource capacity value is a resource that can be provided by the preset service node. And simultaneously reading the number of VNFs in the service chain, comparing the number with a preset number, judging whether the number is smaller than the preset number, and if so, judging that the service chain can be deployed in the same service node. At this time, the physical nodes which meet the resource requirements of each VNF in the service chain and have the advantage of computing power are searched, so that the deployed physical nodes are ensured to have excellent computing power while the service chain is deployed at the physical nodes which meet the resource requirements of the service chain, and the data in the service chain can be rapidly and effectively processed conveniently.

In this embodiment, the process of searching for the physical node is implemented by processing logic of the first algorithm, and before processing of the first algorithm, the VNF instance of the VNF in the service chain is characterized by fi, where fi is { f1... fn }, so that the service chain is formed by the VNF instance fi { f1... fn }Composition is carried out; simultaneously representing a set formed by the preset physical nodes by using Ns, wherein the resource type I of the preset physical nodes is { CPU, bandwidth and storage }; assume that fi has a class I resource requirement of

Presetting the resource capacity of a physical node n as

I.e. a resource capacity value of

The computing capacity of each preset physical node is Ci. Then, taking the resource required by each VNF in the service chain as a function resource required value for reading, taking the function resource required value and a resource capacity value as the input of a first algorithm, and determining a physical node corresponding to the service chain from preset physical nodes through the first algorithm, wherein the processing logic step of the first algorithm comprises the following steps:

(1) demand value for each function resource

Adding and integrating to generate a total function resource demand value of the service chain, namely the total function resource demand value

(2) Comparing the capacity value of each resource with the total function resource demand value one by one, determining the capacity value of the function resource which is greater than the total function resource demand value in the capacity values of each resource, namely searching for the capacity value which meets the requirement of the total function resource

A preset physical node of the condition, and forming each preset physical node satisfying the condition as a set N (N ═ { N1, N2 · · ni };

(3) reading the function computing capacity value corresponding to each function resource capacity value, namely reading the computing capacity Ci of each preset physical node in the set N (N ═ { N1, N2 · · ni }) as the function computing capacity value;

(4) and sequencing the function calculation capacity values, determining the maximum function calculation capacity value with the maximum value, and determining the preset physical node with the maximum function calculation capacity value as the physical node corresponding to the service chain.

The physical node determined by the processing logic of the first algorithm is the physical node which meets the resource requirements of each VNF in the service chain and has the advantage of computing power.

Understandably, the number of VNFs in the constructed service chain is still greater than the preset number, and at this time, processing needs to be performed according to other processing logic; specifically, the step of determining whether the number of VNFs is less than the preset number includes:

step S35, if the number of VNFs is not less than the preset number, reading a first resource requirement value of each first VNF and a second resource requirement value of each second VNF;

step S36, determining a first physical node corresponding to a first VNF in the service chain according to each first resource requirement value;

step S37, determining a third physical node corresponding to a third VNF in the service chain according to each of the second resource requirement values;

step S38, merging the first physical node and the third physical node to form a physical node corresponding to the service chain.

Further, when it is determined through comparison that the number of VNFs in the service chain is greater than the preset number and the service chain needs to be deployed to a plurality of different physical nodes, the first resource requirement value of the first VNF in the service chain is read, and the second resource requirement value of the second VNF in the service chain is read. The first resource requirement value represents the size of the resource required by each first VNF, so that the preset physical node meeting the requirement of each first VNF can be determined from the preset physical nodes according to the first resource requirement value, and the determined preset physical node is used as the first physical node corresponding to the first VNF in the service chain. And meanwhile, the second resource requirement value represents the size of the resource required by each second VNF, and the third VNF is composed of the second VNFs, so that the preset physical nodes meeting the requirements of each third VNF can be determined from the preset physical nodes according to the second resource requirement value, and the determined preset physical nodes are used as third physical nodes corresponding to the third VNFs in the service chain. And then, taking the determined first physical nodes and the third physical nodes as physical nodes corresponding to the service chain, and deploying each VNF in the service chain to the corresponding service node according to the corresponding relation between the VNF and the physical nodes to complete the deployment of the service chain.

Further, in another embodiment of the method for constructing a service chain of the present invention, the step of determining, according to each of the first resource requirement values, a first physical node corresponding to a first VNF in the service chain includes:

step S361, reading each of the first resource requirement values one by one, and executing the following steps for each of the first resource requirement values:

step S362, comparing the first resource requirement value with each of the resource capacity values, and determining a first resource capacity value greater than the first resource requirement value in each of the resource capacity values;

step S363, forming preset physical nodes corresponding to each first resource capacity value into a first physical node set, and reading a first node distance between each adjacent preset physical node in the first physical node set;

step S364, reading a first calculation ability value of each preset physical node in the first physical node set, and generating a first ability score and a first distance score of each preset physical node in the first physical node set according to each first calculation ability value and each first node distance;

further, the first VNF is a VNF in the service chain, which has a serial relationship with previous and subsequent VNFs, and when determining the corresponding first physical node, the first resource requirement values of the first VNFs in the service chain are read one by one and processed. Considering that a service chain divides delay in a data processing process into two parts, namely execution delay and transmission delay, in order to reduce the execution delay, a physical node meeting the condition of resources required by the first VNF needs to be found out, and a physical node with strong computing power (namely the number of cores of the CPU) needs to be selected for deployment; in order to reduce the link transmission delay, the proximity principle is considered when deploying the first VNF, i.e. the latter physical node is as close as possible to the former physical node. Specifically, in order to meet the resource requirement, the currently read first resource requirement value is compared with the resource capacity values of the preset physical nodes, and all first resource capacity values which are greater than the first resource requirement value in the resource capacity values are found; and then forming the preset physical nodes corresponding to the first resource capacity values into a first physical node set, and reading the first node distance between the preset physical nodes with adjacent relation in the first physical node set. And simultaneously reading the first calculation capacity value of each preset physical node in the first physical node set, and generating a first distance score and a first capacity score of each preset physical node according to the first node distance and the first calculation capacity value of each preset physical node. The step of generating a first capability score and a first distance score of each preset physical node in the first physical node set according to each first calculation capability value and each first node distance comprises the following steps:

step q1, according to the sequence of the first computing ability values from big to small, performing ability sorting on each preset physical node in the first physical node set, and according to a first preset computing rule, determining a first ability score of each preset physical node in the first physical node set after the ability sorting;

and q2, performing distance sorting on each preset physical node in the first physical node set according to the sequence of the distances of the first nodes from small to large, and determining the first distance score of each preset physical node in the first physical node set after distance sorting according to a second preset calculation rule.

Furthermore, capacity sequencing is carried out on each preset physical node according to the sequence of the first computing capacity value of each preset physical node in the first physical node set from large to small; because the calculation capacity changes from strong to weak according to the sequence from large to small, the preset physical nodes after capacity sequencing represent the processing capacity changes from strong to weak. Meanwhile, a first preset calculation rule is preset, for example, a scoring rule with a descending numerical value of 1 is set, the score of the first-ranked preset physical node is set as i, and the scores of subsequent preset physical nodes are respectively (i-1), (i-2) · (i-k), and the like. And setting the first capability score of each preset physical node in the first physical node set after capability sorting according to the first preset calculation rule.

In addition, distance sequencing is carried out on each preset physical node according to the sequence of the node distance of each preset physical node in the first physical node set from small to large; because the change of the distance from short to long is represented in the sequence from small to large, the preset physical nodes after distance sorting represent the change of the distance between the preset physical nodes from short to long. Meanwhile, a second preset calculation rule is preset, for example, a scoring rule of a descending numerical value 2 is set, the score of the first-ranked preset physical node is set as i, and the scores of the subsequent preset physical nodes are (i-2), (i-4) - (i- (k-2)), and the like. And setting the first distance scores of all preset physical nodes in the first physical node set after the distance sorting processing according to the second preset calculation rule. It should be noted that the same calculation rule may be set between the first preset calculation rule and the second preset calculation rule, which is not limited herein.

Step S365 of determining, according to each of the first capability scores and each of the first distance scores, a preset physical node with a highest score in the first physical node set, and determining the preset physical node with the highest score in the first physical node set as the first physical node corresponding to the first VNF in the service chain.

Further, adding the first capability score and the first distance score of each preset physical node in the first physical node set to generate a total score of each preset physical node; and comparing the total scores to determine the maximum value, wherein the preset physical node corresponding to the maximum value is the preset physical node with the highest score in the first physical node set. Since the preset physical node with the highest score balances the computation capability and the node distance, the preset physical node has smaller execution delay and transmission delay, and is determined as the first physical node corresponding to the first VNF in the service chain. Considering that there are many first VNFs in a service chain, each first VNF needs to determine a corresponding first physical node; after determining, for the currently read first resource requirement value, a first physical node corresponding to the first VNF having the first resource requirement value, and reading a next first resource requirement value for processing, it is necessary to remove the resource occupied in the first physical node that has been determined, and to process the remaining available resource as a resource of a new preset physical node. If for the currently read first resource requirement value W, its corresponding first VNF is W; if the first physical node corresponding to W is determined to be P and the resource in P is P, because the resource occupied by deploying W into P is W, the remaining available resource in P is (P-W), and thus (P-W) is taken as the resource of P, the next first resource requirement value is continuously read, and the first physical node corresponding to the first VNF from which the first resource requirement value is derived is determined.

It should be noted that, in this embodiment, the processing logic that determines the first physical node is processing logic different from the first algorithm, and may be characterized by a second algorithm; in particular, the first node distance of a physical node from a previous physical node is defined by di,

characterizing a first resource requirement value, Ci characterizing a first computing capability value,

representing the first resource capacity value, wherein the meaning represented by fi is the same as the algorithm one, and is not described herein again; the processing logic of algorithm two can be summarized as:

(1) reading a first resource demand value of fi corresponding to the first VNF

(2) Will be provided with

And

comparing, determining to satisfy

The preset physical nodes of the condition form a first set of physical nodes N ═ { N1, N2 … ni };

(3) reading a first node distance di between every two adjacent preset physical nodes and a first calculation capacity value Ci of each preset physical node;

(4) sorting the set N according to the increasing sequence of di to obtain a set N', and sorting the set N according to the decreasing sequence of Ci to obtain a set N ";

(5) scoring each ni in N 'according to a first preset calculation rule, and scoring each ni in N' according to a second preset calculation rule;

(6) and calculating the scores of ni in the N ═ N1, n2... ni } according to the score scores of ni in the N' and the score scores of ni in the N ″, and determining a preset physical node with the highest score as a first physical node corresponding to the first VNF.

Further, in another embodiment of the method for constructing a service chain, the step of determining a third physical node corresponding to a third VNF in the service chain according to each second resource requirement value includes:

step S371, integrating the second resource demand values to generate a third resource demand value, determining a maximum capacity value of the resource capacity values, and determining whether the maximum capacity value is greater than the third resource demand value;

step S372, if the value is larger than the third resource demand value, comparing the third resource demand value with each resource capacity value, and determining a third resource capacity value larger than the third resource demand value in each resource capacity value;

step S373, forming preset physical nodes corresponding to each of the third resource capacity values into a third physical node set, and reading a third node distance between each adjacent preset physical node in the third physical node set;

step S374, reading a third calculation ability value of each preset physical node in the third physical node set, and generating a third ability score and a third distance score of each preset physical node in the third physical node set according to each third calculation ability value and each third node distance;

step S375, according to each third capability score and each third distance score, determining a preset physical node with the highest score in the third physical node set, and determining the preset physical node with the highest score in the third physical node set as a third physical node corresponding to a third VNF in the service chain.

Further, the third VNF is constructed by the second VNF, and when determining the third physical node corresponding to the third VNF, the third physical node may be determined according to the second resource requirement value of the second VNF. Specifically, the second resource demand values are added and integrated, and the result obtained by the addition is the third resource demand value, which represents the size of the resource required by the third VNF as a whole. Considering that transmission delay exists between different preset physical nodes, when a third VNF formed by a second VNF capable of parallel processing is deployed to the same preset physical node, the transmission delay between different preset physical nodes can be avoided to the greatest extent; the resource capacity value of the preset physical node required to be deployed to the same preset physical node is greater than the third resource requirement value, that is, the resource which can be provided by the preset physical node meets the size of the resource required by the third VNF as a whole. Comparing the resource capacity values of the preset physical nodes to determine the maximum capacity value with the maximum value; it should be noted that, for the preset physical node determined as the first physical node, when comparing the resource capacity values, the resource occupied by the first VNF in the first physical node needs to be removed, and the remaining resource is used as the resource capacity value of the preset physical node determined as the first physical node, and then compared with the resource capacity values of other preset physical nodes, so as to determine the maximum capacity value therein. And comparing the third resource requirement value with the maximum capacity value to judge whether the maximum capacity value is larger than the third resource requirement value. If the resource capacity value is larger than the first resource capacity value, at least one preset physical node meeting the deployment requirement of the third VNF exists, the third resource capacity value larger than the third resource capacity value in each resource capacity value is determined, the preset physical node corresponding to each third resource capacity value can meet the requirement of the third VNF on the resource size, and therefore the corresponding physical node is formed into a third physical node set.

And then reading a third node distance between each preset physical node with adjacent relation in the third physical node set, simultaneously reading a third calculation capacity value of each preset physical node in the third physical node set, and generating a third distance score and a third capacity score of each preset physical node according to the third node distance and the third calculation capacity value of each preset physical node. And generating each third distance score and each third capability score according to the magnitude relationship between the distances of each third node and the magnitude relationship between the third calculation capability values, wherein the generation mode is similar to the first distance score and the first capability score, and is not repeated herein.

Similarly, after the third capability score and the third distance score of each preset physical node in the third physical node set are generated, the third capability score and the third distance score of the same preset physical node are added to generate a total score of each preset physical node; and comparing the total scores to determine the maximum value, wherein the preset physical node corresponding to the maximum value is the preset physical node with the highest score in the third physical node set. Since the preset physical node with the highest score balances the computation capability and the node distance, the preset physical node has smaller execution delay and transmission delay, and is determined as a third physical node corresponding to a third VNF in the service chain.

It should be noted that, in the embodiment, the processing logic for determining the third physical node is different from the first algorithm and the second algorithmThe processing logic of (1) can be characterized by an algorithm three; specifically, in algorithm three

Characterizing a third resource requirement value, di characterizing a third node distance, Ci characterizing a third computation capability value,

representing the third resource capacity value, wherein the meaning represented by fi is the same as the algorithm II, and is not described herein again; the processing logic of algorithm three can be summarized as:

(1) computing a total resource requirement of the third VNF for parallel processing, i.e. a third resource requirement value

(2) Mixing Dm with

Comparing, determining to satisfy

The preset physical nodes of the condition form a third physical node set N ═ { N1, N2 … ni };

(3) reading a third node distance di between every two adjacent preset physical nodes and a third calculation capacity value Ci of each preset physical node;

(6) and calculating the scores of ni in the N ═ N1, n2... ni } according to the score scores of ni in the N' and the score scores of ni in the N ″, and determining a preset physical node with the highest score as a third physical node corresponding to the third VNF.

Further, in another embodiment of the method for constructing a service chain of the present invention, the step of determining whether the maximum capacity value is greater than the third resource requirement value includes:

step S376, if the maximum capacity value is not greater than the third resource requirement value, generating a second physical node set according to each second resource requirement value and each resource capacity value;

further, when the maximum capacity value is determined to be not greater than the third resource requirement value, it indicates that the resources in the preset physical node with the most resources cannot meet the resource requirement of the third VNF, and the third VNF needs to be deployed in a different physical node. At this time, different second VNFs forming the third VNF have different resource requirements, the different resource requirements correspond to different computing requirements, and the larger the resource requirements are, the larger the corresponding computing requirements are; meanwhile, there is a difference in processing time of the preset physical node for each second VNF. If the processing time of a certain preset physical node in the deployed preset physical nodes is too long, the processing time of the whole service chain is influenced; second VNFs with different computing requirements are allocated to preset physical nodes with different processing capacities, so that similar processing time is ensured among the second VNFs, and time cooperation of parallel processing is realized; that is, the second VNF with a larger computational demand is mapped to the preset physical node with a strong processing capability, the second VNF with the second computational demand is mapped to the preset physical node with the second processing capability, and so on, the execution time of the single second VNF is reduced, and the overall execution delay is also reduced accordingly.

Understandably, the processing capacity of the preset physical nodes is represented by the computing capacity values, and in order to realize the coordination of parallel processing time, the computing capacity values between the preset physical nodes are determined while each preset physical node meets the resource requirement of the second VNF. At this time, each preset physical node satisfying the second VNF resource requirement may be determined according to a relationship between the second resource requirement value required by the second VNF and the resource capacity value of each preset physical node, and then each preset physical node is generated as a second physical node set, where each preset physical node in the second physical node set has a different calculation capacity value. Specifically, the step of generating the second physical node set according to the second resource requirement values and the resource capacity values includes:

step p1, comparing the second resource demand values, and determining the maximum value and the minimum value of the demand in the second resource demand values;

step p2, screening each resource capacity value, and determining a second resource capacity value with a numerical value between the minimum demand value and the maximum demand value;

and p3, reading second calculation capacity values corresponding to the second resource capacity values, and sorting preset physical nodes corresponding to the second resource capacity values according to the second calculation capacity values to generate a second physical node set.

And further comparing the second resource demand values, determining the maximum value and the minimum value, and taking the maximum value and the minimum value as the maximum demand value and the minimum demand value respectively. And then, screening the resource capacity value of each preset physical node, comparing each resource capacity value with the maximum capacity value and the minimum capacity value, and determining a second resource capacity value of the numerical value in a numerical value interval formed by the minimum demand value and the maximum demand value. Reading the calculation capacity value of the preset physical node corresponding to each second resource capacity value as a second calculation capacity value, and sequencing each preset physical node according to the magnitude relation of each second calculation capacity value to generate a second physical node set; the preset physical nodes with larger second computing power values in the set are arranged in front, and the preset physical nodes with smaller second computing power values are arranged behind.

Step S377, according to a preset formula, calculating a target second resource requirement value having a minimum difference with each of the resource capacity values, and forming VNF instances of a second VNF corresponding to each of the target second resource requirement values into an instance set;

step 378, reading preset physical nodes in the second physical node set as target physical nodes one by one, taking VNF instances in the instance set as target instances, and establishing a matching relationship between a second VNF corresponding to the target instance and the target physical nodes when a second resource capacity value of the target physical node is greater than a second resource required value corresponding to the target instance;

step S379, determining a third physical node corresponding to a third VNF in the service chain according to the matching relationship.

Further, to avoid the waste of resources, the second VNF with the highest matching degree with the resource of the preset physical node is usually deployed into the preset physical node, where the highest matching degree is represented by a difference between the second resource requirement value of the second VNF and the resource capacity value of the preset physical node. When the difference between the two is smaller, the matching degree is higher; and the larger the difference between the two is, the lower the matching degree is. In order to determine a second VNF with a high matching degree with each preset physical node, a preset formula is preset, and the resource capacity values and the second resource required values are transmitted to the preset formula one by one to be calculated to obtain difference results; and comparing the difference results to determine the difference result with the minimum value, and taking the second resource requirement value generating the difference result with the minimum value as the target second resource requirement value. Wherein the preset formula is shown as the following formula (1):

wherein, the Delta R is the result of the difference,

for each of the second resource requirement values,

for each resource capacity value.

After determining the target second resource requirement values, the VNF instances of the second VNF having the target second resource requirement values are formed into an instance set to characterize the second VNF having a high degree of matching with the preset physical nodes.

Understandably, the second physical node set represents each preset physical node with strong and weak computing capability value, and the example set represents a second VNF with high matching degree with each preset physical node; reading the preset physical nodes in the second physical node set one by one as target physical nodes according to the arrangement sequence of the preset physical nodes in the second physical node set for matching between the second VNF and the preset physical nodes; and simultaneously reading each VNF instance as a target instance one by one according to the arrangement sequence of each VNF instance in the instance set. Comparing the read second resource capacity value of the target physical node with a second resource requirement value of the requirement corresponding to the target instance, judging whether the second resource capacity value is greater than the second resource requirement value, and if so, establishing a matching relationship between a second VNF corresponding to the target instance and the target physical node; if not, it is determined that the target physical node does not meet the resource requirement of the second VNF corresponding to the target instance, and the predetermined physical node meeting the resource requirement of the second VNF corresponding to the target instance needs to be re-determined. After the matching relationship is established, a third physical node corresponding to a third VNF in the service chain can be determined according to the target physical node corresponding to the second VNF in the matching relationship.

It should be noted that, in this embodiment, the processing logic for determining the third physical node is processing logic different from the first algorithm, the second algorithm, and the third algorithm, and may be characterized by the fourth algorithm; specifically, in algorithm four

Characterizing a second resource requirement value, Ci characterizing a second computing capability value,

characterizing a second resource capacity value, cap_nCharacterizing a resource capacity value; in addition, fc (ns) is used to characterize the set of preferences of the second VNF for the preset physical node, i.e. the second set of physical nodes; characterizing a set of preferences, i.e. a set of instances, of a preset physical node for the second VNF with ns (fc); and characterizing the matching relation between the second VNF and the preset physical node by match { fi, ni }. Specifically, the processing logic of algorithm four can be summarized into three parts: 1. second physical nodeAnd 2, constructing a set, 2, constructing an example set, and 3, constructing a matching relationship.

1. Constructing a second physical node set FC (NS);

(1) will each

Comparing, and determining a demand maximum value dmmax and a demand minimum value dmmin;

(2) screening the resource capacity value of each preset physical node to determine a second resource capacity value, wherein the second resource capacity value meets the requirement

The conditions of (a);

(3) and sorting the preset physical nodes according to the Ci to generate a second physical node set, namely FC (NS) { n1, n2..

2. Constructing an example set NS (FC);

(1) calculating each according to a preset formula

And cap_nThe difference result Δ R of (a);

(2) determining a VNF instance fi of the second VNF corresponding to the difference result delta R with the minimum value according to the difference results delta R;

(3) adding fi into the instance set through NS (FC) U { fi }, and marking fi as read;

(4) returning to the step (1) to determine fi in a circulating mode until all fi are marked as read;

(5) and a set of output instances ns (fc).

3. Constructing a matching relation match { fi, ni };

(1) when a preset physical node is not deployed in each VNF of a service chain, reading ni in FC (NS) one by one aiming at the VNF not deployed with the preset physical node;

(2) will be provided with

And

comparing, determining to satisfy

Presetting a physical node under a condition, and establishing a matching relation between fi and ni when the VNF which is not deployed with the preset physical node corresponds to the fi arranged at the first position of NS (FC);

(3) updating the second resource capacity value with the second resource demand value, i.e.

(4) Removing fi from NS (FC), removing ni from FC (NS);

(5) if not, satisfy

If the preset physical node of the condition is used, setting identifiers which do not meet the condition aiming at other VNF instances arranged on the rear column of fi;

(6) returning to the step (1) to circularly establish the matching relationship between each fi and each ni;

(7) and outputting a matching result Mach (fi, ni).

Therefore, the preference set FC (NS) of the second VNF to the preset physical node is bidirectionally matched with the preference set NS (FC) of the second VNF by the preset physical node, and the optimal matching result can be obtained; after the parallel second VNFs are mapped, the balance between the computing power and the processing time can be achieved.

Referring to fig. 2, fig. 2 is a schematic device structure diagram of a hardware operating environment related to a method according to an embodiment of the present invention.

The service chain constructing device in the embodiment of the present invention may be a Personal Computer (PC), or may be a terminal device such as a smart phone, a tablet computer, an electronic book reader, or a portable computer.

As shown in fig. 2, the service chain construction device may include: a processor 1001, such as a CPU (Central Processing Unit), a memory 1005, and a communication bus 1002. The communication bus 1002 is used for realizing connection communication between the processor 1001 and the memory 1005. The memory 1005 may be a Random Access Memory (RAM) or a non-volatile memory (disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.

Optionally, the device for constructing the service chain may further include a user interface, a network interface, a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi (Wireless broadband) module, and the like. The user interface may comprise a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface may also comprise a standard wired interface, a wireless interface. The network interface may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface).

Those skilled in the art will appreciate that the service chain building apparatus configuration shown in fig. 2 does not constitute a limitation of the service chain building apparatus and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components.

As shown in fig. 2, a memory 1005, which is a kind of computer-readable storage medium, may include therein an operating system, a network communication module, and a service chain construction program. The operating system is a program that manages and controls the service chain's building equipment hardware and software resources, supports the service chain's building programs, and the execution of other software and/or programs. The network communication module is used to implement communication between the components within the memory 1005 and with other hardware and software in the service chain building apparatus.

In the service chain construction apparatus shown in fig. 2, the processor 1001 is configured to execute a service chain construction program stored in the memory 1005, and implement the steps in each embodiment of the service chain construction method described above.

The present invention provides a computer readable storage medium storing one or more programs, which are further executable by one or more processors for implementing the steps in the embodiments of the method for building a service chain as described above.

It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a computer-readable storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, and includes several instructions for enabling a terminal device (such as a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all modifications and equivalents of the present invention, which are made by the contents of the present specification and the accompanying drawings, or directly/indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. A method for constructing a service chain is characterized by comprising the following steps:

reading a function calculation capacity value corresponding to each function resource capacity value, determining a maximum function calculation capacity value in each function calculation capacity value, and determining a preset physical node corresponding to the maximum function calculation capacity value as a physical node corresponding to the service chain;

and deploying the service chain to the physical node to complete the construction of the service chain.

2. The method of claim 1, wherein the step of determining whether the number of VNFs is less than a preset number is followed by:

3. The method of claim 2, wherein the step of determining the first physical node corresponding to the first VNF in the service chain based on each of the first resource requirement values comprises:

4. The method for constructing a service chain according to claim 3, wherein the step of generating the first capability score and the first distance score of each preset physical node in the first set of physical nodes according to each of the first calculated capability values and each of the first node distances comprises:

5. The method of claim 2, wherein the step of determining a third physical node corresponding to a third VNF in the service chain based on each of the second resource requirement values comprises:

6. The method of constructing a service chain according to claim 5, wherein said step of determining whether said maximum capacity value is greater than said third resource requirement value is followed by the steps of:

7. The method for constructing a service chain according to claim 6, wherein the step of generating a second set of physical nodes according to each of the second resource requirement values and each of the resource capacity values comprises:

8. A service chain construction apparatus, characterized in that the service chain construction apparatus comprises: a memory, a processor, a communication bus, and a service chain builder stored on the memory;

the processor is configured to execute the service chain building program to implement the steps of the service chain building method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a service chain construction program which, when executed by a processor, implements the steps of the service chain construction method according to any one of claims 1 to 7.