CN112235125B - Networked software shared resource allocation method based on Agent bidding information strategy - Google Patents
Networked software shared resource allocation method based on Agent bidding information strategy Download PDFInfo
- Publication number
- CN112235125B CN112235125B CN202010943596.4A CN202010943596A CN112235125B CN 112235125 B CN112235125 B CN 112235125B CN 202010943596 A CN202010943596 A CN 202010943596A CN 112235125 B CN112235125 B CN 112235125B
- Authority
- CN
- China
- Prior art keywords
- gene
- solution
- agent
- bidding information
- software
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a networked software shared resource allocation method based on an Agent bidding information strategy, which comprises the steps of firstly, obtaining an Agent request set, modeling problems, determining a target function, reducing a candidate range according to conditions and removing a software entity Agent with weak competition belief in order to reduce the operation overhead of time and space; secondly, calculating the obtained Agent candidate set according to a profit maximization principle of a shared resource management module and an Agent demand constraint condition of the maximum competitive advantage; and finally, feeding back information including whether the software entity has the resource use authority, the time length for occupying the resources, the bid price of the winning bid and the like to the competitive software entity according to the calculation result. According to the method, the networked software shared resources are distributed according to the idea of the Agent bidding strategy, so that resource request conflicts of the networked software are avoided, emergencies can be flexibly responded, and the capability of overall resource allocation of users is improved.
Description
Technical Field
The invention belongs to the technical field of computers, and particularly relates to a networked software shared resource allocation method based on an Agent bidding information strategy.
Background
With the rise of distributed network software based on resource sharing, software resource sharing makes networked software a new mode of software existence. In large-scale, distributed and networked software, a software node has a few necessary local resources and also has some unnecessary resources, and in order to reduce the cost, the unnecessary resources are generally used by sharing. In a network formed by actual large-scale software, shared resources are uniformly managed by a centralized resource module. Only necessary local resources and a small number of communication resources adjacent to the software nodes are maintained among the software entities, other unnecessary resources are requested to the shared resource management module through the networked software entities only when used, and the shared resource management module is distributed through a shared resource distribution method.
The current shared resource allocation method comprises a mobile cloud network resource allocation method based on service perception, an optimization method combining scheduling and resource allocation based on a genetic algorithm and the like. The mobile cloud network resource allocation method based on service awareness mainly realizes mobile cloud network resource allocation through a resource allocation framework based on service awareness, and the resource allocation framework comprises a user layer, a request management layer and a resource providing layer. The resource sharing strategy of the method for accepting the disclosure needs the cooperation of different types of virtual machines, actively allocates the operation to the resource with high processing capacity, is a passive acceptance for the allocation object, does not consider the autonomy and the activity of the resource demand object, and therefore has the blindness of resource and task allocation to a certain extent. The optimization method combines scheduling and resource allocation under the condition of meeting scheduling limit and power consumption limit, and effectively optimizes the system performance with lower computational complexity. The method aims at scheduling and resource allocation among a plurality of cooperative units, the condition constraint is too strict, and effective information communication among the cooperative units is limited to a certain extent.
Therefore, it is significant to provide a method for allocating shared resources, which can flexibly and autonomously acquire required resources, avoid the conflict of requests for networked software resources, implement dynamic request and release of shared resources, and maximize the allocation benefit of shared resources.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a networked software shared resource allocation method based on an Agent bidding information strategy. The technical problem to be solved by the invention is realized by the following technical scheme:
the invention provides a networked software shared resource allocation method based on an Agent bidding information strategy, which comprises the following steps:
s1: acquiring an Agent request set, modeling a problem, and determining an objective function, wherein the objective function represents the integral maximum benefit of a shared resource management module;
s2: screening the Agent request set to obtain an Agent candidate set;
s3: carrying out coding design on chromosomes according to a variable length mode on potential feasible solutions of the problem;
s4: generating an initial population according to the Agent candidate set;
s5: constructing a fitness function according to the target function;
s6: generating a new generation population according to genetic operators, wherein the genetic operators comprise a segmentation intersection operator, a replacement operator and a sequence comparison operator;
s7: evaluating the fitness of the new generation population according to the fitness function, and reserving an optimal solution according to an elite reservation strategy;
s8: judging whether the optimal solution meets a preset convergence criterion, if so, executing a step S9, otherwise, directly jumping to the step S6 to continue propagation;
s9: obtaining a shared resource allocation scheme according to the optimal solution, and implementing resource allocation according to the shared resource allocation scheme;
wherein the fitness function is:
wherein, f (X) represents the cost value obtained by the shared resource management module, g (X) represents the energy consumption value of resource saving, sigma represents the calibration parameter, and X represents the feasible solution.
In one embodiment of the present invention, in the S1:
the Agent request setA union represents a set of software entities that request the same shared resource and cannot be satisfied simultaneously, the set of Agent requests being Agent ═ { A }1,A2,...Ai...,An}，AiRepresenting the software entity, setting the software entity AiThe belief of competition is Bi＝{i,ti,di,bi}，tiIndicating the duration of resource occupation, diDenotes the maximum completion deadline, biThe price is shown to be marked,
the objective function is:
Max[f(X)，g(X)]，
g(X)＝λ×(T-∑ti)，
wherein, λ represents an internal fixed reserved utility parameter of the shared resource management module, and T represents a total time length theoretically used by a certain shared resource.
In an embodiment of the present invention, the S2 includes, according to a screening condition, performing corresponding screening on software entities that request the same shared resource and have the same request occupation duration to obtain the Agent candidate set, where the Agent candidate set is a ═ { a ═ a {1,A2,...,AmAnd m is less than n, wherein the screening conditions comprise:
if the resource occupation duration of the software entity meets ti＞(diNowtime)/2, then the same tiSelecting a software entity with the largest price;
if the resource occupation duration of the software entity satisfies (d)i-nowtime)/z≥ri＞(di-nowtime)/(z+1)，z∈Z+Then the same tiSelecting the first z software entities with larger price;
if the resource occupation duration of the software entity meets tiWhen T/2, at most the symbol is selectedSum of ti＞(nowtime-di) All software entities of (a);
wherein nowtime represents the current time, Z+Representing a set of positive integers.
In an embodiment of the present invention, the S3 includes:
the gene expression is expressed by adopting a variable length mode, and the feasible solution X is subjected to chromosome coding by adopting a direct coding method to obtain a potential feasible solution X ═ a1,a2,...,ai,...am) X ∈ D, where D denotes the solution space formed by all feasible solutions, aiRepresenting a gene in a chromosome, for which corresponding software entity AiThe software entity A, the bidding information ofiThe bid information of (c) is set as (t)i,di,τi)，tiIndicating the duration of resource occupation, diDenotes the maximum completion deadline, τiRepresenting a bid;
wherein the feasible solution X satisfies the following conditions:
feasible solution of any gene a in XiT corresponding to bid informationiSatisfies the following conditions: t ≧ Σ { Ti∣ai∈X}，diSatisfies the following conditions:
wherein k is ∈ [1, m ]]。
In an embodiment of the present invention, the S4 includes:
s41: randomly selecting a software entity A from the Agent candidate set AiIf the bidding information corresponding to the bidding candidate is used as the initial gene, the first gene of the feasible solution X is aiWhen X is ═ ai)；
S42: if Σ { ti∣aiIf the element belongs to X } < T, selecting the element candidate set A to satisfy Ti≤T-∑{ti∣aiE.g. X and ∑ ti∣ai∈X}＜Max(di-nowtime) condition, constructing a target set H, if
Jumping to step S45, otherwise executing step S43;
s43: randomly selecting a software entity A from the target set HkIf it corresponds to the bidding information (t)k,dk,τk) The deadline time condition is satisfied: (d)k-nowtime)≥∑{ti∣ai∈X}+tkThen, it is added as a gene to the feasible solution X, in which case X ═ ai,ak) Calculating the total time length of the feasible solution X at the moment, then jumping to step S42, and if the time length condition of the deadline is not met, executing step S44; wherein, the calculation formula of the total duration is as follows: sigma { t }i∣ai∈X}＝∑{ti∣ai∈X}+tk；
S44: deletion of Gene a from the target set HkCorresponding software entity AkThen judging whether the target set H is an empty set, if so, judging whether the target set H is an empty set
Skipping to step S43 to continue searching, otherwise, indicating that the current feasible solution X has been generated, and executing step S45;
s45: and judging whether the generated feasible solution X meets the preset requirement of the population scale, if so, terminating the population generation, otherwise, skipping to the step S41.
In an embodiment of the present invention, the S6 includes:
s61: randomly selecting a feasible solution X in the contemporary population as a father P1And duplicate individuals P2Generating two new solutions P 'after hybridization by using the segment crossover operator'1And P'2；
S62: utilizing the replacement operator to create a front part which has the same resource occupation time but gives higher benefit to the resource allocation module aiming at all gene positions of each solution processed by the segmentation and intersection operator
Is a new set, whichLength (X) represents the length of the chromosome, and r represents the number of genes corresponding to the bidding information with the same resource occupation duration; any bid information c ═ t (t) in the new set obtainedi,di,τi) In, if diIf the maximum completion deadline of the bidding information on the current gene position is longer than the maximum completion deadline of the bidding information on the current gene position, replacing the bidding information on the current gene position with the bidding information c, and if d is greater than the maximum completion deadline of the bidding information on the current gene positioniLess than the maximum completion deadline for the bid information at the current genetic location, and diIf the requirement of the maximum completion deadline is met, replacing the bidding information on the current gene position with the bidding information c;
s63: utilizing a sequencing operator to create a gene set with the same occupied time requirement according to the bidding information corresponding to each gene of each solution processed by the replacement operator, and establishing the bidding information of the set according to diAnd (4) positive sequence arrangement of the nowtime values, if the gene positions corresponding to the bidding information are inconsistent with the sequence, sequentially arranging the sequences as the standard, replacing the inconsistent gene positions, and otherwise, keeping the sequence unchanged.
In an embodiment of the present invention, the S61 includes:
s611: from the parent P1In (b), preselecting any jth gene locus, and selecting a gene according to
Obtaining the total time length S of the resource occupation time length of the bidding information corresponding to the first j gene positions;
s612: reserving the parent P1Determining the genes after the j position according to the following steps to generate new solution P'1：
Step 1, constructing a candidate set A meeting the following conditions from the Agent candidate set A: T-S ≧ Σ Ti，di-nowtime > S, and the parent P1All genes a after the j-th position of (1)iSet I of (1), if
Skipping to the step 4, otherwise, executing the step 2;
step 2, randomly selecting a gene akIf S + tk≤dk-nowtime, then add it to the new solution P'1And recalculate the current new solution P'1S + t, the total duration of the resource occupation durationk；
Step 3, if S is equal to T, executing step 4, otherwise, jumping to step 1;
step 4, outputting newly-solved P'1；
S613: keeping the copy individual P2And determining the gene preceding the j-th position according to the following procedure to generate a new solution P'2：
Step 1. suppose software entity AiThe bid information of (1) is marked starting from 0 and let S be 0; i is 0;
step 2, establishing a product meeting the conditions: sigma ti≤T-S，di-nowtime > S, and said replica individual P2All genes a before the j-th position of (1)iIf the set of
Skipping to the step 5, otherwise, executing the step 3;
step 3, randomly selecting a gene akIf it is maximum completion cut-off time dkSatisfies the following conditions: s + tk≤dkThe nowtime, let i +1, and akAs newly decomposed P'2The current new solution P 'is simultaneously obtained for the ith gene of'2S + t, the total duration of the resource occupation durationk；
Step 4, if S is equal to T, executing step 5, otherwise, jumping to step 2 to continue searching;
step 5, the copy individual P2Sequentially adding the genes after the j-th position of (1) to newly solved P'2At the gene position subsequent to the i-th position of (1), and outputs a new solution P'2。
In an embodiment of the present invention, the S7 includes:
and evaluating the fitness of the new generation population according to the fitness function, reserving alpha feasible solutions with high fitness evaluation of the previous generation population as optimal solutions, forming an optimal solution candidate set with the current generation population, and selecting beta feasible solutions with high fitness evaluation in the optimal solution candidate set as the next generation population.
In an embodiment of the present invention, in S8, the convergence criterion is: if the optimal solution has not evolved for g consecutive generations, o new solutions are randomly added to replace o poor solutions in the current population and continue to evolve, and if the poor solutions are continuously replaced for q times, the optimal solution still has not evolved, and at the moment, the optimal solution is shown to be converged.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the networked software shared resource allocation method based on the Agent bidding information strategy, the networked software shared resource is allocated according to the idea of the Agent bidding strategy, so that the conflict of networked software resource requests is avoided, the emergency can be flexibly coped with, and the capability of performing overall resource allocation by a user is improved.
2. According to the networked software shared resource allocation method based on the Agent bidding information strategy, the new generation of population is formed based on the matching of the proportion and the fitness of the two generation of population, the difference and the excellence are eliminated, the quality of solution group is improved, the convergence of solution group is accelerated, and the allocation method is high in efficiency.
3. According to the networked software shared resource allocation method based on the Agent bidding information strategy, the fact that the resource occupation duration and the maximum completion deadline of bidding are different is considered, the traditional crossover operator is abandoned, the segmented crossover operator, the replacement operator and the order operator are adopted to implement gene variation, the method is enabled to be more adaptive and reliable, the whole allocation method can provide technical reference for distributed, heterogeneous and dynamic software shared resource allocation, and the maximization of shared resource allocation income is achieved.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical means of the present invention more clearly understood, the present invention may be implemented in accordance with the content of the description, and in order to make the above and other objects, features, and advantages of the present invention more clearly understood, the following preferred embodiments are described in detail with reference to the accompanying drawings.
Drawings
FIG. 1 is a schematic diagram of a networked software shared resource allocation method based on an Agent bidding information policy according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for allocating shared resources of networked software based on Agent bidding information policy according to an embodiment of the present invention;
FIG. 3 is a flow chart for generating a new solution based on a segment intersection operator, provided by an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and effects of the present invention adopted to achieve the predetermined invention purpose, the following describes in detail a networked software sharing resource allocation method based on Agent bidding information policy according to the present invention with reference to the accompanying drawings and the detailed embodiments.
The foregoing and other technical matters, features and effects of the present invention will be apparent from the following detailed description of the embodiments, which is to be read in connection with the accompanying drawings. The technical means and effects of the present invention adopted to achieve the predetermined purpose can be more deeply and specifically understood through the description of the specific embodiments, however, the attached drawings are provided for reference and description only and are not used for limiting the technical scheme of the present invention.
Example one
Referring to fig. 1 and fig. 2 in combination, fig. 1 is a schematic diagram of a networked software shared resource allocation method based on an Agent bidding information policy according to an embodiment of the present invention; FIG. 2 is a flowchart of a method for allocating shared resources of networked software based on Agent bidding information policy according to an embodiment of the present invention. As shown, the method of the present invention comprises:
s1: acquiring an Agent request set, modeling a problem, and determining an objective function, wherein the objective function represents the integral maximum benefit of a shared resource management module;
in this embodiment, the software entities are represented by agents, the requests of the software entities are defined as beliefs, and when a plurality of software entities request shared resources, a request is formedAnd (4) solving for Agent set, and generating belief conflict in the process of requesting. In particular, software entity A1And A2The same resource which cannot be satisfied simultaneously is requested, and at the moment, the two resources compete for each other, if the phenomenon is amplified to havenIn a network of software entities, then, the set of Agent requests represents a set of software entities that request the same shared resource and cannot be satisfied simultaneously, the set of Agent requests being Agent { a ═ a1,A2,...Ai...,An}，AiRepresenting a software entity.
At an arbitrary software entity AiIn the method, the resource requirements are constrained from the aspects of users, current environment estimation and the like, a certain belief constraint condition is formed, and a software entity A is setiThe belief of competition is Bi＝{i,ti,di,bi}，tiIndicating the duration of resource occupation, diDenotes the maximum completion deadline, biThe price is shown to be marked,
in general, the belief constraints between the demand agents are different, which just shows that the belief strengths of each software entity Agent are different, and the competitive abilities are different.
If the shared resource management module senses the competition beliefs of the plurality of software entities for the resources at the same time, that is, the software entities initiate bids, the shared resource management module performs allocation according to the shared resource allocation method of the embodiment on the basis of the maximum profit principle, and finally feeds back the allocation results to all the software entities participating in the resource competition, including information such as whether the software entities have resource usage rights, the duration of the resources can be occupied, bid price of winning bid, and the like.
The objective function is:
Max[f(X)，g(X)]，
g(X)＝λ×(T-∑ti)，
wherein, λ represents an internal fixed reserved utility parameter of the shared resource management module, and T represents a total time length theoretically used by a certain shared resource. X represents a feasible solution, namely a solution meeting all the constraint conditions can ensure that the same resource is allocated to one software entity at most at any time and can also meet the constraint conditions expressed in the beliefs of each software entity.
S2: screening the Agent request set to obtain an Agent candidate set;
specifically, the S2 includes:
according to the screening conditions, corresponding screening is carried out on software entities which request the same shared resource and have the same request occupation time length, the software entities with weak competition beliefs are removed, and the Agent candidate set is obtained, wherein the Agent candidate set is A ═ { A ═ A1,A2,...,AmAnd m is less than n, wherein the screening conditions comprise:
if the resource occupation duration of the software entity meets ti＞(diNowtime)/2, then the same tiSelecting a software entity with the largest price;
if the resource occupation duration of the software entity satisfies (d)i-nowtime)/z≥ti＞(di-nowtime)/(z+1)，z∈Z+Then the same tiSelecting the first z software entities with larger price;
if the resource occupation duration of the software entity meets tiIf T/2, then at most T is chosen to be truei＞(nowtime-di) Or selecting any subset of all software entities meeting the condition;
wherein nowtime represents the current time, Z+Representing a set of positive integers.
S3: carrying out coding design on chromosomes according to a variable length mode on potential feasible solutions of the problem;
if a certain chromosome is the optimal solution, the software entity which wins the bid is the software entity object to which the resource is allocated. This means that the resource occupation duration finally obtained for each software entity is the same as that of the request, and the provided profit value conforms to the expectation of the resource management module. Of course, there may be situations where the resulting deadline for the actual execution of the resource is not consistent with the deadline expected by the software entity, but the actual requirement of the software entity is met when the deadline is equal to or less than the deadline required by the software entity.
Specifically, the S3 includes:
the gene expression is expressed by adopting a variable length mode, and the feasible solution X is subjected to chromosome coding by adopting a direct coding method to obtain a potential feasible solution X ═ a1,a2,...,ai,...am) X ∈ D, where D denotes the solution space formed by all feasible solutions, a feasible solution X is also a legal chromosome, aiRepresenting a gene in a chromosome, for which corresponding software entity AiThe software entity A, the bidding information ofiThe bid information of (c) is set as (t)i,di,τi)，tiIndicating the duration of resource occupation, diDenotes the maximum completion deadline, τiRepresenting a bid.
In this embodiment, software entity A is employediThe resource occupation time t in the bid informationiAnd carrying out chromosome coding on the feasible solution X. Because the total time for the shared resource is limited, and the competing software entities have different requirements for the time for the resource usage, the number of genes in the chromosome is not uniformly limited, i.e., the chromosome length is adjusted in time.
Wherein the feasible solution X satisfies the following conditions:
feasible solution of any gene a in XiT corresponding to bid informationiSatisfies the following conditions: t ≧ Σ { Ti∣ai∈X}，diSatisfies the following conditions:
wherein k is ∈ [1, m ]]。
S4: generating an initial population according to the Agent candidate set;
since the length of the chromosome is not fixed, it depends on the number of software entities used for encoding, and at the same time, each bid information item of the gene of the chromosome satisfies the above-mentioned condition.
Specifically, the S4 includes:
s41: randomly selecting a software entity A from the Agent candidate set AiIf the bidding information corresponding to the bidding candidate is used as the initial gene, the first gene of the feasible solution X is aiWhen X is ═ ai)；
S42: if Σ { ti∣aiIf the element belongs to X } < T, selecting the element candidate set A to satisfy Ti≤T-∑{ti∣aiE.g. X and ∑ ti∣ai∈X}＜Max(di-nowtime) condition, constructing a target set H, if
Jumping to step S45, otherwise executing step S43;
s43: randomly selecting a software entity A from the target set HkIf it corresponds to the bidding information (t)k,dk,τk) The deadline time condition is satisfied: (d)k-nowtime)≥∑{ti∣ai∈X}+tkThen, it is added as a gene to the feasible solution X, in which case X ═ ai,ak) Calculating the total time length of the feasible solution X at the moment, then jumping to step S42, and if the time length condition of the deadline is not met, executing step S44; wherein, the calculation formula of the total duration is as follows: sigma { t }i∣ai∈X}＝∑{ti∣ai∈X}+tk；
S44: deletion of Gene a from the target set HkCorresponding software entity AkThen judging whether the target set H is an empty set, if so, judging whether the target set H is an empty set
Skipping to step S43 to continue searching, otherwise, indicating that the current feasible solution X has been generated, and executing step S45;
s45: and judging whether the generated feasible solution X meets the preset requirement of the population scale, if so, terminating the population generation, otherwise, skipping to the step S41.
S5: constructing a fitness function according to the target function;
because the objective function is the maximum value of the gains obtained by solving the shared resource module, the fitness function of the chromosome is constructed by combining the objective function as follows:
wherein, f (X) represents the cost value obtained by the shared resource management module, g (X) represents the energy consumption value of resource saving, sigma represents the calibration parameter, and X represents the feasible solution. In this embodiment, σ may be the minimum objective function in the current generation or the recent generation group, or an empirical value with a small value may be predetermined.
S6: generating a new generation population according to genetic operators, wherein the genetic operators comprise a segmentation intersection operator, a replacement operator and a sequence comparison operator;
because the number of genes of the chromosome weight is not uniformly limited, namely the chromosome length is adjusted timely, each gene corresponds to t of the bidding informationiAnd diAll are different, so that for solutions represented by variable-length chromosomes, traditional genetic operators such as Partial-Mapped crossbar, Cycle crossbar, Ordercrossbar and the like are not suitable for solving the problem, and therefore, a segmented Crossover operator, a replacement operator and an order operator are used and sequentially executed to generate a new generation of population.
Referring to fig. 3, fig. 3 is a flowchart of generating a new solution according to a segment intersection operator according to an embodiment of the present invention, and as shown in the drawing, the S6 includes:
s61: randomly selecting a feasible solution X in the contemporary population as a father P1And duplicate individuals P2Generating two new solutions P 'after hybridization by using the segment crossover operator'1And P'2；
Further, the S61 includes:
s611: from the parent P1In (b), preselecting any jth gene locus, and selecting a gene according to
Obtaining the total time length S of the resource occupation time length of the bidding information corresponding to the first j gene positions;
s612: reserving the parent P1Determining the genes after the j position according to the following steps to generate new solution P'1：
Step 1, constructing a candidate set A meeting the following conditions from the Agent candidate set A: T-S ≧ Σ Ti，di-nowtime > S, and the parent P1All genes a after the j-th position of (1)iSet I of (1), if
Skipping to the step 4, otherwise, executing the step 2;
step 2, randomly selecting a gene akIf S + tk≤dk-nowtime, then add it to the new solution P'1And recalculate the current new solution P'1S + t, the total duration of the resource occupation durationk；
Step 3, if S is equal to T, executing step 4, otherwise, jumping to step 1;
step 4, outputting newly-solved P'1；
S613: keeping the copy individual P2And determining the gene preceding the j-th position according to the following procedure to generate a new solution P'2：
Step 1. suppose software entity AiThe bid information of (1) is marked starting from 0 and let S be 0; i is 0;
step 2, establishing a product meeting the conditions: sigma ti≤T-S，di-nowtime > S, and said replica individual P2All genes a before the j-th position of (1)iIf the set of
Jump to step 5Otherwise, executing step 3;
step 3, randomly selecting a gene akIf it is maximum completion cut-off time dkSatisfies the following conditions: s + tk≤dkThe nowtime, let i +1, and akAs newly decomposed P'2The current new solution P 'is simultaneously obtained for the ith gene of'2S + t, the total duration of the resource occupation durationk；
Step 4, if S is equal to T, executing step 5, otherwise, jumping to step 2 to continue searching;
step 5, the copy individual P2Sequentially adding the genes after the j-th position of (1) to newly solved P'2At the gene position subsequent to the i-th position of (1), and outputs a new solution P'2。
S62: utilizing the replacement operator to create a front part which has the same resource occupation time but gives higher benefit to the resource allocation module aiming at all gene positions of each solution processed by the segmentation and intersection operator
The new set is obtained, wherein length (X) represents the length of the chromosome, and r represents the number of genes corresponding to the bidding information with the same resource occupation duration; any bid information c ═ t (t) in the new set obtainedi,di,τi) In, if diIf the maximum completion deadline of the bidding information on the current gene position is longer than the maximum completion deadline of the bidding information on the current gene position, replacing the bidding information on the current gene position with the bidding information c, and if d is greater than the maximum completion deadline of the bidding information on the current gene positioniLess than the maximum completion deadline for the bid information at the current genetic location, and diIf the requirement of the maximum completion deadline is met, replacing the bidding information on the current gene position with the bidding information c;
s63: utilizing a sequencing operator to create a gene set with the same occupied time requirement according to the bidding information corresponding to each gene of each solution processed by the replacement operator, and establishing the bidding information of the set according to diPositive sequence arrangement of the nowtime values, if the gene positions corresponding to each bidding information are inconsistent with the sequence arrangement, the sequence arrangement is accurate, and the inconsistent gene positions are replaced, so that the solution is better, otherwise, the solution is kept unchanged。
In this embodiment, the order comparing operator is adopted to aim at adjusting the sequence of the gene, so that not only is the essential information of each gene position not changed, but also the requirement of the deadline of each bidding information in the solution can be met as much as possible through the operator, and meanwhile, the deadline of each bidding information is softened, so that the possibility is provided for better matching with the segment crossing operator and the replacing operator, and a better solution is obtained.
S7: evaluating the fitness of the new generation population according to the fitness function, and reserving an optimal solution according to an elite reservation strategy;
specifically, the S7 includes:
and evaluating the fitness of the new generation population according to the fitness function, reserving alpha feasible solutions with high fitness evaluation of the previous generation population as optimal solutions, forming an optimal solution candidate set with the current generation population, and selecting beta feasible solutions with high fitness evaluation in the optimal solution candidate set as the next generation population. The entire population can evolve towards the optimal direction of the solution by adopting an elite retention strategy, the convergence rate of the algorithm is accelerated, wherein alpha and beta are preset parameters, and are set according to actual requirements without limitation.
S8: judging whether the optimal solution meets a preset convergence criterion, if so, executing a step S9, otherwise, directly jumping to the step S6 to continue propagation;
specifically, in S8, the convergence criterion is:
if the optimal solution has not evolved for g consecutive generations, o new solutions are randomly added to replace o poor solutions in the current population and continue to evolve, and if the poor solutions are continuously replaced for q times, the optimal solution still has not evolved, and at the moment, the optimal solution is shown to be converged. G, o, and q are preset parameters, and may be set according to the reliability requirement of the algorithm, which is not limited herein.
S9: and obtaining a shared resource allocation scheme according to the optimal solution, and implementing resource allocation according to the shared resource allocation scheme.
According to the networked software shared resource allocation method based on the Agent bidding information strategy, the new generation of population is formed based on the matching of the proportion and the fitness of the two generation of population, the difference and the excellence are eliminated, the quality of solution group is improved, the convergence of solution group is accelerated, and the allocation method is high in efficiency. And moreover, the networked software shared resources are distributed according to the idea of the Agent bidding strategy, so that the conflict of networked software resource requests is avoided, the emergency can be flexibly coped with, and the capability of performing overall resource allocation by a user is improved.
The method of the embodiment has better adaptability and reliability, the whole distribution method can provide technical reference for the distribution of distributed, heterogeneous and dynamic software shared resources, and the maximization of the benefit of the distribution of the shared resources is realized.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.
Claims (6)
1. A networked software shared resource allocation method based on an Agent bidding information strategy is characterized by comprising the following steps:
s1: acquiring an Agent request set, modeling a problem, and determining an objective function, wherein the objective function represents the integral maximum benefit of a shared resource management module;
in the S1:
the Agent request set represents a set of software entities that request the same shared resource and cannot be satisfied simultaneously, the Agent request set being Agent { A } {1,A2,...Ai...,An}，AiRepresenting the software entity, setting the software entity AiThe belief of competition is Bi＝{i,ti,di,bi}，tiIndicating the duration of resource occupation, diDenotes the maximum completion deadline, biThe price is shown to be marked,
the objective function is:
Max[f(X)，g(X)]，
g(X)＝λ×(T-∑ti)，
wherein, λ represents the default reserved utility parameter of the shared resource management module, T represents the total time length theoretically used by a certain shared resource,
indicating a bid price biIs associated with the resource occupation duration tiAnd a maximum completion deadline diThe function in question is a function of the relationship,
is a function symbol;
s2: screening the Agent request set to obtain an Agent candidate set;
s3: carrying out coding design on chromosomes according to a variable length mode on potential feasible solutions of the problem;
in the S3:
the gene expression is expressed by adopting a variable length mode, and the feasible solution X is subjected to chromosome coding by adopting a direct coding method to obtain a potential feasible solution X ═ a1,a2,...,ai,...am) X ∈ D, where D denotes the solution space formed by all feasible solutions, aiRepresenting a gene in a chromosome, for which corresponding software entity AiThe software entity A, the bidding information ofiThe bid information of (c) is set as (t)i,di,τi)，tiIndicating the duration of resource occupation, diDenotes the maximum completion deadline, τiRepresenting a bid;
wherein the feasible solution X satisfies the following conditions:
feasibleSolving any gene a in XiT corresponding to bid informationiSatisfies the following conditions: t ≧ Σ { Ti∣ai∈X}，diSatisfies the following conditions:
wherein k is ∈ [1, m ]]；
S4: generating an initial population according to the Agent candidate set;
s5: constructing a fitness function according to the target function;
s6: generating a new generation population according to genetic operators, wherein the genetic operators comprise a segmentation intersection operator, a replacement operator and a sequence comparison operator;
the S6 includes:
s61: randomly selecting a feasible solution X in the contemporary population as a father P1And duplicate individuals P2Generating two new solutions P after hybridization by using the segment intersection operator1'and P'2；
S62: using the replacement operator to create front [ r ] with the same resource occupation time but giving higher benefit to the resource allocation module for all gene positions of each solution processed by the segmentation and intersection operator2/Length(X)]The new set is obtained, wherein length (X) represents the length of the chromosome, and r represents the number of genes corresponding to the bidding information with the same resource occupation duration; any bid information c ═ t (t) in the new set obtainedi,di,τi) In, if diIf the maximum completion deadline of the bidding information on the current gene position is longer than the maximum completion deadline of the bidding information on the current gene position, replacing the bidding information on the current gene position with the bidding information c, and if d is greater than the maximum completion deadline of the bidding information on the current gene positioniLess than the maximum completion deadline for the bid information at the current genetic location, and diIf the requirement of the maximum completion deadline is met, replacing the bidding information on the current gene position with the bidding information c;
s63: utilizing a sequencing operator to create a gene set with the same occupied time requirement according to the bidding information corresponding to each gene of each solution processed by the replacement operator, and establishing the bidding information of the set according to diPositive ordering of the nowtime values, if each bid is dueIf the corresponding gene positions are inconsistent with the sequence, sequentially ordering the corresponding gene positions as a standard, replacing the inconsistent gene positions, otherwise, keeping the gene positions unchanged;
s7: evaluating the fitness of the new generation population according to the fitness function, and reserving an optimal solution according to an elite reservation strategy;
s8: judging whether the optimal solution meets a preset convergence criterion, if so, executing a step S9, otherwise, directly jumping to the step S6 to continue propagation;
s9: obtaining a shared resource allocation scheme according to the optimal solution, and implementing resource allocation according to the shared resource allocation scheme;
wherein the fitness function is:
wherein, f (X) represents the cost value obtained by the shared resource management module, g (X) represents the energy consumption value of resource saving, sigma represents the calibration parameter, and X represents the feasible solution.
2. The method of claim 1, wherein: the step S2 includes, according to the screening condition, performing corresponding screening on software entities that request the same shared resource and have the same request occupation duration to obtain the Agent candidate set, where the Agent candidate set is a ═ { a ═ a1,A2,...,AmAnd m is less than n, wherein the screening conditions comprise:
if the resource occupation duration of the software entity meets ti＞(diNowtime)/2, then the same tiSelecting a software entity with the largest price;
if the resource occupation duration of the software entity satisfies (d)i-nowtime)/z≥ti＞(di-nowtime)/(z+1)，z∈Z+Then the same tiSelecting the first z software entities with larger price;
if the resource occupation duration of the software entity meets tiWhen T/2, thenAt most, choose to meet ti＞(nowtime-di) All software entities of (a);
wherein nowtime represents the current time, Z+Representing a set of positive integers.
3. The method of claim 2, wherein: the S4 includes:
s41: randomly selecting a software entity A from the Agent candidate set AiIf the bidding information corresponding to the bidding candidate is used as the initial gene, the first gene of the feasible solution X is aiWhen X is ═ ai)；
S42: if Σ { ti∣aiIf the element belongs to X } < T, selecting the element candidate set A to satisfy Ti≤T-∑{ti∣aiE.g. X and ∑ ti∣ai∈X}＜Max(di-nowtime) condition, constructing a target set H, if
Jumping to step S45, otherwise executing step S43;
s43: randomly selecting a software entity A from the target set HkIf it corresponds to the bidding information (t)k,dk,τk) The deadline time condition is satisfied: (d)k-nowtime)≥∑{ti∣ai∈X}+tkThen, it is added as a gene to the feasible solution X, in which case X ═ ai,ak) Calculating the total time length of the feasible solution X at the moment, then jumping to step S42, and if the time length condition of the deadline is not met, executing step S44; wherein, the calculation formula of the total duration is as follows: sigma { t }i∣ai∈X}＝∑{ti∣ai∈X}+tk；
S44: deletion of Gene a from the target set HkCorresponding software entity AkThen judging whether the target set H is an empty set, if so, judging whether the target set H is an empty set
Skipping to step S43 to continue searching, otherwise, indicating that the current feasible solution X has been generated, and executing step S45;
s45: and judging whether the generated feasible solution X meets the preset requirement of the population scale, if so, terminating the population generation, otherwise, skipping to the step S41.
4. The method of claim 3, wherein: the S61 includes:
s611: from the parent P1In (b), preselecting any jth gene locus, and selecting a gene according to
Obtaining the total time length S of the resource occupation time length of the bidding information corresponding to the first j gene positions;
s612: reserving the parent P1The first j genes are determined according to the following steps, and a new solution P is generated1′：
Step 1, constructing a candidate set A meeting the following conditions from the Agent candidate set A: T-S ≧ Σ Ti，di-nowtime > S, and the parent P1All genes a after the j-th position of (1)iSet I of (1), if
Skipping to the step 4, otherwise, executing the step 2;
step 2, randomly selecting a gene akIf S + tk≤dk-nowtime, then add it to the new solution P1' middle, and recalculate the current new solution P1' total duration of resource occupation duration, S ═ S + tk；
Step 3, if S is equal to T, executing step 4, otherwise, jumping to step 1;
step 4, outputting new solution P1′；
S613: keeping the copy individual P2And the gene before the j-th position is determined according to the following procedure to generateNovel solution of P'2：
Step 1. suppose software entity AiThe bid information of (1) is marked starting from 0 and let S be 0; i is 0;
step 2, establishing a product meeting the conditions: sigma ti≤T-S，di-nowtime > S, and said replica individual P2All genes a before the j-th position of (1)iIf the set of
Skipping to the step 5, otherwise, executing the step 3;
step 3, randomly selecting a gene akIf it is maximum completion cut-off time dkSatisfies the following conditions: s + tk≤dkThe nowtime, let i +1, and akAs newly decomposed P'2The current new solution P 'is simultaneously obtained for the ith gene of'2S + t, the total duration of the resource occupation durationk；
Step 4, if S is equal to T, executing step 5, otherwise, jumping to step 2 to continue searching;
step 5, the copy individual P2Sequentially adding the genes after the j-th position of (1) to newly solved P'2At the gene position subsequent to the i-th position of (1), and outputs a new solution P'2。
5. The method of claim 4, wherein: the S7 includes:
and evaluating the fitness of the new generation population according to the fitness function, reserving alpha feasible solutions with high fitness evaluation of the previous generation population as optimal solutions, forming an optimal solution candidate set with the current generation population, and selecting beta feasible solutions with high fitness evaluation in the optimal solution candidate set as the next generation population.
6. The method of claim 5, wherein: in S8, the convergence criterion is: if the optimal solution has not evolved for g consecutive generations, o new solutions are randomly added to replace o poor solutions in the current population and continue to evolve, and if the poor solutions are continuously replaced for q times, the optimal solution still has not evolved, and at the moment, the optimal solution is shown to be converged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010943596.4A CN112235125B (en) | 2020-09-09 | 2020-09-09 | Networked software shared resource allocation method based on Agent bidding information strategy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010943596.4A CN112235125B (en) | 2020-09-09 | 2020-09-09 | Networked software shared resource allocation method based on Agent bidding information strategy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112235125A CN112235125A (en) | 2021-01-15 |
| CN112235125B true CN112235125B (en) | 2022-04-19 |
Family
ID=74116113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010943596.4A Active CN112235125B (en) | 2020-09-09 | 2020-09-09 | Networked software shared resource allocation method based on Agent bidding information strategy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112235125B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107229971A (en) * | 2017-06-06 | 2017-10-03 | 西安电子科技大学 | Optimal adaptive strategy decision-making technique based on GAPSO algorithms |
| CN107895225A (en) * | 2017-11-01 | 2018-04-10 | 北京邮电大学 | A kind of cooperation type method for allocating tasks of multi-Agent Lothrus apterus |
| CN108416523A (en) * | 2018-03-08 | 2018-08-17 | 中国人民解放军陆军工程大学 | Task scheduling method and device, electronic equipment and storage medium |
| CN110188785A (en) * | 2019-03-28 | 2019-08-30 | 山东浪潮云信息技术有限公司 | A kind of data clusters analysis method based on genetic algorithm |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170116522A1 (en) * | 2015-10-05 | 2017-04-27 | Telekom Malaysia Berhad | Method For Task Scheduling And Resources Allocation And System Thereof |
-
2020
- 2020-09-09 CN CN202010943596.4A patent/CN112235125B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107229971A (en) * | 2017-06-06 | 2017-10-03 | 西安电子科技大学 | Optimal adaptive strategy decision-making technique based on GAPSO algorithms |
| CN107895225A (en) * | 2017-11-01 | 2018-04-10 | 北京邮电大学 | A kind of cooperation type method for allocating tasks of multi-Agent Lothrus apterus |
| CN108416523A (en) * | 2018-03-08 | 2018-08-17 | 中国人民解放军陆军工程大学 | Task scheduling method and device, electronic equipment and storage medium |
| CN110188785A (en) * | 2019-03-28 | 2019-08-30 | 山东浪潮云信息技术有限公司 | A kind of data clusters analysis method based on genetic algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 《A Multiagent-based Framework for Self-adaptive Software with Search-based Optimization》;王璐;《IEEE》;20161231;全文 * |
| 《Event-based Evolution Mechanism in Dynamic Environment for Multi-Agent System》;李青山等;《IEEE》;20141231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112235125A (en) | 2021-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rathnayaka et al. | Framework to manage multiple goals in community-based energy sharing network in smart grid | |
| CN109165808B (en) | Power communication network on-site operation and maintenance work order distribution method | |
| Prayogo et al. | Optimization model for construction project resource leveling using a novel modified symbiotic organisms search | |
| CN113821318B (en) | Internet of things cross-domain subtask combination collaborative computing method and system | |
| CN113885555A (en) | Multi-machine task allocation method and system for power transmission line dense channel routing inspection | |
| WO2023087658A1 (en) | Task scheduling method, apparatus and device, and readable storage medium | |
| CN113191619A (en) | Emergency rescue material distribution and vehicle dispatching dynamic optimization method | |
| CN112187535B (en) | Server deployment method and device in fog computing environment | |
| CN110162390B (en) | Task allocation method and system for fog computing system | |
| CN109597682A (en) | A kind of cloud computing workflow schedule method using heuristic coding strategy | |
| CN104506576B (en) | A kind of wireless sensor network and its node tasks moving method | |
| Chen et al. | Genetic algorithm-based design and simulation of manufacturing flow shop scheduling | |
| CN114565239A (en) | Comprehensive low-carbon energy scheduling method and system for industrial park | |
| CN113139639B (en) | MOMBI-oriented smart city application multi-target computing migration method and device | |
| Bai et al. | A manufacturing task scheduling method based on public goods game on cloud manufacturing model | |
| He | Optimization of edge delay sensitive task scheduling based on genetic algorithm | |
| CN116985146B (en) | Robot parallel disassembly planning method for retired electronic products | |
| Manavi et al. | Resource allocation in cloud computing using genetic algorithm and neural network | |
| CN112235125B (en) | Networked software shared resource allocation method based on Agent bidding information strategy | |
| CN115421885B (en) | Distributed multi-target cloud task scheduling method and device and cloud service system | |
| CN104540171A (en) | Wireless sensor network and node task distribution method thereof | |
| CN110633784B (en) | Multi-rule artificial bee colony improvement algorithm | |
| CN114371915A (en) | Manufacturing enterprise data space system task scheduling method based on decomposition strategy | |
| CN115361392A (en) | Control method, system and storage medium of computing power network based on block chain | |
| CN111290853B (en) | Cloud data center scheduling method based on self-adaptive improved genetic algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |