CN114390102B - Internet of things resource allocation method, system, terminal and storage medium - Google Patents
Internet of things resource allocation method, system, terminal and storage medium Download PDFInfo
- Publication number
- CN114390102B CN114390102B CN202210026373.0A CN202210026373A CN114390102B CN 114390102 B CN114390102 B CN 114390102B CN 202210026373 A CN202210026373 A CN 202210026373A CN 114390102 B CN114390102 B CN 114390102B
- Authority
- CN
- China
- Prior art keywords
- moth
- flame
- vector
- internet
- matching
- 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
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a method, a system, a terminal and a storage medium for resource allocation of the Internet of things. The method comprises the following steps: initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on the received resource allocation task request to obtain an initialized moth vector set; evaluating the health value of each moth in the initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector; based on the new flame vector, calculating the corresponding cascade flame of the moth by adopting a progressive cascade flame matching algorithm and an exploring type moth flame matching algorithm, and matching the moth with the corresponding cascade flame; when only one survival flame exists, generating a resource allocation scheme of a task request according to the survival flame; the resource abstract architecture of the Internet of things based on the digital object adopts a resource scheduling and allocation algorithm to execute a resource allocation scheme. The method and the device can better approach the optimal solution, and avoid the algorithm from sinking into the local optimal solution.
Description
Technical Field
The application belongs to the technical field of application of the Internet of things, and particularly relates to a method, a system, a terminal and a storage medium for resource allocation of the Internet of things.
Background
With the rapid development of the internet of things technology and the rapid popularization of internet of things equipment in the production and life of people, the internet of things equipment becomes more intelligent in a plurality of application scenes. For example, in intelligent medical treatment, patient health monitoring is performed through the internet of things equipment, automatic parcel delivery is performed through the internet of things equipment in intelligent logistics, and the like.
In real life, internet of things devices are often heterogeneous. For example, different internet of things devices may have different availability times and usage costs. Likewise, the scenes driven by the devices of the internet of things have heterogeneity, such as a scene with sufficient resources, a scene with insufficient resources, and the like. In an actual internet of things application task, each application task may request one or more internet of things devices to accomplish this task. Also, each application task will typically specify a period of demand for each resource required by the task. Along with the proliferation of tasks of the internet of things, how to efficiently allocate resources with isomerism for the tasks with resource requirements, so that the cost generated by meeting the tasks is reduced through a more effective resource allocation mode while the income obtained by meeting the tasks is maximized, and the task allocation method becomes a problem to be solved urgently.
The resource allocation scheme of the internet of things in the prior art mainly starts from two angles, and specifically comprises the following steps:
1. the resource allocation model angle of the Internet of things; for example, zhao et al in [ Zhao, l., wang, j., liu, j., kato, n.,2019.Optimal edge resource allocation in iot-based smart technologies, ieee Network 33,30-35 ], propose two algorithms, EOERA and chura, to optimize computing resources of an internet of things device in a smart city scenario, thereby optimizing average service response time. Sangaiah et al in [ Sangaiah, A.K., hosseina bardi, A.A.R., shareh, M.B., bozorgi Rad, S.Y., zolfagharian, A., chilamkurti, N.,2020.Iot resource allocation and optimization based on heuristic algorithm.Sensors 20,539 ] propose the use of whale optimization algorithms to optimize the allocation of Internet of things computing resources, thereby minimizing communication costs. Tsai et al in [ Tsai, C.W.,2018.Seira:An effective algorithm for iot resource allocation problem.Computer Communications 119,156-166 ] propose an SEIRA algorithm that also focuses on optimization of Internet of things computing and communication resources, thereby optimizing overall communication costs. However, the above-described internet of things resource allocation model has the following drawbacks:
Firstly, the above-mentioned internet of things resource allocation model is only aimed at the internet of things resource allocation in a single scene. Secondly, the above-mentioned resource allocation model of the internet of things aims at the calculation, storage and communication resource allocation in the internet of things equipment, but the resources of the internet of things equipment are not regarded as a whole, and the resource allocation is carried out on the level of the internet of things equipment. Meanwhile, the allocation models do not consider that the task requests for resources are time-limited, and do not adopt a reserved resource allocation mode, so that resource deadlock can occur. In addition, the above-mentioned internet of things resource allocation model does not adopt any unified internet of things equipment resource management framework, and along with the rapid development and evolution of internet of things technology and internet of things equipment, management of various internet of things equipment resources becomes very difficult. Finally, the resource allocation model of the Internet of things only optimizes a certain cost, and does not consider the cost generated by simultaneously optimizing and satisfying the service and the benefit brought by the service.
2. Optimizing the method angle; for example: yang in [ Yang, X.,2010.Firefly algorithm in engineering optimization ] presents a firefly algorithm that simulates the flash pattern of fireflies and their attractive behavior to guide the way fireflies move in the search space. The whale optimization algorithm is presented by mirjallii et al [ mirjallii, s., lewis, a.,2016.The whale optimization algorithm.Advances in engineering software 95,51-67 ]. The algorithm mimics the behavior of the air bubble network of whales during predation, thereby guiding whales to search for a better solution in the solution space. Mirjallii et al in [ Mirjallii, S.2015. Mol-flame optimization algorithm: A novel nature-inspired heuristic paradigm. Knowledgebase-based systems 89,228-249 ], propose a moth fire suppression algorithm that simulates the moth's flight navigation mechanism, thereby guiding the moth to perform an optimal search in solution space. In the optimization method, the performance is the moth fire suppression algorithm, however, the moth fire suppression algorithm still has the following defects:
Firstly, the moth fire suppression algorithm adopts a pure random moth initializing mechanism which does not utilize additional information at all, and does not consider any additional model information, so that the effect is poor. In addition, each moth is matched with only one flame, so that a local optimal solution dilemma is easily caused, and the algorithm cannot effectively approach the optimal solution. Finally, the moth fire suppression algorithm adopts a matching mechanism for uniformly matching the worst flame for all the moth whose corresponding flame is eliminated. The matching mechanism for matching the worst flame is not reasonable, and the worst flame is often far from the optimal solution. Matching multiple moths to the same flame can also greatly cause the algorithm to fall into a local optimal solution, thereby damaging the optimization effect of the algorithm.
Disclosure of Invention
The application provides a method, a system, a terminal and a storage medium for resource allocation of the Internet of things, which aim to at least solve one of the technical problems in the prior art to a certain extent.
In order to solve the above problems, the present application provides the following technical solutions:
an internet of things resource allocation method, comprising:
initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on the received resource allocation task request to obtain an initialized moth vector set;
Evaluating the health value of each moth in the initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector;
based on the new flame vector, calculating the corresponding cascade flame of the moth by adopting a progressive cascade flame matching algorithm and an exploring type moth flame matching algorithm, and matching the moth with the corresponding cascade flame;
when only one survival flame exists, generating a resource allocation scheme of the task request according to the survival flame;
and executing the resource allocation scheme by adopting a resource scheduling and allocation algorithm based on the resource abstraction architecture of the Internet of things of the digital object, and performing resource scheduling and allocation on the task request.
The technical scheme adopted by the embodiment of the application further comprises: the method for initializing the moth vector set of the moth fire suppression algorithm by adopting the target and density sensing algorithm comprises the following specific steps:
for the received task request, respectively calculating the income, the cost and the optimization targets of each task under different resource scenes; the resource scenes comprise a scene with sufficient resources and a scene with insufficient resources;
judging whether the initialized moth vectors in the initialized moth vector set reach the set times of the set number, if not, distributing each task request according to the optimized target size, and repeating the judgment; if the set number of the set times is reached, two initialization moth vectors with the minimum distance are found out from the initialization moth vector set by adopting a target and density sensing algorithm, the mean value vector of the two initialization moth vectors is used as a new moth vector to be added into the initialization moth vector set, and the two initialization moth vectors are deleted from the initialization moth vector set; the length of each initialization moth vector is the number of task requests of resources to be allocated;
Judging whether the initialized moth vectors in the initialized moth vector set reach the set number, if not, continuing to execute the target and density sensing algorithm; and outputting an initialized moth vector set if the set number is reached.
The technical scheme adopted by the embodiment of the application further comprises: the step flame corresponding to the moth is calculated by adopting a progressive step flame matching algorithm and an exploring type moth flame matching algorithm based on the new flame vector comprises the following steps:
judging whether the flame vector corresponding to the moth is eliminated, if not, calculating the cascade flame according to the progressive coefficient by adopting a progressive cascade flame matching algorithm, and matching the moth with the corresponding cascade flame;
if the flame vector corresponding to the moth is eliminated, adopting an exploring moth flame matching algorithm to calculate the cascade flame according to the progressive coefficient, and matching the moth with the corresponding cascade flame.
The technical scheme adopted by the embodiment of the application further comprises: the step flame is calculated according to the progressive coefficient by adopting a progressive step flame matching algorithm, and the matching of the moths and the corresponding step flames is specifically as follows:
Selecting three flames with highest health values from k new flame vectors;
updating the progressive coefficient;
calculating the cascade flame by combining the three flames with the highest progressive coefficient and the highest health value, and matching the moth with the corresponding cascade flame; the matching mode is as follows:
if the number of the current surviving flames is more than or equal to 3, performing flame matching according to the flames corresponding to the moths and the three flames with the highest health values:
where w is a progressive coefficient that increases linearly from 0 to 1 as the iteration progresses, F i Flame corresponding to the moths, F 1 、F 2 、F 3 Respectively three flames with highest health values, F 1 、F 2 、F 3 The flames with the highest health values have the highest weights, which are respectively given with weights of 0.15, 0.1 and 0.05;
when only two flames exist in three flames with highest health values, the matching mode is as follows:
and updating the position of the moth according to the flame matching result.
The technical scheme adopted by the embodiment of the application further comprises: the adoption of the exploratory moth flame matching algorithm calculates the cascade flame according to the progressive coefficient, and the matching of the moth and the corresponding cascade flame is specifically as follows:
selecting three flames with highest health values from k new flame vectors;
randomly selecting live flames F I ；
Updating the progressive coefficient;
calculating a cascade flame by combining the progressive coefficient and the selected survival flame, and matching the moth with the corresponding cascade flame; the matching mode is as follows:
if the current number of the surviving flames is more than or equal to 3, performing moth-flame matching according to the randomly selected surviving flames and three flames with highest health values:
when only two flames exist in the three flames with the highest health value, the matching formula is as follows:
and updating the position of the moth according to the flame matching result.
The technical scheme adopted by the embodiment of the application further comprises: the resource allocation model adopts a resource scheduling and allocation algorithm to execute a resource allocation scheme, and the resource allocation scheme is specifically as follows:
the resource allocation model transmits a task request to a digital object of the Internet of things equipment, the digital object comprises an API module and a communication module, the API module is used for judging whether the Internet of things equipment can meet the task request, and if so, information meeting the task request is transmitted back to the resource allocation model; if the Internet of things equipment cannot meet the task request, the communication module is used for communicating with other similar Internet of things equipment, and the task request is sent to the other similar Internet of things equipment.
The technical scheme adopted by the embodiment of the application further comprises: the resource allocation model adopts a resource scheduling and allocation algorithm to execute a resource allocation scheme, which comprises the following steps:
acquiring a moth vector and a flame vector; the lengths of the moth vector and the flame vector are the number of task requests;
setting a constant ub so that all values in the moth vector and the flame vector are between 0 and ub, and if the values in the moth vector and the flame vector are between the range of [0, ub/2], distributing the task request to a scene with sufficient resources; if the values in the moth vector and the flame vector are between (ub/2, ub), the task request is distributed to the scene of lack of resources;
judging whether all required resources of the task request in an allocated resource scene can be met as required for the task request after being allocated, if so, allocating the requested resources for the task request as required in the allocated resource scene, updating the available conditions of the resources in the allocated resource scene, and calculating the benefits, the costs and the optimization targets of meeting the task in the allocated resource scene; otherwise, if the task request cannot be satisfied as required, the task request is disserviced.
The embodiment of the application adopts another technical scheme that: an internet of things resource allocation system, comprising:
an initialization module: the method comprises the steps of initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on a received resource allocation task request to obtain an initialized moth vector set;
health value evaluation module: the method comprises the steps of evaluating the health value of each moth in the initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector;
moth-flame matching module: the resource allocation scheme is used for calculating the corresponding cascade flame of the moths by adopting a progressive cascade flame matching algorithm and an exploring type moths flame matching algorithm based on the new flame vector, matching the moths with the corresponding cascade flame, and generating the resource allocation scheme of the task request according to the survival flame when only one survival flame exists;
a resource allocation module: the resource allocation scheme is used for executing the resource allocation scheme by adopting a resource scheduling and allocation algorithm based on the resource abstraction architecture of the Internet of things of the digital object, and performing resource scheduling and allocation on the task request.
The embodiment of the application adopts the following technical scheme: a terminal comprising a processor, a memory coupled to the processor, wherein,
The memory stores program instructions for implementing the internet of things resource allocation method;
the processor is used for executing the program instructions stored by the memory to control the resource allocation of the Internet of things.
The embodiment of the application adopts the following technical scheme: a storage medium storing program instructions executable by a processor for performing the internet of things resource allocation method.
Compared with the prior art, the beneficial effect that this application embodiment produced lies in: according to the method, the system, the terminal and the storage medium for allocating the resources of the Internet of things, the abstract framework of the resources of the Internet of things based on the digital objects is introduced into the resource allocation model, and the heterogeneous Internet of things equipment forms the digital object framework by utilizing the concept of the digital objects, so that the management of the resources of the Internet of things equipment and the cooperation between the resources in the resource allocation process are facilitated. When the resource allocation is carried out, the resource allocation model is utilized to simultaneously optimize the benefits generated by meeting the tasks and the costs caused by meeting the tasks, and the optimization targets simultaneously consider the benefits and the costs, so that the method has higher practicability. And the improved moth fire suppression optimizer is adopted to optimize, so that the income and the cost of the task are met as a common optimization target, and the resource allocation model is optimally allocated, so that a better resource allocation scheme of the Internet of things equipment is obtained. Compared with the prior art, the application has the following beneficial effects:
1. The target and the density sensing algorithm are introduced to initialize on the basis of the moth fire suppression algorithm, so that denser moth vectors in the initialized moth vector set can be eliminated, the efficiency of the group-based algorithm is enhanced, the initialization process can be more refined under the guidance of the optimization target, and the problem that the effect of the optimization algorithm is damaged due to random initialization is solved.
2. By introducing a progressive cascade flame matching algorithm and an exploratory moth flame matching algorithm, the method can better balance exploration and discovery in the training process, so that the optimal solution is better approximated, a single matching mechanism is avoided, and the algorithm is less prone to sinking into a local optimal solution.
3. The Internet of things resource abstraction framework based on the digital object utilizes the concept of the digital object to uniformly manage and schedule various Internet of things devices, and covers the diversity of the various Internet of things devices, so that the management of the Internet of things device resources and the cooperative coordination among the resources in the resource allocation process become more convenient, and the resource management difficulty caused by the isomerism of the Internet of things devices is solved.
4. By adopting a reserved type internet of things equipment resource allocation mode under a specific internet of things multi-scene mode, resources can be reasonably allocated under various internet of things scenes with resource isomerism, and when the resources are allocated, the time requirements of tasks on the requested resources are considered, so that the method has stronger practicability, and the occurrence of resource deadlock in the resource allocation can be avoided.
5. And the cost generated by meeting the task is optimized, and meanwhile, the income generated by meeting the task is optimized, so that the optimization target is more comprehensive and has practicability.
Drawings
Fig. 1 is a flowchart of an internet of things resource allocation method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of an initialization process based on a target and density sensing algorithm in an embodiment of the present application;
FIG. 3 is a schematic diagram of an asymptotic cascade flame matching algorithm in an embodiment of the present application;
fig. 4 is a schematic diagram of an exploratory moth flame matching algorithm according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a resource scheduling and allocation algorithm according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an internet of things resource allocation system according to an embodiment of the present application;
fig. 7 is a schematic diagram of a terminal structure according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of a storage medium according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Please refer to fig. 1, which is a flowchart of an internet of things resource allocation method according to an embodiment of the present application. The resource allocation method of the Internet of things comprises the following steps:
S10: receiving a task request for requesting to allocate resources through a resource allocation model, and initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on the received task request to obtain an initialized moth vector set;
in the step, as the moth fire suppression algorithm is a genetic algorithm based on a group, the probability of approaching a global optimal solution is higher by using a plurality of searchers. Therefore, in the initial stage of the algorithm, a set of moth vectors needs to be initialized to generate a set number of moth vectors. According to the invention, the moth vector set of the moth fire suppression algorithm is initialized based on the target and the density sensing algorithm, the algorithm guides the initialization by using the optimization target as priori knowledge, and the remote degree of the initialized moth vector is increased according to the density sensing characteristic of the optimization target, so that the efficiency of the algorithm is improved. Specifically, as shown in fig. 2, an initialization process based on the target and density sensing algorithm in the embodiment of the application is shown, and the initialization process includes the following steps:
s11: for the received task request, respectively calculating the income, the cost and the optimization targets of each task under different resource scenes;
The embodiment of the application includes two resource scenes with resource isomerism, namely a scene R with sufficient resources and a scene S with insufficient resources, so as to reflect the situation of uneven resource allocation which usually occurs in the actual application scene. For example, in a smart medical scenario, a large hospital is a resource-rich scenario, and a community clinic is a resource-starved scenario; in the wisdom logistics scenario, the central warehouse is a resource-rich scenario, while the regional post office is a resource-starved scenario.
Further, the calculation modes of the income, the cost and the optimization target of the task under different resource scenes are specifically as follows:
in the resource allocation model, M different resources are considered altogether, defined as:
wherein, the liquid crystal display device comprises a liquid crystal display device,
representing the mth resource in scenario X. Every resource has->
Copy of resource, resource->
The j-th copy of (2) is:
for each copy of the resource, it has its start availability time and end availability time, respectively, noted as
And->
The available period of each resource copy is recorded as +.>
It is initially expressed as:
for example, a mobile health monitor may have a start availability time of 9 early and an end availability time of 6 late, and may have an availability period of 9 early to 6 late. If the mobile health monitor is requested to be occupied between 10 and 11 early points, the available time period of the device is updated to 9 to 10 early points and 11 to 6 late points.
Let the resource allocation model consider N tasks, denoted S n Each task can request at most K Internet of things devices to be
Denoted as task S n Requested set of resources, +.>
Marked as task->
I-th resource request in +.>
Representing task S n The number of resources requested. Thus, there are:
at task S n In each of its resource requests
Will specify a start time and end causeThe time of use is recorded as
And->
For example, the resource request of a certain task is: and using the Internet of things equipment from 9 to 9 in the early stage, namely, the starting use time designated by the request is 9 in the early stage, and the ending use time is 9 in the half stage.
Each task, when satisfied in scenario X, generates a benefit, denoted P (S n ，X)，
Thus, under allocation scheme a, the total yield generated is defined as:
the unit time of using each type of Internet of things equipment can generate a cost which is recorded as
The use period length len of each resource is defined as: />
The total cost incurred under allocation scheme a is defined as:
wherein, the liquid crystal display device comprises a liquid crystal display device,
at task S n Is equal to 1 when allocated in a resource-efficient scenario R, otherwise equal to 0.
Because the internet of things equipment has isomerism, namely the internet of things equipment has different available quantity and available time, different internet of things equipment has different use cost. The usage costs may also include other costs such as energy usage costs, such that heterogeneous cost definitions are commonplace.
The optimization objective of the resource allocation model in the embodiment of the application simultaneously considers the cost and the benefit generated by meeting the task, so that the final optimization objective is to optimize the benefit and simultaneously minimize the cost. The final optimization objective is:
where w is a trade-off coefficient, is a negative value,
representing the optimized resource allocation scheme.
According to the method and the device for initializing the moth vector set, the initialization process of initializing the moth vector set is guided by utilizing the benefits, the costs and the optimization targets, so that the initialization mode is more reasonable, and the effectiveness of the algorithm can be effectively improved.
S12: judging whether the number of the initialized moth vectors in the initialized moth vector set reaches a set multiple of the set number, and executing S13 if the number of the initialized moth vectors does not reach the set multiple of the set number; otherwise, executing S14;
the length of each initialization moth vector is the number of tasks to be allocated with resources. When generating the initialization moth vector, the initialization moth vector is generated to be 1.5 times more than the set quantity, and the specific times can be set according to practical application.
S13: distributing each task request according to the scene optimization target size, and re-executing S12;
s14: finding out two initializing moth vectors with minimum distance (namely closest distance) from the initializing moth vector set by adopting a target and density sensing algorithm, adding the mean value vector of the two initializing moth vectors into the initializing moth vector set as a new moth vector, and deleting the two initializing moth vectors from the initializing moth vector set;
S15: judging whether the initialized moth vectors in the initialized moth vector set reach the set number, and if the initialized moth vectors do not reach the set number, re-executing S14; if the set number is reached, S16 is executed;
s16: outputting an initialized moth vector set.
Based on the above, in the embodiment of the present application, by adopting the target and density sensing algorithm to continuously find two initializing moth vectors with the smallest distance from the generated initializing moth vector set, adding the average value vector of the two initializing moth vectors into the initializing moth vector set, deleting the two initializing moth vectors from the set, thereby detecting the initializing moth vector set with larger density and closer distance, and merging and removing the initializing moth vector set until the number of the initializing moth vectors in the initializing moth vector set reaches the set number, so that the moth vectors in the final initializing moth vector set are as far away from each other as possible, thereby enhancing the exploration capability of the algorithm.
S20: evaluating the health value of each moth in the initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector;
s30: updating the number of flame vectors and judging whether there is only one surviving flame, if there are more surviving flames, executing S40; if there is only one live flame vector, execute S70;
S40: judging whether the flame vector corresponding to the moth is eliminated, if not, executing S50; otherwise, executing S60;
s50: calculating the corresponding cascade flame of the moths according to the progressive coefficients by using a progressive cascade flame matching algorithm, matching the moths with the corresponding cascade flame, and re-executing S30;
in this step, as shown in fig. 3, a schematic diagram of an asymptotic cascade flame matching algorithm in the embodiment of the present application is shown, which specifically includes the following steps:
s51: selecting three flames with highest health values from k new flame vectors;
s52: updating the progressive coefficient;
s53: calculating the cascade flame by combining the three flames with the highest progressive coefficient and the highest health value, and matching the moth with the corresponding cascade flame;
in order to solve the problem of sinking into a local optimal solution in an original moth fire suppression algorithm, the embodiment of the invention adopts an asymptotic cascade flame matching mode, and the specific matching mode is as follows:
if the number of the current surviving flames is more than or equal to 3, performing flame matching according to the flames corresponding to the moths and the three flames with the highest health values, wherein a matching formula is as follows:
wherein w is a progressive coefficient, F i Flame corresponding to the moths, F 1 、F 2 、F 3 Respectively three flames with highest health values, F 1 、F 2 、F 3 The flames with the highest health values have the highest weights (i.e., influence) to represent the rank hierarchy each flame has, respectively given weights of 0.15, 0.1, and 0.05.
When only two flames exist in the three flames with the highest health value, the matching formula is as follows:
in the asymptotic cascade flame matching algorithm, the asymptotic coefficient w is a coefficient which increases linearly from 0 to 1 as the iteration progresses, and the use of the asymptotic coefficient enables the influence of the flame with the top three health values on the moth matching to increase gradually as the iteration progresses. The reasons are as follows: in the initial stage of iteration, the flames with the top three health values are not necessarily very close to the global optimal solution, so that the influence degree of the three flames on the matching of the moths and the flames can be reduced by using a smaller progressive coefficient w, and the search of the moths on the search space is encouraged. The flame with the top three health value ranks becomes closer to the optimal solution as the iteration is continuously carried out, and at this time, the progressive coefficient which is continuously increased along with the iteration is gradually used for enhancing the influence degree of the three flames on the moth flame matching process, so that the algorithm is gradually converted into the mining stage from the exploring stage. Therefore, the use of the asymptotic coefficient w enables the algorithm to effectively balance exploration and discovery when the asymptotic cascade flame matching algorithm is adopted, and the optimal solution is better approximated. After updating the asymptotic coefficients, the algorithm calculates the rank flames that each moth matches and matches the moth.
S54: and updating the position of the moth according to the flame matching result.
S60: using an exploring type moth flame matching algorithm, calculating corresponding cascade flames of the moth according to the progressive coefficient, and matching the moth with the corresponding cascade flames;
in this step, as shown in fig. 4, a schematic diagram of an exploratory moth flame matching algorithm according to an embodiment of the present application specifically includes the following steps:
s61: selecting three flames with highest health values from k new flame vectors;
s62: randomly selecting survival flames;
in order to prevent all the moths losing the corresponding flame from uniformly matching with the worst surviving moths, the embodiment of the application adopts a exploring mode to randomly select the surviving flame as the flame corresponding to the moths losing the flame, and the flame is marked as F I 。
S63: updating the progressive coefficient;
s64: calculating a cascade flame by combining the progressive coefficient and the selected survival flame, and matching the moth with the corresponding cascade flame;
in order to solve the problem of the original moth fire suppression algorithm that the local optimal solution is trapped, the embodiment of the invention adopts a search-level flame matching mode, and the search-level flame matching algorithm is similar to an asymptotic-level flame matching algorithm, and uses the randomly searched flames to influence the asymptotic-level flames to be used as matching flames corresponding to the moth. The specific matching mode is as follows:
If the current number of the surviving flames is more than or equal to 3, performing moth-flame matching according to the randomly selected surviving flames and three flames with highest health values, wherein a matching formula is as follows:
when only two flames exist in the three flames with the highest health value, the matching formula is as follows:
s65: and updating the position of the moth according to the flame matching result.
In the above steps, the algorithm continuously repeats the above operations, and in the iterative process, the flame quantity parameter k gradually decreases and finally decreases to 1, that is, only one flame vector can survive, and then the algorithm outputs the finally survived flame vector and guides the resource scheduling and allocation module to perform resource scheduling and allocation.
S70: generating a resource allocation scheme of the task request according to the final survival flame;
s80: the resource allocation model adopts a resource scheduling and allocation algorithm to execute a resource allocation scheme based on the resource abstraction architecture of the Internet of things of the digital object, and performs resource scheduling and allocation on the received task request;
in the step, the Internet of things equipment with different characteristics is digitized and unified by adopting an Internet of things resource abstraction architecture based on the digital object, so that communication and distribution with the Internet of things equipment and cooperation between the equipment are facilitated. The abstract architecture of the Internet of things equipment based on the digital object comprises the following concrete steps: each Internet of things device is abstracted into a digital object, so that management and collaboration of the Internet of things device are converted into management and collaboration of the digital object with unified architecture, and the isomerism of the Internet of things device is covered through the abstraction of the digital object. For the Internet of things equipment in different resource scenes, when the Internet of things equipment is used as the Internet of things equipment to communicate with the resource allocation algorithm, the resource allocation algorithm is communicated with the digital object abstract architecture of the Internet of things equipment, and the Internet of things equipment cannot be affected by the difference of the Internet of things equipment.
Specifically, for each digital object, two modules, namely an API module and a collaboration module, are provided, where the API module is responsible for reflecting the current state of each internet of things device, including whether the device is available, the cost generated by using the device, and whether a specified task request is acceptable. The collaboration module is used for being responsible for collaborative communication between the Internet of things equipment and other similar Internet of things equipment. When the resource allocation model receives a task request, the task request is transmitted to a digital object of the internet of things device. And judging whether the Internet of things equipment can meet the task request or not through an API module of the digital object, and if so, transmitting the information meeting the task request back to the resource allocation model through the digital object. Otherwise, if the internet of things equipment cannot meet the task request, the internet of things equipment communicates with other internet of things equipment of the same kind through a communication module of the digital object of the internet of things equipment, and the task request is sent to the other internet of things equipment of the same kind so as to try to meet the task request, so that the internet of things equipment of the same kind can cooperate in the process of resource allocation.
Further, as shown in fig. 5, a schematic diagram of a resource scheduling and allocation algorithm in the embodiment of the present application specifically includes the following steps:
S81: acquiring a moth vector and a flame vector;
s82: sequentially distributing each task according to the resource distribution schemes in the moth vector and the flame vector;
the resource allocation scheme specifically comprises the following steps: a constant ub is set so that all values in the moth vector and the flame vector are between 0-ub, and the lengths of the moth vector and the flame vector are the number of tasks. Taking each value in the moth vector and the flame vector as an allocation mode for the task, and if the value in the moth vector and the flame vector is in the range of [0, ub/2], allocating the task to a scene with sufficient resources; for example, if ub is 10, the value of the moth vector is [3,7,2,9], which indicates that the value of the moth vector is 3, the value of the task 2 is 7, the value of the task 3 is 2, the value of the task 4 is 9, since the value of the task 1 and the task 3 is between [0,10/2], the task 1 and the task 3 are allocated to the scene with sufficient resources, and since the value of the task 2 and the task 4 is between (10/2, 10], the task 2 and the task 4 are allocated to the scene with insufficient resources.
S83: judging whether the task is allocated in a scene with sufficient resources or a scene with insufficient resources, if the task is allocated in a scene with sufficient resources, executing S84; if the task is allocated in the resource starvation scenario, S87 is performed;
s84: judging whether all required resources of the task can be met as required in a scene of sufficient resources, and executing S85 if the required resources can be met as required; otherwise, S86 is performed;
s85: the requested resources are distributed for the task on demand in a scene with sufficient resources, the resource availability in the scene with sufficient resources is updated, and the income, the cost and the optimization target of the task are calculated to be met in the scene with sufficient resources;
s86: to avoid deadlock, the task is taken out of service;
s87: judging whether all required resources of the task can be met as required in a resource deficiency scene, and executing S88 if the required resources can be met as required; otherwise, S86 is performed;
s88: the requested resources are allocated on demand for the task in a resource starved scenario, the resource availability in the resource starved scenario is updated, and the benefits, costs and optimization goals of the task are calculated to be met in the resource starved scenario.
Based on the above, the embodiment of the application adopts a reserved allocation mode, for the tasks allocated in different scenes, it is necessary to determine whether all the resources requested by the task in the scene can be allocated on time as required, if so, the requested resources can be allocated on time as required for the task in the scene, otherwise, the task is out of service, and the occurrence of resource deadlock caused by that a certain task only obtains partial resource allocation is effectively avoided. After the resource allocation is carried out on the task, the resource availability condition in the scene is updated, and the income, cost and optimization target meeting the task are calculated. Finally, the optimization objective under such resource allocation scheme is returned.
Based on the above, the method for allocating resources of the internet of things according to the embodiment of the application introduces an abstract frame of resources of the internet of things based on the digital object in the resource allocation model, and utilizes the concept of the digital object, and the internet of things equipment with isomerism forms the digital object architecture, so that the management of resources of the internet of things equipment and the cooperation between the resources in the resource allocation process are facilitated. When the resource allocation is carried out, the resource allocation model is utilized to simultaneously optimize the benefits generated by meeting the tasks and the costs caused by meeting the tasks, and the optimization targets simultaneously consider the benefits and the costs, so that the method has higher practicability. And the improved moth fire suppression optimizer is adopted to optimize, so that the income and the cost of the task are met as a common optimization target, and the resource allocation model is optimally allocated, so that a better resource allocation scheme of the Internet of things equipment is obtained. Compared with the prior art, the application has the following beneficial effects:
1. the target and the density sensing algorithm are introduced to initialize on the basis of the moth fire suppression algorithm, so that denser moth vectors in the initialized moth vector set can be eliminated, the efficiency of the group-based algorithm is enhanced, the initialization process can be more refined under the guidance of the optimization target, and the problem that the effect of the optimization algorithm is damaged due to random initialization is solved.
2. By introducing a progressive cascade flame matching algorithm and an exploratory moth flame matching algorithm, the method can better balance exploration and discovery in the training process, so that the optimal solution is better approximated, a single matching mechanism is avoided, and the algorithm is less prone to sinking into a local optimal solution.
3. The Internet of things resource abstraction framework based on the digital object utilizes the concept of the digital object to uniformly manage and schedule various Internet of things devices, and covers the diversity of the various Internet of things devices, so that the management of the Internet of things device resources and the cooperative coordination among the resources in the resource allocation process become more convenient, and the resource management difficulty caused by the isomerism of the Internet of things devices is solved.
4. By adopting a reserved type internet of things equipment resource allocation mode under a specific internet of things multi-scene mode, resources can be reasonably allocated under various internet of things scenes with resource isomerism, and when the resources are allocated, the time requirements of tasks on the requested resources are considered, so that the method has stronger practicability, and the occurrence of resource deadlock in the resource allocation can be avoided.
5. And the cost generated by meeting the task is optimized, and meanwhile, the income generated by meeting the task is optimized, so that the optimization target is more comprehensive and has practicability.
Fig. 6 is a schematic structural diagram of an internet of things resource allocation system according to an embodiment of the present application. The internet of things resource allocation system 40 of the embodiment of the present application includes:
initialization module 41: the method comprises the steps of initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on a received resource allocation task request to obtain an initialized moth vector set;
health value assessment module 42: the method comprises the steps of evaluating the health value of each moth in an initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector;
moth-flame matching module 43: the resource allocation scheme is used for calculating the corresponding cascade flame of the moths by adopting a progressive cascade flame matching algorithm and an exploring type moths flame matching algorithm based on the new flame vector, matching the moths with the corresponding cascade flame, and generating a task request according to the survival flame when only one survival flame exists;
resource allocation module 44: the resource allocation method is used for an Internet of things resource abstraction architecture based on the digital object, a resource allocation scheme is executed by adopting a resource scheduling and allocation algorithm, and resource scheduling and allocation are carried out on task requests.
Fig. 7 is a schematic diagram of a terminal structure according to an embodiment of the present application. The terminal 50 includes a processor 51, a memory 52 coupled to the processor 51.
The memory 52 stores program instructions for implementing the above-described internet of things resource allocation method.
The processor 51 is configured to execute program instructions stored in the memory 52 to control the allocation of resources of the internet of things.
The processor 51 may also be referred to as a CPU (Central Processing Unit ). The processor 51 may be an integrated circuit chip with signal processing capabilities. Processor 51 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Fig. 8 is a schematic structural diagram of a storage medium according to an embodiment of the present application. The storage medium of the embodiment of the present application stores a program file 61 capable of implementing all the methods described above, where the program file 61 may be stored in the storage medium in the form of a software product, and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the methods in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, an optical disk, or other various media capable of storing program codes, or a terminal device such as a computer, a server, a mobile phone, a tablet, or the like.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The method for allocating the resources of the Internet of things is characterized by comprising the following steps:
initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on the received resource allocation task request to obtain an initialized moth vector set;
evaluating the health value of each moth in the initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector;
based on the new flame vector, calculating the corresponding cascade flame of the moth by adopting a progressive cascade flame matching algorithm and an exploring type moth flame matching algorithm, and matching the moth with the corresponding cascade flame;
When only one survival flame exists, generating a resource allocation scheme of the task request according to the survival flame;
and executing the resource allocation scheme by adopting a resource scheduling and allocation algorithm based on the resource abstraction architecture of the Internet of things of the digital object, and performing resource scheduling and allocation on the task request.
2. The method for allocating resources of the internet of things according to claim 1, wherein initializing the moth vector set of the moth fire suppression algorithm by using the target and density sensing algorithm is specifically as follows:
for the received task request, respectively calculating the income, the cost and the optimization targets of each task under different resource scenes; the resource scenes comprise a scene with sufficient resources and a scene with insufficient resources;
judging whether the initialized moth vectors in the initialized moth vector set reach the set times of the set number, if not, distributing and repeatedly judging each task request according to the optimized target size; if the set number of the set times is reached, two initialization moth vectors with the minimum distance are found out from the initialization moth vector set by adopting a target and density sensing algorithm, the mean value vector of the two initialization moth vectors is used as a new moth vector to be added into the initialization moth vector set, and the two initialization moth vectors are deleted from the initialization moth vector set; the length of each initialization moth vector is the number of task requests of resources to be allocated;
Judging whether the initialized moth vectors in the initialized moth vector set reach the set number, if not, continuing to execute the target and density sensing algorithm; and outputting an initialized moth vector set if the set number is reached.
3. The method of claim 2, wherein the calculating the cascade flame corresponding to the moth using a progressive cascade flame matching algorithm and a heuristic moth flame matching algorithm based on the new flame vector comprises:
judging whether the flame vector corresponding to the moth is eliminated, if not, calculating the cascade flame according to the progressive coefficient by adopting a progressive cascade flame matching algorithm, and matching the moth with the corresponding cascade flame;
if the flame vector corresponding to the moth is eliminated, adopting an exploring moth flame matching algorithm to calculate the cascade flame according to the progressive coefficient, and matching the moth with the corresponding cascade flame.
4. The internet of things resource allocation method according to claim 3, wherein the step flame is calculated according to a progressive coefficient by adopting a progressive step flame matching algorithm, and the matching of the moths with the corresponding step flames is specifically:
Selecting three flames with highest health values from k new flame vectors;
updating the progressive coefficient;
calculating the cascade flame by combining the three flames with the highest progressive coefficient and the highest health value, and matching the moth with the corresponding cascade flame; the matching mode is as follows:
if the number of the current surviving flames is more than or equal to 3, performing flame matching according to the flames corresponding to the moths and the three flames with the highest health values:
where w is a progressive coefficient that increases linearly from 0 to 1 as the iteration progresses, F i Flame corresponding to the moths, F 1 、F 2 、F 3 Respectively three flames with highest health values, F 1 、F 2 、F 3 The flames with the highest health values have the highest weights, which are respectively given with weights of 0.15, 0.1 and 0.05;
when only two flames exist in three flames with highest health values, the matching mode is as follows:
and updating the position of the moth according to the flame matching result.
5. The method for allocating resources of the internet of things according to claim 4, wherein the step flames are calculated according to the progressive coefficients by adopting an exploratory moth flame matching algorithm, and the matching of the moth and the corresponding step flames is specifically as follows:
selecting three flames with highest health values from k new flame vectors;
Randomly selecting live flames F I ；
Updating the progressive coefficient;
calculating a cascade flame by combining the progressive coefficient and the selected survival flame, and matching the moth with the corresponding cascade flame; the matching mode is as follows:
if the current number of the surviving flames is more than or equal to 3, performing moth-flame matching according to the randomly selected surviving flames and three flames with highest health values:
when only two flames exist in the three flames with the highest health value, the matching formula is as follows:
and updating the position of the moth according to the flame matching result.
6. The method for allocating resources of the internet of things according to any one of claims 1 to 5, wherein the resource allocation model adopts a resource scheduling and allocation algorithm to execute a resource allocation scheme specifically as follows:
the resource allocation model transmits a task request to a digital object of the Internet of things equipment, the digital object comprises an API module and a communication module, the API module is used for judging whether the Internet of things equipment can meet the task request, and if so, information meeting the task request is transmitted back to the resource allocation model; if the Internet of things equipment cannot meet the task request, the communication module is used for communicating with other similar Internet of things equipment, and the task request is sent to the other similar Internet of things equipment.
7. The method for allocating resources of the internet of things according to claim 6, wherein the resource allocation model adopts a resource scheduling and allocation algorithm to execute a resource allocation scheme specifically comprising:
acquiring a moth vector and a flame vector; the lengths of the moth vector and the flame vector are the number of task requests;
setting a constant ub so that all values in the moth vector and the flame vector are between 0 and ub, and if the values in the moth vector and the flame vector are between the range of [0, ub/2], distributing the task request to a scene with sufficient resources; if the values in the moth vector and the flame vector are between (ub/2, ub), the task request is distributed to the scene of lack of resources;
judging whether all required resources of the task request in an allocated resource scene can be met as required for the task request after being allocated, if so, allocating the requested resources for the task request as required in the allocated resource scene, updating the available conditions of the resources in the allocated resource scene, and calculating the benefits, the costs and the optimization targets of meeting the task in the allocated resource scene; otherwise, if the task request cannot be satisfied as required, the task request is disserviced.
8. The utility model provides an thing networking resource allocation system which characterized in that includes:
an initialization module: the method comprises the steps of initializing a moth vector set of a moth fire suppression algorithm by adopting a target and density sensing algorithm based on a received resource allocation task request to obtain an initialized moth vector set;
health value evaluation module: the method comprises the steps of evaluating the health value of each moth in the initialized moth vector set, and selecting the first k moth with the highest health value as a new flame vector;
moth-flame matching module: the resource allocation scheme is used for calculating the corresponding cascade flame of the moths by adopting a progressive cascade flame matching algorithm and an exploring type moths flame matching algorithm based on the new flame vector, matching the moths with the corresponding cascade flame, and generating the resource allocation scheme of the task request according to the survival flame when only one survival flame exists;
a resource allocation module: the resource allocation scheme is used for executing the resource allocation scheme by adopting a resource scheduling and allocation algorithm based on the resource abstraction architecture of the Internet of things of the digital object, and performing resource scheduling and allocation on the task request.
9. A terminal comprising a processor, a memory coupled to the processor, wherein,
The memory stores program instructions for implementing the internet of things resource allocation method of any one of claims 1-7;
the processor is used for executing the program instructions stored by the memory to control the resource allocation of the Internet of things.
10. A storage medium storing program instructions executable by a processor for performing the internet of things resource allocation method of any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210026373.0A CN114390102B (en) | 2022-01-11 | 2022-01-11 | Internet of things resource allocation method, system, terminal and storage medium |
| PCT/CN2022/140066 WO2023134403A1 (en) | 2022-01-11 | 2022-12-19 | Internet of things resource allocation method and system, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210026373.0A CN114390102B (en) | 2022-01-11 | 2022-01-11 | Internet of things resource allocation method, system, terminal and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114390102A CN114390102A (en) | 2022-04-22 |
| CN114390102B true CN114390102B (en) | 2023-07-14 |
Family
ID=81201951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210026373.0A Active CN114390102B (en) | 2022-01-11 | 2022-01-11 | Internet of things resource allocation method, system, terminal and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114390102B (en) |
| WO (1) | WO2023134403A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113052514B (en) * | 2021-05-21 | 2021-10-29 | 深圳市海柔创新科技有限公司 | Method, apparatus, device, storage medium and program product for resource allocation |
| CN114390102B (en) * | 2022-01-11 | 2023-07-14 | 中国科学院深圳先进技术研究院 | Internet of things resource allocation method, system, terminal and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106802822A (en) * | 2016-12-30 | 2017-06-06 | 南京邮电大学 | A kind of cloud data center cognitive resources dispatching method based on moth algorithm |
| CN113222096A (en) * | 2021-04-30 | 2021-08-06 | 桂林理工大学 | Improved particle swarm algorithm for cloud computing task scheduling |
| CN113656153A (en) * | 2021-06-29 | 2021-11-16 | 桂林理工大学 | Improved artificial bee colony algorithm for cloud computing task scheduling |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108062585A (en) * | 2017-12-29 | 2018-05-22 | 航天科工智慧产业发展有限公司 | A kind of method that Function Extreme value is calculated based on a flying moth darts into the fire algorithm |
| US11129196B2 (en) * | 2018-05-11 | 2021-09-21 | Qualcomm Incorporated | Shared channel design around reserved resources |
| CN109598296A (en) * | 2018-11-26 | 2019-04-09 | 长安大学 | One kind is based on a flying moth darts into the fire the K mean cluster method of improvement |
| CN114390102B (en) * | 2022-01-11 | 2023-07-14 | 中国科学院深圳先进技术研究院 | Internet of things resource allocation method, system, terminal and storage medium |
-
2022
- 2022-01-11 CN CN202210026373.0A patent/CN114390102B/en active Active
- 2022-12-19 WO PCT/CN2022/140066 patent/WO2023134403A1/en unknown
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106802822A (en) * | 2016-12-30 | 2017-06-06 | 南京邮电大学 | A kind of cloud data center cognitive resources dispatching method based on moth algorithm |
| CN113222096A (en) * | 2021-04-30 | 2021-08-06 | 桂林理工大学 | Improved particle swarm algorithm for cloud computing task scheduling |
| CN113656153A (en) * | 2021-06-29 | 2021-11-16 | 桂林理工大学 | Improved artificial bee colony algorithm for cloud computing task scheduling |
Non-Patent Citations (2)
| Title |
|---|
| 基于改进飞蛾优化算法的云计算任务调度策略;李宏伟;;太原学院学报(自然科学版)(第01期);全文 * |
| 基于自适应交叉变异的飞蛾算法云计算任务调度策略;李宏伟;;赤峰学院学报(自然科学版)(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023134403A1 (en) | 2023-07-20 |
| CN114390102A (en) | 2022-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114390102B (en) | Internet of things resource allocation method, system, terminal and storage medium | |
| CN111625361B (en) | Joint learning framework based on cooperation of cloud server and IoT (Internet of things) equipment | |
| CN113242568B (en) | Task unloading and resource allocation method in uncertain network environment | |
| Wang et al. | Task scheduling algorithm based on improved firework algorithm in fog computing | |
| CN108804227B (en) | Method for computing-intensive task unloading and optimal resource allocation based on mobile cloud computing | |
| CN111459628B (en) | Spark platform task scheduling method based on improved quantum ant colony algorithm | |
| CN111382868B (en) | Neural network structure searching method and device | |
| JP2022137182A (en) | Federated learning method, device, equipment and storage medium | |
| CN110968426B (en) | Edge cloud collaborative k-means clustering model optimization method based on online learning | |
| Kaur et al. | Deep‐Q learning‐based heterogeneous earliest finish time scheduling algorithm for scientific workflows in cloud | |
| CN114787824A (en) | Combined hybrid model | |
| KR20130020050A (en) | Apparatus and method for managing bucket range of locality sensitivie hash | |
| CN110347515B (en) | Resource optimization allocation method suitable for edge computing environment | |
| CN112463337B (en) | Workflow task migration method used in mobile edge computing environment | |
| Muthusamy et al. | Cluster-based task scheduling using K-means clustering for load balancing in cloud datacenters | |
| CN113821318A (en) | Internet of things cross-domain subtask combined collaborative computing method and system | |
| CN115330189A (en) | Workflow optimization scheduling method based on improved moth flame algorithm | |
| CN114741955A (en) | Multi-objective optimization task scheduling method based on security cloud | |
| CN117670005A (en) | Super-computing internet multi-objective workflow optimization method and system based on ant colony algorithm | |
| CN113360245A (en) | Internet of things equipment task downloading method based on mobile cloud computing deep reinforcement learning | |
| WO2022267765A1 (en) | Keep-alive method for video application, and electronic device | |
| CN115185660A (en) | Unloading and buffer storage method and system for MAR task in multi-access edge calculation | |
| CN114691327A (en) | Multi-objective group intelligent optimization method and system for two-stage task scheduling | |
| CN116450658A (en) | Node cluster gain-based maximum block storage method and device | |
| WO2021106176A1 (en) | Software deployment system, software deployment apparatus, software deployment method, basepoint computer, and program |
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 |