CN111324444A - Cloud computing task scheduling method and device - Google Patents

Cloud computing task scheduling method and device Download PDF

Info

Publication number
CN111324444A
CN111324444A CN202010212709.3A CN202010212709A CN111324444A CN 111324444 A CN111324444 A CN 111324444A CN 202010212709 A CN202010212709 A CN 202010212709A CN 111324444 A CN111324444 A CN 111324444A
Authority
CN
China
Prior art keywords
cloud computing
task
pheromone
pheromone concentration
node
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.)
Granted
Application number
CN202010212709.3A
Other languages
Chinese (zh)
Other versions
CN111324444B (en
Inventor
谭超奖
徐康康
杨海东
朱成就
印四华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong University of Technology
Original Assignee
Guangdong University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong University of Technology filed Critical Guangdong University of Technology
Priority to CN202010212709.3A priority Critical patent/CN111324444B/en
Publication of CN111324444A publication Critical patent/CN111324444A/en
Application granted granted Critical
Publication of CN111324444B publication Critical patent/CN111324444B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5072—Grid computing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a cloud computing task scheduling method and device, wherein ant colony path searching is carried out based on heuristic factors obtained through calculation according to performance parameters of cloud computing equipment and task load states of the cloud computing equipment, load balance of the cloud computing equipment is introduced and serves as node searching conditions of an ant colony algorithm model, so that a task scheduling scheme based on the cloud computing equipment load states is obtained according to an optimal solution output by the model, and the technical problem that energy consumption is overhigh due to the fact that the load balance is not considered in the existing cloud computing task scheduling mode is solved.

Description

Cloud computing task scheduling method and device
Technical Field
The application relates to the technical field of cloud computing, in particular to a cloud computing task scheduling method and device.
Background
At present, with the rapid development and the increasing demand of the internet, cloud computing services are promoted to be widely applied to the related fields of the internet. The data center is a cloud computing unified management and resource scheduling service platform, the scale of the data center is increasingly huge with the development of a cloud computing technology, but meanwhile, the problems of task scheduling and energy consumption of the data center are increasingly shown, and the data center becomes a bottleneck for limiting the further development of the cloud computing technology.
At present, a cloud computing task scheduling method is a rotating wheel scheduling algorithm, the algorithm adopts a mode of polling and calculating nodes, and then tasks can be sequentially deployed on the nodes which are in accordance with task execution requirements. It is a stateless schedule, i.e. it is treated as one for all tasks, with no priority. Before the algorithm executes, it is necessary to give a specified time that all tasks waiting to execute can take. If the task is not completely executed within the time period, the execution of the task is suspended and moved to the end of the queue, and the CPU is dispatched to another task. If the task is blocked or completed within the current time period, the CPU should switch to continue processing at this time.
The cycle scheduling algorithm does not consider other objective factors such as energy consumption, load balance and the like in the task scheduling process, so that the average power of the system is kept to be maximum, the power consumption is serious, and the energy consumption of the system is greatly increased.
Disclosure of Invention
The application provides a cloud computing task scheduling method and device, which are used for solving the technical problem of overhigh energy consumption caused by the fact that load balance is not considered in the existing cloud computing task scheduling mode.
In view of this, a first aspect of the present application provides a cloud computing task scheduling method, including:
acquiring a task to be scheduled;
according to the task to be scheduled and the preset initial pheromone concentration, node path searching operation is carried out through a preset ant colony algorithm model to obtain a node path searching result, wherein the node path searching result comprises the following steps: actual pheromone concentration between nodes and path length between nodes, wherein a heuristic factor of the ant colony algorithm model is a coefficient obtained by calculation according to the performance parameters of the cloud computing equipment and the task load state of the cloud computing equipment;
updating the actual pheromone concentration according to the node path searching result;
and determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to carry out task scheduling according to the optimal path result.
Optionally, the method further comprises:
according to the performance parameters and the task load information of the cloud computing equipment, a load coefficient of the cloud computing equipment is obtained through a preset cloud computing equipment load computing formula;
inverting the load coefficient of the cloud computing equipment to obtain a heuristic factor of the ant colony algorithm model;
the cloud computing equipment load computing formula specifically comprises:
Figure BDA0002423361100000021
in the formula, the tlength(Vj) The total length of the task already running on the cloud computing device j, tlength_iFor the length of task i, said Vcomp_jIs the performance of the cloud computing device j, the task represents the total number of tasks already running on the cloud computing device j, and the LBijAnd calculating the load coefficient of the cloud computing equipment.
Optionally, the performing, according to the task to be scheduled and a preset initial pheromone concentration, a node path search operation through a preset ant colony algorithm model to obtain a node path search result specifically includes:
according to the task to be scheduled and the initial pheromone concentration, node path searching operation is carried out through a pseudorandom proportion transition probability calculation formula in a preset ant colony algorithm model to obtain a node path searching result, wherein the pseudorandom proportion transition probability calculation formula specifically comprises the following steps:
Figure BDA0002423361100000022
in the formula, ρ0(0≤ρ0≦ 1) is a constant value, ρ is obedient [0,1 ≦]Uniformly distributing the random value, τij(t) is the actual pheromone concentration between node i and node j, ηijIs the heuristic factor.
Optionally, the updating the actual pheromone concentration according to the node path search result specifically includes:
according to the pheromone concentration between the nodes in the node path search result and the path length between the nodes, the actual pheromone concentration is updated through a preset pheromone reward punishment updating formula, wherein the pheromone reward punishment updating formula specifically comprises the following steps:
Figure BDA0002423361100000031
in the formula (I), the
Figure BDA0002423361100000032
Represents the updated actual pheromone concentration, the
Figure BDA0002423361100000033
Representing the actual pheromone concentration before updating, wherein omega is an added reward and punishment coefficient, and LcurFor the total length of the currently searched path, LlastThe Q is the total length of the last searched path, and is the total amount of pheromone released by ants on all paths.
Optionally, the method further comprises:
performing mean operation according to the obtained performance parameters of the cloud computing equipment to obtain average performance parameters, and obtaining the initial pheromone concentration according to the ratio of the performance parameters to the average performance parameters.
A second aspect of the present application provides a cloud computing task scheduling device, including:
the parameter acquiring unit is used for acquiring a task to be scheduled;
the ant colony search execution unit is used for carrying out node path search operation through a preset ant colony algorithm model according to the task to be scheduled and a preset initial pheromone concentration to obtain a node path search result, wherein the node path search result comprises: actual pheromone concentration between nodes and path length between nodes, wherein a heuristic factor of the ant colony algorithm model is a coefficient obtained by calculation according to the performance parameters of the cloud computing equipment and the task load state of the cloud computing equipment;
the pheromone updating unit is used for updating the actual pheromone concentration according to the node path searching result;
and the optimal solution output unit is used for determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to carry out task scheduling according to the optimal path result.
Optionally, the method further comprises:
and the heuristic factor calculation unit is used for obtaining the cloud computing equipment load factor according to the performance parameters and the task load information of the cloud computing equipment and inverting the cloud computing equipment load factor through a preset cloud computing equipment load calculation formula to obtain the heuristic factor of the ant colony algorithm model.
The cloud computing equipment load computing formula specifically comprises:
Figure BDA0002423361100000041
in the formula, the tlength(Vj) The total length of the task already running on the cloud computing device j, tlength_iFor the length of task i, said Vcomp_jIs the performance of the cloud computing device j, the task represents the total number of tasks already running on the cloud computing device j, and the LBijAnd calculating the load coefficient of the cloud computing equipment.
Optionally, the ant colony search executing unit is specifically configured to:
according to the task to be scheduled and the initial pheromone concentration, node path searching operation is carried out according to a pseudorandom proportion transition probability calculation formula in a preset ant colony algorithm model to obtain a node path searching result, wherein the pseudorandom proportion transition probability calculation formula specifically comprises the following steps:
Figure BDA0002423361100000042
wherein ρ is0(0≤ρ0≦ 1) is a constant value, and ρ is obedient [0,1 ≦]Uniformly distributing random values, said τij(t) is the actual pheromone concentration between node i and node j, ηijIs the heuristic factor.
Optionally, the pheromone updating unit is specifically configured to:
according to the pheromone concentration between the nodes in the node path search result and the path length between the nodes, the actual pheromone concentration is updated through a preset pheromone reward punishment updating formula, wherein the pheromone reward punishment updating formula specifically comprises the following steps:
Figure BDA0002423361100000051
in the formula (I), the
Figure BDA0002423361100000052
Represents the updated actual pheromone concentration, the
Figure BDA0002423361100000053
Representing the actual pheromone concentration before updating, wherein omega is an added reward and punishment coefficient, and LcurFor the total length of the currently searched path, LlastThe Q is the total length of the last searched path, and is the total amount of pheromone released by ants on all paths.
Optionally, the method further comprises: the initial pheromone concentration calculating unit is used for carrying out mean value operation according to the obtained performance parameters of the cloud computing equipment to obtain average performance parameters, and obtaining the initial pheromone concentration according to the ratio of the performance parameters to the average performance parameters.
According to the technical scheme, the embodiment of the application has the following advantages:
the application provides a cloud computing task scheduling method, which comprises the following steps: acquiring a task to be scheduled; according to the task to be scheduled and the preset initial pheromone concentration, node path searching operation is carried out through a preset ant colony algorithm model, and a node path searching result is obtained, wherein the node path searching result comprises the following steps: actual pheromone concentration between nodes and path length between nodes, wherein a heuristic factor of the ant colony algorithm model is a coefficient obtained by calculation according to performance parameters of the cloud computing equipment and a task load state of the cloud computing equipment; updating the actual pheromone concentration according to the node path search result; and determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to carry out task scheduling according to the optimal path result.
According to the ant colony path search method and device, ant colony path search is carried out based on heuristic factors obtained through calculation according to performance parameters of the cloud computing device and the task load state of the cloud computing device, load balance of the cloud computing device is introduced and serves as node search conditions of an ant colony algorithm model, so that a task scheduling scheme based on the load state of the cloud computing device is obtained according to an optimal solution output by the model, and the technical problem that energy consumption is too high due to the fact that the load balance is not considered in the existing cloud computing task scheduling mode is solved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 is a schematic flowchart of a cloud computing task scheduling method according to a first embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a cloud computing task scheduling method according to a second embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a cloud computing task scheduling apparatus according to a first embodiment of the present disclosure.
Detailed Description
The embodiment of the application provides a cloud computing task scheduling method and device, and aims to solve the technical problem of overhigh energy consumption of the existing cloud computing task scheduling mode.
The ant colony algorithm is an intelligent bionic algorithm inspired by collective foraging behavior of ant colonies, is widely applied to the research of key problems of combinatorial optimization, and is proposed by Dorigo, an Italian scholaro, in his doctor's paper at first. Ants search for food while walking can secrete a chemical substance called 'pheromone' on a path through which the ants pass, and the chemical substance is volatile, so that information content communication among the ants can be realized. The ants searched later can sense the pheromone left by the former ants and the concentration, and the direction of the next search is selected according to the pheromone and the concentration. The foraging behavior of ants has a remarkable positive feedback mechanism, which is specifically shown in that the pheromone concentration of the path where more ants pass is higher and is selected probably, so that more ants pass through the path, and the pheromone accumulation concentration is higher. Eventually, it is found that all ants search along the path with the highest pheromone concentration, i.e. the direction with the shortest path.
In the ant colony algorithm, the number of ant colonies is initialized to m, and the initial pheromone concentration on each path is set. In order to prevent ants from repeatedly accessing nodes, a tabu table is respectively set for each ant kkAdding the searched node i into the set tabu after each timekIn (1), the subsequent search is in the non-tabu set, i.e., allowedk＝N-tabukIs carried out in (1). And when the ants reach the node i, selecting the next searching node according to the set transition probability, wherein the transition probability is calculated according to the pheromone concentration in the path and the heuristic information. While avoiding premature algorithm recoveryIn the case of convergence, the pheromone needs to be updated in time.
In order to make the objects, features and advantages of the present invention more apparent and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the embodiments described below are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, a first embodiment of the present application provides a cloud computing task scheduling method, including:
step 101, obtaining a task to be scheduled.
It should be noted that before the cloud computing task scheduling is implemented, task information of a task i to be scheduled and an initial pheromone concentration between nodes need to be acquired.
And 102, carrying out node path search operation through a preset ant colony algorithm model according to the task to be scheduled and the preset initial pheromone concentration to obtain a node path search result.
The heuristic factor of the ant colony algorithm model is a coefficient obtained through calculation according to the performance parameters of the cloud computing equipment and the task load state of the cloud computing equipment, and the initial pheromone concentration of the embodiment is the pheromone parameter obtained through calculation according to the performance parameters of each cloud computing equipment.
It should be noted that, according to the task to be scheduled and the preset initial pheromone obtained in step 101, a node path search operation is performed through a preset ant colony algorithm model, and when all ants complete the search, a node path search result is obtained, where the node path search result in this embodiment includes: the longer the path between the task node i and the cloud computing device node j is, the longer the execution time of the task i in the cloud computing device j is represented, and accordingly, the higher the energy consumption for executing the task is.
And 103, updating the actual pheromone concentration according to the node path search result.
And step 104, determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to carry out task scheduling according to the optimal path result.
It should be noted that, according to the updated actual pheromone concentration obtained in step 103, an optimal path result is determined, so that task scheduling is performed according to the optimal path result.
According to the method and the device, the ant colony path search is carried out based on heuristic factors obtained through calculation according to the performance parameters of the cloud computing equipment and the task load state of the cloud computing equipment, the load balance of the cloud computing equipment is introduced and serves as a node search condition of the ant colony algorithm model, so that a task scheduling scheme based on the load state of the cloud computing equipment is obtained according to an optimal solution output by the model, and the technical problem that the energy consumption is too high due to the fact that the load balance is not considered in the existing cloud computing task scheduling mode is solved.
The above is a detailed description of a first embodiment of a cloud computing task scheduling method provided by the present application, and the following is a detailed description of a second embodiment of the cloud computing task scheduling method provided by the present application.
Referring to fig. 2, a second embodiment of the present application provides a cloud computing task scheduling method, including:
step 201, obtaining a cloud computing device load coefficient through a preset cloud computing device load computing formula according to the performance parameters and the task load information of the cloud computing device.
202, inverting the load coefficient of the cloud computing equipment to obtain a heuristic factor of the ant colony algorithm model.
It should be noted that, as can be seen from the principle of the standard ant colony algorithm, the heuristic factor is an important parameter that can control the ant colony node search result, and the value of the conventional heuristic factor is represented by the distance between two resource nodes. However, in the cloud computing resource scheduling process, the heuristic factor is only represented by the distance between the nodes, and an optimal load balancing scheduling scheme cannot be obtained. In cloud computing resource scheduling, when an ant selects a cloud computing device to execute the ant for a task, whether load is balanced is also considered in addition to time and cost, and for this reason, the cloud computing device load calculation formula of this embodiment is specifically:
Figure BDA0002423361100000081
in the formula, tlength(Vj) Is the total length of the task already running on the cloud computing device j, tlength_iIs the length of task i, Vcomp_jIs the performance of the cloud computing device j, task represents the total number of tasks that have been run on the cloud computing device j, LBijThe cloud computing device load factor.
From the above formula, the heuristic factor calculation formula is:
Figure BDA0002423361100000082
step 203, performing mean value operation according to the obtained performance parameters of the cloud computing device to obtain average performance parameters, and obtaining initial pheromone concentration according to the ratio of the performance parameters to the average performance parameters.
Note that τ is used as an initial value of pheromone on the path between the task and the cloud computing device in this embodiment0It is shown that,
Figure BDA0002423361100000091
average computing power V for cloud computing deviceavgcompIt is shown that,
Figure BDA0002423361100000092
and step 204, acquiring a task to be scheduled.
And step 205, performing node path search operation through a preset pseudo random proportion transfer probability calculation formula in the ant colony algorithm model according to the task to be scheduled and the initial pheromone concentration to obtain a node path search result.
It should be noted that the present embodiment preferably adopts a pseudo-random proportion rule to improve the transition probability to increase its randomness. The method comprises the following specific steps:
Figure BDA0002423361100000093
in the formula, ρ0(0≤ρ0≦ 1) is a constant value, ρ is obedient [0,1 ≦]The random values are evenly distributed. If rho is less than or equal to rho0Then [ tau ] is selectedij(t)]α·[ηij(t)]βThe node path with the largest value; if ρ > ρ0The ant will be based on
Figure BDA0002423361100000094
The node at the next time is selected.
Figure BDA0002423361100000095
The calculation formula of (a) is as follows:
Figure BDA0002423361100000096
in the formula, α and β respectively represent heuristic factor expectation factors ηij(t) is a heuristic factor function.
And step 206, updating the actual pheromone concentration through a preset pheromone reward punishment updating formula according to the pheromone concentration between the nodes and the path length between the nodes in the node path searching result.
It should be noted that, in order to further improve the search efficiency of the ant colony algorithm model, in this embodiment, a reward and punishment coefficient is added to the pheromone update formula of the ant colony algorithm, and if an optimal path is found in the mobile search process of ants, a reward is given to the current path immediately according to the reward and punishment coefficient, that is, a corresponding value is added to the pheromone, so that the pheromone concentration of the path is improved. And if the current path is longer than the last searched path, punishing the current path according to the reward and punishment coefficient, namely reducing the pheromone value by a corresponding value, thereby reducing the pheromone concentration of the path.
Specifically, the pheromone reward and punishment updating formula specifically includes:
Figure BDA0002423361100000101
in the formula (I), the compound is shown in the specification,
Figure BDA0002423361100000102
indicating the updated actual pheromone concentration,
Figure BDA0002423361100000103
representing the actual pheromone concentration before updating, omega is the added reward and punishment coefficient, LcurIs the total length of the currently searched path, LlastQ is the total length of the last searched path, and Q is the total amount of pheromone released by the ant on all paths.
When the optimal path appears, the current pheromone concentration is increased
Figure BDA0002423361100000104
As a reward, when a relatively poor path occurs, the current pheromone concentration is subtracted
Figure BDA0002423361100000105
As a penalty, when the worst path occurs, then the current pheromone concentration is subtracted
Figure BDA0002423361100000106
As a severe penalty. And circularly iterating the target function until the maximum iteration number is reached or the optimal solution is found, and outputting the optimal path.
And step 207, determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to perform task scheduling according to the optimal path result.
According to the cloud computing task scheduling method, based on the improved heuristic factor coefficient, the energy consumption of the cloud platform is considered in the scheduling scheme, and the problem of whether the load is balanced is also considered, so that the cloud platform is guaranteed to exert the strength of high performance and low energy consumption, the resource waste caused by overlong task waiting time with short execution time is avoided, and the advantage of distributed processing is embodied. Meanwhile, a pseudo-random proportion rule is adopted to improve the transition probability so as to increase the randomness of the transition probability, and meanwhile, a reward and punishment coefficient is added so as to improve the updating mode of the pheromone, so that the optimal result is obtained more quickly, and the cloud platform with low energy consumption and high performance is realized.
The foregoing is a detailed description of a second embodiment of the cloud computing task scheduling method provided in the present application, and the following is a detailed description of a first embodiment of the cloud computing task scheduling apparatus provided in the present application.
Referring to fig. 3, a third embodiment of the present application provides a cloud computing task scheduling device, including:
a parameter obtaining unit 301, configured to obtain a task to be scheduled;
the ant colony search execution unit 302 is configured to perform node path search operation through a preset ant colony algorithm model according to a task to be scheduled and a preset initial pheromone concentration, so as to obtain a node path search result, where the node path search result includes: actual pheromone concentration between nodes and path length between nodes, wherein a heuristic factor of the ant colony algorithm model is a coefficient obtained by calculation according to performance parameters of the cloud computing equipment and a task load state of the cloud computing equipment;
the pheromone updating unit 303 is used for updating the actual pheromone concentration according to the node path searching result;
and an optimal solution output unit 304, configured to determine an optimal path result from the node path search results according to the updated actual pheromone concentration, so as to perform task scheduling according to the optimal path result.
Further, still include:
a heuristic factor calculating unit 305, configured to obtain a cloud computing device load factor according to the performance parameters and the task load information of the cloud computing device and obtain a cloud computing device load factor by using a preset cloud computing device load calculation formula, and obtain a heuristic factor of the ant colony algorithm model by inverting the cloud computing device load factor;
the cloud computing equipment load computing formula is specifically as follows:
Figure BDA0002423361100000111
in the formula, tlength(Vj) Is the total length of the task already running on the cloud computing device j, tlength_iIs the length of task i, Vcomp_jIs the performance of the cloud computing device j, task represents the total number of tasks that have been run on the cloud computing device j, LBijThe cloud computing device load factor.
Further, the ant colony search executing unit 302 is specifically configured to:
according to the task to be scheduled and the initial pheromone concentration, node path searching operation is carried out according to a pseudorandom proportion transfer probability calculation formula in a preset ant colony algorithm model to obtain a node path searching result, wherein the pseudorandom proportion transfer probability calculation formula specifically comprises the following steps:
Figure BDA0002423361100000112
in the formula, ρ0(0≤ρ0≦ 1) is a constant value, ρ is obedient [0,1 ≦]Uniformly distributing the random value, τij(t) is the actual pheromone concentration between node i and node j, ηijIs a heuristic factor.
Further, the pheromone updating unit 303 is specifically configured to:
according to the pheromone concentration between the nodes and the path length between the nodes in the node path search result, the actual pheromone concentration is updated through a preset pheromone reward punishment updating formula, wherein the pheromone reward punishment updating formula specifically comprises the following steps:
Figure BDA0002423361100000121
in the formula (I), the compound is shown in the specification,
Figure BDA0002423361100000122
indicating the updated actual pheromone concentration,
Figure BDA0002423361100000123
representing the actual pheromone concentration before updating, omega is the added reward and punishment coefficient, LcurIs the total length of the currently searched path, LlastQ is the total length of the last searched path, and Q is the total amount of pheromone released by the ant on all paths.
Further, still include: the initial pheromone concentration calculating unit 306 is configured to perform mean value operation according to the obtained performance parameters of the cloud computing device to obtain average performance parameters, and obtain the initial pheromone concentration according to a ratio of the performance parameters to the average performance parameters.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The terms "first," "second," "third," "fourth," and the like in the description of the application and the above-described figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to 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), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (10)

1. A cloud computing task scheduling method is characterized by comprising the following steps:
acquiring a task to be scheduled;
according to the task to be scheduled and the preset initial pheromone concentration, node path searching operation is carried out through a preset ant colony algorithm model to obtain a node path searching result, wherein the node path searching result comprises the following steps: actual pheromone concentration between nodes and path length between nodes, wherein a heuristic factor of the ant colony algorithm model is a coefficient obtained by calculation according to the performance parameters of the cloud computing equipment and the task load state of the cloud computing equipment;
updating the actual pheromone concentration according to the node path searching result;
and determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to carry out task scheduling according to the optimal path result.
2. The cloud computing task scheduling method according to claim 1, further comprising:
according to the performance parameters and the task load information of the cloud computing equipment, a load coefficient of the cloud computing equipment is obtained through a preset cloud computing equipment load computing formula;
inverting the load coefficient of the cloud computing equipment to obtain a heuristic factor of the ant colony algorithm model;
the cloud computing equipment load computing formula specifically comprises:
Figure FDA0002423361090000011
in the formula, the tlength(Vj) The total length of the task already running on the cloud computing device j, tlength_iFor the length of task i, said Vcomp_jIs the performance of the cloud computing device j, the task represents the total number of tasks already running on the cloud computing device j, and the LBijAnd calculating the load coefficient of the cloud computing equipment.
3. The cloud computing task scheduling method according to claim 1, wherein the obtaining of the node path search result by performing node path search operation through a preset ant colony algorithm model according to the task to be scheduled and a preset initial pheromone concentration specifically comprises:
according to the task to be scheduled and the initial pheromone concentration, node path searching operation is carried out through a pseudorandom proportion transition probability calculation formula in a preset ant colony algorithm model to obtain a node path searching result, wherein the pseudorandom proportion transition probability calculation formula specifically comprises the following steps:
Figure FDA0002423361090000021
wherein ρ is0(0≤ρ0≦ 1) is a constant value, and ρ is obedient [0,1 ≦]Uniformly distributing random values, said τij(t) is the actual pheromone concentration between node i and node j, ηijIs the heuristic factor.
4. The cloud computing task scheduling method according to claim 3, wherein the updating the actual pheromone concentration according to the node path search result specifically includes:
according to the pheromone concentration between the nodes in the node path search result and the path length between the nodes, the actual pheromone concentration is updated through a preset pheromone reward punishment updating formula, wherein the pheromone reward punishment updating formula specifically comprises the following steps:
Figure FDA0002423361090000022
in the formula (I), the
Figure FDA0002423361090000023
Represents the updated actual pheromone concentration, the
Figure FDA0002423361090000024
Representing the actual pheromone concentration before updating, wherein omega is an added reward and punishment coefficient, and LcurFor the total length of the currently searched path, LlastThe Q is the total length of the last searched path, and is the total amount of pheromone released by ants on all paths.
5. The cloud computing task scheduling method according to claim 1, further comprising:
performing mean operation according to the obtained performance parameters of the cloud computing equipment to obtain average performance parameters, and obtaining the initial pheromone concentration according to the ratio of the performance parameters to the average performance parameters.
6. A cloud computing task scheduling apparatus, comprising:
the parameter acquiring unit is used for acquiring a task to be scheduled;
the ant colony search execution unit is used for carrying out node path search operation through a preset ant colony algorithm model according to the task to be scheduled and a preset initial pheromone concentration to obtain a node path search result, wherein the node path search result comprises: actual pheromone concentration between nodes and path length between nodes, wherein a heuristic factor of the ant colony algorithm model is a coefficient obtained by calculation according to the performance parameters of the cloud computing equipment and the task load state of the cloud computing equipment;
the pheromone updating unit is used for updating the actual pheromone concentration according to the node path searching result;
and the optimal solution output unit is used for determining an optimal path result from the node path search result according to the updated actual pheromone concentration so as to carry out task scheduling according to the optimal path result.
7. The cloud computing task scheduling device of claim 6, further comprising:
a heuristic factor calculation unit, configured to obtain a cloud computing device load factor according to the performance parameters and the task load information of the cloud computing device and obtain a heuristic factor of the ant colony algorithm model by inverting the cloud computing device load factor through a preset cloud computing device load calculation formula;
the cloud computing equipment load computing formula specifically comprises:
Figure FDA0002423361090000031
in the formula, the tlength(Vj) The total length of the task already running on the cloud computing device j, tlength_iFor the length of task i, said Vcomp_jIs the performance of the cloud computing device j, the task represents the total number of tasks already running on the cloud computing device j, and the LBijAnd calculating the load coefficient of the cloud computing equipment.
8. The cloud computing task scheduling device according to claim 6, wherein the ant colony search executing unit is specifically configured to:
according to the task to be scheduled and the initial pheromone concentration, node path searching operation is carried out according to a pseudorandom proportion transition probability calculation formula in a preset ant colony algorithm model to obtain a node path searching result, wherein the pseudorandom proportion transition probability calculation formula specifically comprises the following steps:
Figure FDA0002423361090000041
wherein ρ is0(0≤ρ0≦ 1) is a constant value, and ρ is obedient [0,1 ≦]Uniformly distributing random values, said τij(t) is the actual pheromone concentration between node i and node j, ηijIs the heuristic factor.
9. The cloud computing task scheduling device according to claim 8, wherein the pheromone updating unit is specifically configured to:
according to the pheromone concentration between the nodes in the node path search result and the path length between the nodes, the actual pheromone concentration is updated through a preset pheromone reward punishment updating formula, wherein the pheromone reward punishment updating formula specifically comprises the following steps:
Figure FDA0002423361090000042
in the formula (I), the
Figure FDA0002423361090000043
Represents the updated actual pheromone concentration, the
Figure FDA0002423361090000044
Representing the actual pheromone concentration before updating, wherein omega is an added reward and punishment coefficient, and LcurFor the total length of the currently searched path, LlastThe Q is the total length of the last searched path, and is the total amount of pheromone released by ants on all paths.
10. The cloud computing task scheduling device of claim 6, further comprising: the initial pheromone concentration calculating unit is used for carrying out mean value operation according to the obtained performance parameters of the cloud computing equipment to obtain average performance parameters, and obtaining the initial pheromone concentration according to the ratio of the performance parameters to the average performance parameters.
CN202010212709.3A 2020-03-24 2020-03-24 Cloud computing task scheduling method and device Active CN111324444B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010212709.3A CN111324444B (en) 2020-03-24 2020-03-24 Cloud computing task scheduling method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010212709.3A CN111324444B (en) 2020-03-24 2020-03-24 Cloud computing task scheduling method and device

Publications (2)

Publication Number Publication Date
CN111324444A true CN111324444A (en) 2020-06-23
CN111324444B CN111324444B (en) 2023-05-09

Family

ID=71167752

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010212709.3A Active CN111324444B (en) 2020-03-24 2020-03-24 Cloud computing task scheduling method and device

Country Status (1)

Country Link
CN (1) CN111324444B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112148482A (en) * 2020-09-11 2020-12-29 电子科技大学 Edge computing task scheduling method based on combination with load balancing

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102932422A (en) * 2012-09-29 2013-02-13 南京邮电大学 Cloud environment task scheduling method based on improved ant colony algorithm
CN103345657A (en) * 2013-04-02 2013-10-09 江苏大学 Task scheduling method based on heredity and ant colony in cloud computing environment
CN106951059A (en) * 2017-03-28 2017-07-14 中国石油大学(华东) Based on DVS and the cloud data center power-economizing method for improving ant group algorithm

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102932422A (en) * 2012-09-29 2013-02-13 南京邮电大学 Cloud environment task scheduling method based on improved ant colony algorithm
CN103345657A (en) * 2013-04-02 2013-10-09 江苏大学 Task scheduling method based on heredity and ant colony in cloud computing environment
CN106951059A (en) * 2017-03-28 2017-07-14 中国石油大学(华东) Based on DVS and the cloud data center power-economizing method for improving ant group algorithm

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112148482A (en) * 2020-09-11 2020-12-29 电子科技大学 Edge computing task scheduling method based on combination with load balancing
CN112148482B (en) * 2020-09-11 2023-08-22 电子科技大学 Edge computing task scheduling method based on combination of load balancing

Also Published As

Publication number Publication date
CN111324444B (en) 2023-05-09

Similar Documents

Publication Publication Date Title
US20170329643A1 (en) Distributed node intra-group task scheduling method and system
CN104901989B (en) A kind of Site Service offer system and method
CN112685170B (en) Dynamic optimization of backup strategies
CN112416585B (en) Deep learning-oriented GPU resource management and intelligent scheduling method
CN108694077A (en) Based on the distributed system method for scheduling task for improving binary system bat algorithm
Mondal et al. Scheduling of time-varying workloads using reinforcement learning
Ma et al. Real-time virtual machine scheduling in industry IoT network: A reinforcement learning method
CN112783607A (en) Task deployment method and device in container cluster
CN107566535B (en) Self-adaptive load balancing method based on concurrent access timing sequence rule of Web map service
CN116302389A (en) Task scheduling method based on improved ant colony algorithm
CN111324444A (en) Cloud computing task scheduling method and device
Kalai Arasan et al. Energy‐efficient task scheduling and resource management in a cloud environment using optimized hybrid technology
Devagnanam et al. Design and development of exponential lion algorithm for optimal allocation of cluster resources in cloud
CN115794358A (en) Cloud workflow task scheduling method and device, electronic equipment and storage medium
CN106874215B (en) Serialized storage optimization method based on Spark operator
CN116089083A (en) Multi-target data center resource scheduling method
CN116010051A (en) Federal learning multitasking scheduling method and device
CN114980216A (en) Dependent task unloading system and method based on mobile edge calculation
CN112698911B (en) Cloud job scheduling method based on deep reinforcement learning
Talha et al. A chaos opposition‐based dwarf mongoose approach for workflow scheduling in cloud
CN110297704B (en) Particle swarm optimization method and system integrating reverse learning and heuristic perception
CN114625493A (en) Kubernetes cluster resource scheduling method based on improved intelligent longicorn stigma method
CN112328364B (en) Computing-intensive cloud workflow scheduling method based on farmland fertility algorithm
Gong et al. Research and Simulation Implementation of Fog Calculation Resource Allocation Algorithm
CN113448687B (en) Hyper-heuristic task scheduling method and system based on reinforcement learning in cloud environment

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