CN115344388B - Power real-time simulation calculation task allocation method and device considering communication and calculation force - Google Patents
Power real-time simulation calculation task allocation method and device considering communication and calculation force Download PDFInfo
- Publication number
- CN115344388B CN115344388B CN202211007632.1A CN202211007632A CN115344388B CN 115344388 B CN115344388 B CN 115344388B CN 202211007632 A CN202211007632 A CN 202211007632A CN 115344388 B CN115344388 B CN 115344388B
- Authority
- CN
- China
- Prior art keywords
- task
- solving
- calculation
- thread
- cpu
- 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
- 238000004364 calculation method Methods 0.000 title claims abstract description 226
- 238000004088 simulation Methods 0.000 title claims abstract description 126
- 238000004891 communication Methods 0.000 title claims abstract description 116
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000009826 distribution Methods 0.000 claims abstract description 31
- 230000001052 transient effect Effects 0.000 claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims description 104
- 238000012545 processing Methods 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 9
- 235000006679 Mentha X verticillata Nutrition 0.000 claims description 5
- 235000002899 Mentha suaveolens Nutrition 0.000 claims description 5
- 235000001636 Mentha x rotundifolia Nutrition 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000005457 optimization Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a power real-time simulation calculation task allocation method and device taking communication and calculation force into account, wherein the method comprises the following steps: acquiring an electric power calculation task set and a task solution topological structure thereof, determining the solution calculation force of each solution task in the electric power calculation task set, constructing a calculation force constraint condition of a thread group formed by every two threads, constructing an objective function in a CPU multithread parallel mode based on the task solution topological structure and communication performance parameters of each CPU and each core thereof, and distributing each solution task to each thread according to the objective function under the calculation force constraint condition. Therefore, the communication performance between the CPU and the cores is analyzed, the multithreading optimization distribution is carried out on the multithreading calculation solving tasks by utilizing the multithreading calculation capability of the CPU, the multithreading parallel calculation capability of the CPU is fully exerted, the calculation scale of the electromagnetic transient real-time simulation is favorably increased, and the simulation performance of the electromagnetic transient simulation system real-time simulation is improved.
Description
Technical Field
The application relates to the field of power system simulation, in particular to a power real-time simulation calculation task distribution method taking communication and calculation force into consideration.
Background
With the increasing demand for electricity, power systems are also rapidly developing, and more power electronic devices are created to enrich the power systems, so that the power systems become more complex, and therefore, simulation of the power systems before operation is necessary. The simulation of the power system is an effective means for knowing the characteristics of the power system, supporting the research, planning, operation, production and equipment manufacturing of the power system and guaranteeing the safe and reliable operation of the power system.
In the simulation process of the electromagnetic transient simulation system, a plurality of electric power calculation tasks need to be calculated. If the historical current solving task is to calculate the current amount of the last simulation step length, and if the equivalent conductance matrix is to calculate the conductance matrix of the simulation object, the conductance matrix is decoupled into an upper triangular matrix and a lower triangular matrix, and if the secondary control solving task is to calculate the calculation task of the secondary control system for digital signal simulation.
At present, the electric power calculation tasks of the electromagnetic transient simulation system are usually manually distributed to different cores of different CPUs for calculation and solving, so that the communication time among the different cores of the CPUs, the communication time among the different CPUs and the calculation power of the CPUs are ignored, the characteristic of parallel calculation among the multiple cores of the CPUs can not be fully utilized, the task distribution mode is not coordinated, and the calculation scale of electromagnetic transient real-time simulation is small.
And the power calculation tasks are cooperatively distributed to each thread for processing by combining the factors such as communication among different cores of the CPU, communication among different CPUs, calculation power of the CPU and the like, so that the distribution of the power calculation tasks to the threads can be optimized.
Disclosure of Invention
In view of the above problems, the present application is provided to provide a method and apparatus for distributing power real-time simulation computing tasks in consideration of communication and computing power, so as to optimize the distribution of the power computing tasks to threads.
In order to achieve the above object, the following specific solutions are proposed:
a power real-time simulation calculation task allocation method taking communication and calculation force into consideration comprises the following steps:
acquiring an electric power calculation task set in real-time simulation of an electromagnetic transient simulation system and a task solving topological structure of each task in the electric power calculation task set, wherein the electric power calculation task set comprises a plurality of historical current solving tasks, a plurality of equivalent conductivity matrix solving tasks and a plurality of secondary control solving tasks;
determining the solving force of each historical current solving task, the solving force of each equivalent conductivity matrix solving task and the solving force of each secondary control solving task;
when the number of threads currently used for power calculation task allocation is larger than 1, based on the solving forces of each historical current solving task, the solving forces of each equivalent conductivity matrix solving task and the solving forces of each secondary control solving task, generating the solving force constraint conditions of a thread group formed by every two threads;
Constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode based on the task solving topological structure and communication performance parameters of each CPU and each core thereof;
and under the condition that each thread meets the computational force constraint condition of a thread group formed by two threads, according to the minimum simulation time objective function, distributing each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread.
Optionally, the generating the computational force constraint condition of the thread group formed by every two threads based on the computational force of each historical current solving task, the computational force of each equivalent conductivity matrix solving task and the computational force of each secondary control solving task includes:
determining the computing power of each thread currently used for computing task allocation;
based on the calculation force of each thread, the calculation force constraint condition of the thread group formed by every two threads is obtained by using the following formula:
wherein,for the calculation of thread s, C hc-s Solving the calculation force of a first part of historical current solving tasks in each historical current solving task for the thread s, C cm-s Solving a first partial equivalent conductivity matrix solving task of each historical current solving task for the thread sForce C sc-s Solving the calculation force of a first part of the quadratic control solving tasks in each historical current solving task for the thread s, C hc-p Solving the calculation force of a second part of historical current solving tasks in each historical current solving task for the thread p, C cm-p Solving the calculation force of a second part of equivalent conductivity matrix solving task in each historical current solving task for the thread p, C sc-p And solving the computational force of a second part of secondary control solving tasks in each historical current solving task for the thread p, wherein q is a preset deviation threshold, no intersection exists between the first part of historical current solving tasks and the second part of historical current solving tasks, no intersection exists between the first part of equivalent conductivity matrix solving tasks and the second part of equivalent conductivity matrix solving tasks, and no intersection exists between the first part of secondary control solving tasks and the second part of secondary control solving tasks.
Optionally, the method further comprises:
when the number of threads currently used for power calculation task allocation is 1, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are allocated to one thread currently used for power calculation task allocation.
Optionally, constructing a minimum simulation time objective function of the power calculation task set in a multi-thread parallel mode of the CPU based on the task solving topology structure and the communication performance parameters of each CPU and each core thereof, including:
analyzing the task solving topological structure, and determining the total number of times of communication among the CPUs and the total number of times of communication among the CPU cores;
according to the first data volume of each inter-CPU communication, the time of transmitting the first data volume between every two cores, the total number of inter-CPU communication, the second data volume of each inter-CPU communication, the time of transmitting the second data volume between every two CPUs, the total number of inter-CPU communication, the minimum simulation time objective function of the power calculation task set under the CPU multithread parallel mode is constructed.
Optionally, the time for transmitting the first data amount between each two cores according to the first data amount of each inter-CPU communication, the total number of inter-CPU communication times, the second data amount of each inter-CPU communication, and the time for transmitting the second data amount between each two CPUs, the total number of inter-CPU communication times, constructs a minimum simulation time objective function of the power calculation task set, including:
Constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode by using the following steps:
wherein mint is the minimum simulation time of the power calculation task set, n 1 R is the total number of communication between the CPU cores core-i A first data amount, t, for the ith CPU inter-core communication core-c For the time of transmission of a first data volume unit between every two cores, n 2 R is the total number of communication between the CPUs CPU-j Second data amount, t, for jth inter-CPU communication CPU-c The time for transferring a unit of the second data amount between each two CPUs.
Optionally, the method further comprises:
determining an execution sequence constraint condition set of the electric power calculation task set, wherein the task execution constraint condition set comprises a first constraint condition, a second constraint condition and a third constraint condition, the first constraint condition is that each secondary control solving task executes calculation according to a preset task layer sequence, the second constraint condition is that the starting processing time of each secondary control solving task is later than the ending processing time of the secondary control solving task positioned before the sub-task execution sequence of the secondary control solving task, and the third constraint condition is that the execution sequence of each equivalent conductivity matrix solving task is later than the historical current solving task associated with the equivalent conductivity matrix solving task;
Under the condition that each thread meets the computational power constraint condition of a thread group formed by two threads, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread, and the method comprises the following steps:
and under the condition that each thread meets the computational power constraint condition of a thread group formed by two threads and the electric power calculation task set meets each constraint condition in the execution sequence constraint condition set, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread.
Optionally, under the condition that each thread meets the computational force constraint condition of a thread group formed by two threads, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread, and then the method further comprises the following steps:
and determining the minimum time of the minimum simulation time objective function.
A power real-time simulation computing task allocation device that allows for communication and computing power, comprising:
The system comprises a solving task acquisition unit, a power computing task set and a task solving topological structure, wherein the solving task acquisition unit is used for acquiring a power computing task set in real-time simulation of an electromagnetic transient simulation system and a task solving topological structure of each task in the power computing task set, and the power computing task set comprises a plurality of historical current solving tasks, a plurality of equivalent conductivity matrix solving tasks and a plurality of secondary control solving tasks;
a solution computing force determining unit for determining a solution computing force of each historical current solution task, a solution computing force of each equivalent conductance matrix solution task, and a solution computing force of each secondary control solution task;
the computing force constraint generating unit is used for generating computing force constraint conditions of a thread group formed by every two threads based on the computing force of each historical current computing task, the computing force of each equivalent conductivity matrix computing task and the computing force of each secondary control computing task when the number of threads currently used for power computing task allocation is larger than 1;
the objective function construction unit is used for constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode based on the task solving topological structure and communication performance parameters of each CPU and each core of the CPU;
And the calculation task allocation unit is used for allocating each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread according to the minimum simulation time objective function under the condition that each thread meets the calculation force constraint condition of a thread group formed by two threads.
Optionally, the computational force constraint generating unit includes:
a thread computing force determining unit, configured to determine a computing force of each thread currently used for computing task allocation when the number of threads currently used for power computing task allocation is greater than 1;
the constraint condition determining unit is used for obtaining the calculation force constraint condition of the thread group formed by every two threads according to the following formula based on the calculation force of each thread:
wherein,for the calculation of thread s, C hc-s Solving the calculation force of a first part of historical current solving tasks in each historical current solving task for the thread s, C cm-s Solving the calculation force of a first partial equivalent conductivity matrix solving task in each historical current solving task for the thread s, C sc-s Solving the calculation force of a first part of the quadratic control solving tasks in each historical current solving task for the thread s, C hc-p Solving the calculation force of a second part of historical current solving tasks in each historical current solving task for the thread p, C cm-p Solving each historical current solution for the thread pCalculating force of second partial equivalent conductivity matrix solving task in task, C sc-p And solving the computational force of a second part of secondary control solving tasks in each historical current solving task for the thread p, wherein q is a preset deviation threshold, no intersection exists between the first part of historical current solving tasks and the second part of historical current solving tasks, no intersection exists between the first part of equivalent conductivity matrix solving tasks and the second part of equivalent conductivity matrix solving tasks, and no intersection exists between the first part of secondary control solving tasks and the second part of secondary control solving tasks.
Optionally, the method further comprises:
the single-thread distribution unit is used for distributing each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to one thread currently used for power computing task distribution when the number of threads currently used for power computing task distribution is 1.
Optionally, the objective function construction unit includes:
The first objective function construction subunit is used for analyzing the task solving topological structure and determining the total number of times of communication among the CPU and the total number of times of communication among the CPU cores;
the second objective function constructing subunit is configured to construct, according to the first data amount of each inter-CPU-core communication, a minimum simulation time objective function of the power computing task set in the multi-thread parallel mode of the CPU, where the time is a unit of the first data amount transmitted between every two cores, the total number of inter-CPU-core communication, the second data amount of each inter-CPU-core communication, and the time is a unit of the second data amount transmitted between every two CPUs.
Optionally, the second objective function building subunit includes:
an objective function determining unit, configured to construct a minimum simulation time objective function of the power calculation task set in the CPU multithreading parallel mode by using:
wherein mint is the minimum simulation time of the power calculation task set, n 1 R is the total number of communication between the CPU cores core-i A first data amount, t, for the ith CPU inter-core communication core-c For the time of transmission of a first data volume unit between every two cores, n 2 R is the total number of communication between the CPUs CPU-j Second data amount, t, for jth inter-CPU communication CPU-c The time for transferring a unit of the second data amount between each two CPUs.
Optionally, the apparatus further comprises:
the order constraint determining unit is used for determining an execution order constraint condition set of the electric power calculation task set, the task execution constraint condition set comprises a first constraint condition, a second constraint condition and a third constraint condition, the first constraint condition is that each secondary control solving task executes calculation according to a preset task layer order, the second constraint condition is that the starting processing time of each secondary control solving task is later than the ending processing time of the secondary control solving task before the sub-task execution order of the secondary control solving task, and the third constraint condition is that the execution order of each equivalent conductivity matrix solving task is later than the historical current solving task associated with the equivalent conductivity matrix solving task;
the computing task allocation unit includes:
and the calculation task allocation subunit is used for allocating each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread according to the minimum simulation time objective function under the condition that each thread meets the calculation force constraint condition of a thread group consisting of two threads and the electric power calculation task set meets each constraint condition in the execution sequence constraint condition set.
Optionally, the apparatus further comprises:
and the minimum time determining unit is used for determining the minimum time of the minimum simulation time objective function.
By means of the technical scheme, the method and the system acquire the electric power calculation task set in the electromagnetic transient simulation system real-time simulation and the task solving topological structure of each task in the electric power calculation task set, wherein the electric power calculation task set comprises a plurality of historical current solving tasks, a plurality of equivalent conductivity matrix solving tasks and a plurality of secondary control solving tasks, the solving force of each historical current solving task is determined, the solving force of each equivalent conductivity matrix solving task and the solving force of each secondary control solving task are determined, when the number of threads currently distributed for the electric power calculation task is greater than 1, the solving force of each equivalent conductivity matrix solving task and the solving force of each secondary control solving task are generated based on the solving force of each historical current solving task, the constraint condition of each thread group formed by each two threads is generated based on the task topological structure and the communication performance parameters of each CPU and each core, the minimum simulation time objective function of the electric power calculation task set is constructed under the condition that each thread meets two groups, and the solving time objective function is distributed to each secondary control task according to the multiple thread constraint time constraint function. Therefore, the communication performance between the CPU and the cores is analyzed, the multithreading optimization distribution is carried out on the multithreading calculation solving tasks by utilizing the multithreading calculation capability of the CPU, the multithreading parallel calculation capability of the CPU is fully exerted, the calculation scale of the electromagnetic transient real-time simulation is increased, and the simulation performance of the electromagnetic transient simulation system is improved.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 is a schematic flow chart of a real-time simulation calculation task distribution of electric power, which is provided by an embodiment of the application and takes communication and calculation force into consideration;
fig. 2 is a schematic diagram of a device structure for real-time simulation calculation task distribution of electric power, which is provided by an embodiment of the application and takes communication and calculation force into consideration.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The scheme of the application can be realized based on the terminal with the data processing capability, and the terminal can be a computer, a server, a cloud end and the like.
Next, as described in connection with fig. 1, the power real-time simulation calculation task allocation method of the present application, which allows for communication and calculation, may include the steps of:
step S110, acquiring an electric power calculation task set in real-time simulation of an electromagnetic transient simulation system and a task solving topological structure of each task in the electric power calculation task set.
Specifically, the power calculation task set may include a plurality of historical current solving tasks, a plurality of equivalent conductance matrix solving tasks, and a plurality of secondary control solving tasks.
The historical current solving task can be a task of calculating the current amount of the last simulation step in the real-time simulation of the electromagnetic transient simulation system. The equivalent conductivity matrix solving task can be a task of decoupling a simulation conductivity matrix into an upper triangular matrix and a lower triangular matrix in real-time simulation of an electromagnetic transient simulation system. The secondary control solving task can be a calculation task of the secondary control system in real-time simulation of the electromagnetic transient simulation system.
It can be understood that the historical current solving task, the equivalent conductivity matrix solving task and the secondary control solving task can be the minimum task unit disassembled in the simulation task, so that the electric power computing task set can protect a plurality of historical current solving tasks, a plurality of equivalent conductivity matrix solving tasks and a plurality of secondary control solving tasks.
Step S120, determining the solving force of each historical current solving task, the solving force of each equivalent conductivity matrix solving task and the solving force of each secondary control solving task.
Specifically, the solving force of each historical current solving task may represent the amount of computing resources required to be consumed for solving each historical current solving task, the solving force of each equivalent conductivity matrix solving task may represent the amount of computing resources for solving each equivalent conductivity matrix solving task, and the solving force of each secondary control solving task may represent the amount of computing resources for solving each secondary control solving task.
And step S130, when the number of threads currently used for power calculation task allocation is greater than 1, generating a computational force constraint condition of a thread group formed by every two threads based on the computational force of each historical current computational task, the computational force of each equivalent conductivity matrix computational task and the computational force of each secondary control computational task.
It is understood that when the number of threads used for power calculation task allocation is greater than 1, each task in the power calculation task set may be allocated to the threads, and when the allocation is planned, the computing power of each thread and the computing power of each task may be matched, so as to obtain constraint conditions about a thread group formed by every two threads, wherein the constraint conditions are used for balancing the computing power.
And step 140, constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode based on the task solving topological structure and the communication performance parameters of each CPU and each core thereof.
Specifically, the communication performance parameter of each CPU and each core thereof may include the communication time between CPUs and the communication time between cores of the CPUs.
It can be understood that in the process of solving or calculating each historical current solving task, each equivalent conductance matrix solving task or each secondary control solving task, communication between cores of the CPU and between the CPUs takes up time, so that the communication time between cores of the CPU and between the CPUs needs to be analyzed when constructing the minimum simulation time objective function.
And step S150, under the condition that each thread meets the computational constraint condition of a thread group formed by two threads, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread.
Specifically, when each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread, the minimum simulation time in the minimum simulation time objective function can be taken as a target, and the task distribution result of the power computing task set is obtained under the condition that each thread meets the calculation force constraint condition of a thread group consisting of two threads, wherein the task distribution result can comprise a plurality of task distribution schemes, each task distribution scheme can meet the calculation force constraint condition of the thread group consisting of two threads, and the minimum simulation time in the minimum simulation time objective function is met.
According to the electric power real-time simulation calculation task distribution method considering communication and calculation force, when the number of threads currently used for electric power calculation task distribution is larger than 1, the calculation force of each equivalent electric conductivity matrix calculation task and the calculation force of each secondary control calculation task are obtained, a calculation force constraint condition of a thread group formed by each two threads is generated, based on the calculation force of each historical electric current calculation task, the calculation force of each equivalent electric conductivity matrix calculation task and the communication performance parameters of each CPU and each core, a minimum simulation time objective function of the electric power calculation task set in a CPU multi-thread mode is established, when the number of threads currently used for electric power calculation task distribution is larger than 1, the calculation force of each equivalent electric conductivity matrix calculation task is calculated based on the calculation force of each historical electric current calculation task, and the calculation force of each secondary control calculation task, the calculation force of each thread group formed by each two threads is calculated, and the calculation time objective function of each CPU multi-thread group is calculated based on the calculation topological structure of the task and the communication performance parameters of each CPU and each core, and the calculation time objective function is calculated according to the two threads and the calculation time constraint conditions of each thread group is calculated in parallel. Therefore, the communication performance between the CPU and the cores is analyzed, the multithreading optimization distribution is carried out on the multithreading calculation solving tasks by utilizing the multithreading calculation capability of the CPU, the multithreading parallel calculation capability of the CPU is fully exerted, the calculation scale of the electromagnetic transient real-time simulation is increased, and the simulation performance of the electromagnetic transient simulation system is improved.
In some embodiments of the present application, a process for generating a computational force constraint condition of a thread group formed by every two threads according to the computational force of each historical current solving task, the computational force of each equivalent conductivity matrix solving task, and the computational force of each quadratic control solving task mentioned in the above embodiments may be described, where the process includes:
s1, determining the calculation power of each thread currently used for calculating task allocation.
Specifically, the computing power of each thread may represent the maximum amount of computing resources used to provide a solution task to perform a solution calculation, and the computing power of each thread may be determined by the number of CPUs, the performance of the CPUs, and the current state of the processing system in which the electromagnetic transient simulation system is located.
S2, based on the calculation power of each thread, obtaining the calculation power constraint condition of a thread group formed by every two threads by using the following formula:
wherein,for the calculation of thread s, C hc-s Solving the calculation force of a first part of historical current solving tasks in each historical current solving task for the thread s, C cm-s Solving the calculation force of a first partial equivalent conductivity matrix solving task in each historical current solving task for the thread s, C sc-s Solving the calculation force of a first part of the quadratic control solving tasks in each historical current solving task for the thread s, C hc-p Solving the calculation force of a second part of historical current solving tasks in each historical current solving task for the thread p, C cm-p Solving the calculation force of a second part of equivalent conductivity matrix solving task in each historical current solving task for the thread p, C sc-p And solving the computational force of a second part of secondary control solving tasks in each historical current solving task for the thread p, wherein q is a preset deviation threshold, no intersection exists between the first part of historical current solving tasks and the second part of historical current solving tasks, no intersection exists between the first part of equivalent conductivity matrix solving tasks and the second part of equivalent conductivity matrix solving tasks, and no intersection exists between the first part of secondary control solving tasks and the second part of secondary control solving tasks.
Specifically, the preset deviation threshold may represent a minimum degree of uniformity in the tasks that the two threads each assume.
According to the power real-time simulation calculation task distribution method considering communication and calculation power, which is provided by the embodiment, the calculation power of each thread used for calculating task distribution is analyzed, so that the processing uniformity degree between the two threads is obtained, and the calculation power constraint condition is constructed, so that each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task can be performed uniformly in parallel when distributed to each thread.
Considering that the electromagnetic transient simulation system has more current processing tasks in the processing system, the number of threads capable of being used for power calculation task allocation is only 1, in some embodiments of the present application, the power real-time simulation calculation task allocation method considering communication and calculation force may further include the following processes:
when the number of threads currently used for power calculation task allocation is 1, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are allocated to one thread currently used for power calculation task allocation.
It will be appreciated that when there are only 1 thread for power computing task allocation, all tasks in the power computing task set are allocated to be executed by the thread, and the thread may perform computing or solving tasks serially for all tasks in the power computing task set.
In some embodiments of the present application, the process of constructing the minimum simulation time objective function of the power calculation task set in the CPU multithreading parallel mode is described in the step S140, based on the task solution topology, and the communication performance parameters of each CPU and each core thereof, where the process may include:
S1, analyzing the task solving topological structure, and determining the total number of times of communication among the CPUs and the total number of times of communication among the CPU cores.
It can be understood that when each historical current solving task, each equivalent conductance matrix solving task or each secondary control solving task is calculated/solved, communication occurs between the CPUs for a plurality of times and communication occurs between the CPU cores for a plurality of times, and the specific occurrence times of communication can be based on the task solving content of each historical current solving task, each equivalent conductance matrix solving task or each secondary control solving task, so that the total times of communication between the CPUs and the total times of communication between the CPU cores can be analyzed from the task solving topological structure.
S2, according to the first data volume of each time of inter-CPU inter-core communication, the time of transmitting the first data volume of each two cores, the total number of times of inter-CPU communication, the second data volume of each time of inter-CPU communication, the time of transmitting the second data volume of each two CPUs, the total number of times of inter-CPU communication, and constructing a minimum simulation time objective function of the power calculation task set in a CPU multithread parallel mode.
Specifically, the minimum simulation time objective function of the power computation task set in the CPU multithreading parallel mode may be constructed using:
Wherein mint is the minimum simulation time of the power calculation task set, n 1 R is the total number of communication between the CPU cores core-i A first data amount, t, for the ith CPU inter-core communication core-c For the time of transmission of a first data volume unit between every two cores, n 2 R is the total number of communication between the CPUs CPU-j Second data amount, t, for jth inter-CPU communication CPU-c The time for transferring a unit of the second data amount between each two CPUs.
Further, a minimum time of the minimum simulation time objective function may be determined.
Specifically, after determining the minimum time of the minimum simulation time objective function, the simulation calculation may be performed according to the minimum time when each task in the power calculation task set is allocated to each thread, so as to simulate the power system.
According to the power real-time simulation calculation task allocation method taking communication and calculation power into consideration, a minimum simulation time objective function is generated through the second data volume and the total times of communication between the CPUs and the first data volume and the total times of communication between the CPU cores, so that the minimum time when each task in the power calculation task set is optimally allocated can be obtained, and the simulation performance of the electromagnetic transient simulation system is improved.
In view of the provision of each task in a power computing task set with its execution order, in some embodiments of the present application, there is provided a power computing task allocation method under execution order constraints, the method may include the following processes:
A set of execution order constraints for the set of power computing tasks is determined.
Specifically, the set of task execution constraints may include a first constraint, a second constraint, and a third constraint.
The first constraint condition may be that each secondary control solving task performs calculation according to a preset task layer sequence, the second constraint task may be that a start processing time of each secondary control solving task is later than an end processing time of a secondary control solving task located before an execution sequence of a subtask of the secondary control solving task, and the third constraint condition is that an execution sequence of each equivalent conductivity matrix solving task is after a historical current solving task associated with the equivalent conductivity matrix solving task.
The task topological structure formed by the secondary control solving tasks can comprise a plurality of task levels, each task level can comprise a plurality of task branches, and each task branch can comprise a plurality of secondary control solving tasks in serial execution sequence. In this case, the third constraint condition may be that the secondary control solution task with the earlier serial execution sequence in the same task branch is preferentially executed, and the secondary control solution task with the later serial execution sequence is executed later. The first constraint condition can be that the secondary control solving task with the high task level is executed preferentially, and the secondary control solving task with the low task level is executed later.
For example, the task level of the secondary control solving task a is a first level, the task level of the secondary control solving task B is a second level, and then the execution order of B is after a.
Based on this, in the step S150, under the condition that each thread satisfies the computational constraint condition of the thread group formed by two threads, according to the minimum simulation time objective function, the process of distributing each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread may include:
and under the condition that each thread meets the computational power constraint condition of a thread group formed by two threads and the electric power calculation task set meets each constraint condition in the execution sequence constraint condition set, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread.
According to the power real-time simulation calculation task distribution method considering communication and calculation power, the execution sequence of each task in the power calculation task set is analyzed to generate the execution sequence constraint condition set of the power calculation task set, so that the distribution of each task in the power calculation task set to each thread is constrained, and the multi-thread distribution is carried out on multiple calculation solving tasks more optimally.
The device for realizing the power real-time simulation calculation task allocation taking into account the communication and the calculation force, which is provided by the embodiment of the application, is described below, and the device for realizing the power real-time simulation calculation task allocation taking into account the communication and the calculation force, which is described below, and the method for realizing the power real-time simulation calculation task allocation taking into account the communication and the calculation force, which are described above, can be correspondingly referred to each other.
Referring to fig. 2, fig. 2 is a schematic diagram of a device structure of a method for distributing power real-time simulation calculation tasks in consideration of communication and calculation power according to an embodiment of the present application.
As shown in fig. 2, the apparatus may include:
the solving task obtaining unit 11 is configured to obtain an electric power computing task set in real-time simulation of the electromagnetic transient simulation system, and a task solving topology structure of each task in the electric power computing task set, where the electric power computing task set includes a plurality of historical current solving tasks, a plurality of equivalent conductance matrix solving tasks, and a plurality of secondary control solving tasks;
a solution force determination unit 12 that determines a solution force for each historical current solution task, a solution force for each equivalent conductance matrix solution task, and a solution force for each secondary control solution task;
A calculation force constraint generating unit 13, configured to generate a calculation force constraint condition of a thread group formed by every two threads, based on a calculation force of each historical current calculation task, a calculation force of each equivalent conductivity matrix calculation task, and a calculation force of each secondary control calculation task when the number of threads currently used for power calculation task allocation is greater than 1;
an objective function construction unit 14, configured to construct a minimum simulation time objective function of the power calculation task set in a multi-thread parallel mode of the CPU, based on the task solution topology structure and the communication performance parameters of each CPU and each core thereof;
and the calculation task allocation unit 15 is used for allocating each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread according to the minimum simulation time objective function under the condition that each thread meets the calculation force constraint condition of a thread group formed by two threads.
Optionally, the computing force constraint generating unit 13 includes:
a thread computing force determining unit, configured to determine a computing force of each thread currently used for computing task allocation when the number of threads currently used for power computing task allocation is greater than 1;
The constraint condition determining unit is used for obtaining the calculation force constraint condition of the thread group formed by every two threads according to the following formula based on the calculation force of each thread:
wherein,for the calculation of thread s, C hc-s Solving the calculation force of a first part of historical current solving tasks in each historical current solving task for the thread s, C cm-s Solving the calculation force of a first partial equivalent conductivity matrix solving task in each historical current solving task for the thread s, C sc-s Solving the calculation force of a first part of the quadratic control solving tasks in each historical current solving task for the thread s, C hc-p Solving the calculation force of a second part of historical current solving tasks in each historical current solving task for the thread p, C cm-p Solving for the thread p a second partial equivalent conductance in each historical current solving taskComputing power of matrix solving task, C sc-p And solving the computational force of a second part of secondary control solving tasks in each historical current solving task for the thread p, wherein q is a preset deviation threshold, no intersection exists between the first part of historical current solving tasks and the second part of historical current solving tasks, no intersection exists between the first part of equivalent conductivity matrix solving tasks and the second part of equivalent conductivity matrix solving tasks, and no intersection exists between the first part of secondary control solving tasks and the second part of secondary control solving tasks.
Optionally, the method further comprises:
the single-thread distribution unit is used for distributing each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to one thread currently used for power computing task distribution when the number of threads currently used for power computing task distribution is 1.
Optionally, the objective function construction unit 14 includes:
the first objective function construction subunit is used for analyzing the task solving topological structure and determining the total number of times of communication among the CPU and the total number of times of communication among the CPU cores;
the second objective function constructing subunit is configured to construct, according to the first data amount of each inter-CPU-core communication, a minimum simulation time objective function of the power computing task set in the multi-thread parallel mode of the CPU, where the time is a unit of the first data amount transmitted between every two cores, the total number of inter-CPU-core communication, the second data amount of each inter-CPU-core communication, and the time is a unit of the second data amount transmitted between every two CPUs.
Optionally, the second objective function building subunit includes:
an objective function determining unit, configured to construct a minimum simulation time objective function of the power calculation task set in the CPU multithreading parallel mode by using:
Wherein mint is the minimum simulation time of the power calculation task set, n 1 R is the total number of communication between the CPU cores core-i A first data amount, t, for the ith CPU inter-core communication core-c For the time of transmission of a first data volume unit between every two cores, n 2 R is the total number of communication between the CPUs CPU-j Second data amount, t, for jth inter-CPU communication CPU-c The time for transferring a unit of the second data amount between each two CPUs.
Optionally, the apparatus further comprises:
the order constraint determining unit is used for determining an execution order constraint condition set of the electric power calculation task set, the task execution constraint condition set comprises a first constraint condition, a second constraint condition and a third constraint condition, the first constraint condition is that each secondary control solving task executes calculation according to a preset task layer order, the second constraint condition is that the starting processing time of each secondary control solving task is later than the ending processing time of the secondary control solving task before the sub-task execution order of the secondary control solving task, and the third constraint condition is that the execution order of each equivalent conductivity matrix solving task is later than the historical current solving task associated with the equivalent conductivity matrix solving task;
The calculation task allocation unit 15 includes:
and the calculation task allocation subunit is used for allocating each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread according to the minimum simulation time objective function under the condition that each thread meets the calculation force constraint condition of a thread group consisting of two threads and the electric power calculation task set meets each constraint condition in the execution sequence constraint condition set.
Optionally, the apparatus further comprises:
and the minimum time determining unit is used for determining the minimum time of the minimum simulation time objective function.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the present specification, each embodiment is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and may be combined according to needs, and the same similar parts may be referred to each other.
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 (7)
1. The power real-time simulation calculation task allocation method taking communication and calculation force into consideration is characterized by comprising the following steps of:
acquiring an electric power calculation task set in real-time simulation of an electromagnetic transient simulation system and a task solving topological structure of each task in the electric power calculation task set, wherein the electric power calculation task set comprises a plurality of historical current solving tasks, a plurality of equivalent conductivity matrix solving tasks and a plurality of secondary control solving tasks;
Determining the solving force of each historical current solving task, the solving force of each equivalent conductivity matrix solving task and the solving force of each secondary control solving task;
when the number of threads currently used for power calculation task allocation is larger than 1, based on the solving forces of each historical current solving task, the solving forces of each equivalent conductivity matrix solving task and the solving forces of each secondary control solving task, generating the solving force constraint conditions of a thread group formed by every two threads;
constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode based on the task solving topological structure and communication performance parameters of each CPU and each core thereof;
under the condition that each thread meets the computational constraint condition of a thread group formed by two threads, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread;
the generating the computational force constraint condition of the thread group formed by every two threads based on the computational force of each historical current solving task, the computational force of each equivalent conductivity matrix solving task and the computational force of each secondary control solving task comprises the following steps:
Determining the computing power of each thread currently used for computing task allocation;
based on the calculation force of each thread, the calculation force constraint condition of the thread group formed by every two threads is obtained by using the following formula:
wherein,for thread sCalculating the force, C hc-s Solving the calculation force of a first part of historical current solving tasks in each historical current solving task for the thread s, C cm-s Solving the calculation force of a first part of equivalent conductivity matrix solving tasks in all the equivalent conductivity matrix solving tasks for the thread s, and C sc-s Solving the calculation force of a first part of the secondary control solving tasks in each secondary control solving task for the thread s, C hc-p Solving the calculation force of a second part of historical current solving tasks in each historical current solving task for the thread p, C cm-p Solving the calculation force of a second part of equivalent conductivity matrix solving task in each equivalent conductivity matrix solving task for the thread p, C sc-p Solving the computational power of a second part of secondary control solving tasks in each secondary control solving task for the thread p, wherein q is a preset deviation threshold value, no intersection exists between the first part of historical current solving tasks and the second part of historical current solving tasks, no intersection exists between the first part of equivalent conductivity matrix solving tasks and the second part of equivalent conductivity matrix solving tasks, and no intersection exists between the first part of secondary control solving tasks and the second part of secondary control solving tasks;
The constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode based on the task solving topological structure and communication performance parameters of each CPU and each core thereof comprises the following steps:
analyzing the task solving topological structure, and determining the total number of times of communication among the CPUs and the total number of times of communication among the CPU cores;
according to the first data volume of each inter-CPU communication, the time of transmitting the first data volume between every two cores, the total number of inter-CPU communication, the second data volume of each inter-CPU communication, the time of transmitting the second data volume between every two CPUs, the total number of inter-CPU communication, the minimum simulation time objective function of the power calculation task set under the CPU multithread parallel mode is constructed.
2. The method as recited in claim 1, further comprising:
when the number of threads currently used for power calculation task allocation is 1, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are allocated to one thread currently used for power calculation task allocation.
3. The method of claim 1, wherein the constructing the minimum simulation time objective function for the power computing task set based on the first amount of data per inter-CPU communication, the total number of inter-CPU communication, the second amount of data per inter-CPU communication, and the time per second amount of data per two inter-CPU communication between each two cores, the total number of inter-CPU communication comprises:
Constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode by using the following steps:
wherein mint is the minimum simulation time of the power calculation task set, n 1 R is the total number of communication between the CPU cores core-i A first data amount, t, for the ith CPU inter-core communication core-c For the time of transmission of a first data volume unit between every two cores, n 2 R is the total number of communication between the CPUs CPU-j Second data amount, t, for jth inter-CPU communication CPU-c The time for transferring a unit of the second data amount between each two CPUs.
4. The method as recited in claim 1, further comprising:
determining an execution sequence constraint condition set of the electric power calculation task set, wherein the execution sequence constraint condition set comprises a first constraint condition, a second constraint condition and a third constraint condition, the first constraint condition is that each secondary control solving task executes calculation according to a preset task layer sequence, the second constraint condition is that the starting processing time of each secondary control solving task is later than the ending processing time of the secondary control solving task positioned before the sub-task execution sequence of the secondary control solving task, and the third constraint condition is that the execution sequence of each equivalent conductivity matrix solving task is later than the historical current solving task associated with the equivalent conductivity matrix solving task;
Under the condition that each thread meets the computational power constraint condition of a thread group formed by two threads, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread, and the method comprises the following steps:
and under the condition that each thread meets the computational power constraint condition of a thread group formed by two threads and the electric power calculation task set meets each constraint condition in the execution sequence constraint condition set, according to the minimum simulation time objective function, each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task are distributed to each thread.
5. The method of any of claims 1-4, further comprising, after constructing a minimum simulation time objective function for the set of power computation tasks in a CPU multithreaded parallel mode based on the task solution topology and the communication performance parameters of each CPU and each core thereof:
and determining the minimum time of the minimum simulation time objective function.
6. An electric power real-time simulation calculation task distribution device taking communication and calculation force into consideration is characterized by comprising:
The system comprises a solving task acquisition unit, a power computing task set and a task solving topological structure, wherein the solving task acquisition unit is used for acquiring a power computing task set in real-time simulation of an electromagnetic transient simulation system and a task solving topological structure of each task in the power computing task set, and the power computing task set comprises a plurality of historical current solving tasks, a plurality of equivalent conductivity matrix solving tasks and a plurality of secondary control solving tasks;
a solution computing force determining unit for determining a solution computing force of each historical current solution task, a solution computing force of each equivalent conductance matrix solution task, and a solution computing force of each secondary control solution task;
the computing force constraint generating unit is used for generating computing force constraint conditions of a thread group formed by every two threads based on the computing force of each historical current computing task, the computing force of each equivalent conductivity matrix computing task and the computing force of each secondary control computing task when the number of threads currently used for power computing task allocation is larger than 1;
the objective function construction unit is used for constructing a minimum simulation time objective function of the power calculation task set in a CPU multithreading parallel mode based on the task solving topological structure and communication performance parameters of each CPU and each core of the CPU;
The computing task allocation unit is used for allocating each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to each thread according to the minimum simulation time objective function under the condition that each thread meets the computational constraint condition of a thread group consisting of two threads;
the computational force constraint generation unit includes:
a thread computing force determining unit, configured to determine a computing force of each thread currently used for computing task allocation when the number of threads currently used for power computing task allocation is greater than 1;
the constraint condition determining unit is used for obtaining the calculation force constraint condition of the thread group formed by every two threads according to the following formula based on the calculation force of each thread:
wherein,for the calculation of thread s, C hc-s Solving the calculation force of a first part of historical current solving tasks in each historical current solving task for the thread s, C cm-s Solving the calculation force of a first part of equivalent conductivity matrix solving tasks in all the equivalent conductivity matrix solving tasks for the thread s, and C sc-s Solving the calculation force of a first part of the secondary control solving tasks in each secondary control solving task for the thread s, C hc-p Solving the calculation force of a second part of historical current solving tasks in each historical current solving task for the thread p, C cm-p Solving the calculation force of a second part of equivalent conductivity matrix solving task in each equivalent conductivity matrix solving task for the thread p, C sc-p Solving the computational power of a second part of secondary control solving tasks in each secondary control solving task for the thread p, wherein q is a preset deviation threshold value, no intersection exists between the first part of historical current solving tasks and the second part of historical current solving tasks, no intersection exists between the first part of equivalent conductivity matrix solving tasks and the second part of equivalent conductivity matrix solving tasks, and no intersection exists between the first part of secondary control solving tasks and the second part of secondary control solving tasks;
the objective function construction unit includes:
the first objective function construction subunit is used for analyzing the task solving topological structure and determining the total number of times of communication among the CPU and the total number of times of communication among the CPU cores;
the second objective function constructing subunit is configured to construct, according to the first data amount of each inter-CPU-core communication, a minimum simulation time objective function of the power computing task set in the multi-thread parallel mode of the CPU, where the time is a unit of the first data amount transmitted between every two cores, the total number of inter-CPU-core communication, the second data amount of each inter-CPU-core communication, and the time is a unit of the second data amount transmitted between every two CPUs.
7. The apparatus as recited in claim 6, further comprising:
the single-thread distribution unit is used for distributing each historical current solving task, each equivalent conductivity matrix solving task and each secondary control solving task to one thread currently used for power computing task distribution when the number of threads currently used for power computing task distribution is 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211007632.1A CN115344388B (en) | 2022-08-22 | 2022-08-22 | Power real-time simulation calculation task allocation method and device considering communication and calculation force |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211007632.1A CN115344388B (en) | 2022-08-22 | 2022-08-22 | Power real-time simulation calculation task allocation method and device considering communication and calculation force |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115344388A CN115344388A (en) | 2022-11-15 |
| CN115344388B true CN115344388B (en) | 2023-12-01 |
Family
ID=83954707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211007632.1A Active CN115344388B (en) | 2022-08-22 | 2022-08-22 | Power real-time simulation calculation task allocation method and device considering communication and calculation force |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115344388B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012026582A1 (en) * | 2010-08-27 | 2012-03-01 | 日本電気株式会社 | Simulation device, distributed computer system, simulation method and program |
| CN109299531A (en) * | 2018-09-12 | 2019-02-01 | 清华四川能源互联网研究院 | Electromagnetical transient emulation method and device |
| CN113835866A (en) * | 2021-10-09 | 2021-12-24 | 南方电网数字电网研究院有限公司 | Multithreading task scheduling optimization method |
| CN114911579A (en) * | 2022-06-13 | 2022-08-16 | 南方电网科学研究院有限责任公司 | Simulation parallel computing method, device and equipment for power system and readable storage medium |
-
2022
- 2022-08-22 CN CN202211007632.1A patent/CN115344388B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012026582A1 (en) * | 2010-08-27 | 2012-03-01 | 日本電気株式会社 | Simulation device, distributed computer system, simulation method and program |
| CN109299531A (en) * | 2018-09-12 | 2019-02-01 | 清华四川能源互联网研究院 | Electromagnetical transient emulation method and device |
| CN113835866A (en) * | 2021-10-09 | 2021-12-24 | 南方电网数字电网研究院有限公司 | Multithreading task scheduling optimization method |
| CN114911579A (en) * | 2022-06-13 | 2022-08-16 | 南方电网科学研究院有限责任公司 | Simulation parallel computing method, device and equipment for power system and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115344388A (en) | 2022-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10474504B2 (en) | Distributed node intra-group task scheduling method and system | |
| CN108415763B (en) | Distribution method of edge computing system | |
| US8392572B2 (en) | Method for scheduling cloud-computing resource and system applying the same | |
| US9386086B2 (en) | Dynamic scaling for multi-tiered distributed systems using payoff optimization of application classes | |
| CN109981744B (en) | Data distribution method and device, storage medium and electronic equipment | |
| CN102724103A (en) | Proxy server, hierarchical network system and distributed workload management method | |
| CN108845874A (en) | The dynamic allocation method and server of resource | |
| CN114253735B (en) | Task processing method and device and related equipment | |
| CN116302568A (en) | Computing power resource scheduling method and system, scheduling center and data center | |
| CN114610474A (en) | Multi-strategy job scheduling method and system in heterogeneous supercomputing environment | |
| CN111556516A (en) | Distributed wireless network task cooperative distribution method facing delay and energy efficiency sensitive service | |
| CN111861793A (en) | Distribution and utilization electric service distribution method and device based on cloud edge cooperative computing architecture | |
| Delavar et al. | A synthetic heuristic algorithm for independent task scheduling in cloud systems | |
| CN112130966A (en) | Task scheduling method and system | |
| CN108132840A (en) | Resource regulating method and device in a kind of distributed system | |
| CN105488134A (en) | Big data processing method and big data processing device | |
| CN115421930A (en) | Task processing method, system, device, equipment and computer readable storage medium | |
| CN114500339A (en) | Node bandwidth monitoring method and device, electronic equipment and storage medium | |
| KR101065436B1 (en) | Stochastic scheduling of a real-time parallel task with uncertain computation amount on mulit-core processors | |
| CN116932198A (en) | Resource scheduling method, device, electronic equipment and readable storage medium | |
| Zhou et al. | DPS: Dynamic pricing and scheduling for distributed machine learning jobs in edge-cloud networks | |
| CN115344388B (en) | Power real-time simulation calculation task allocation method and device considering communication and calculation force | |
| Fang et al. | Renewable prediction-driven service offloading for IoT-enabled energy systems with edge computing | |
| US20090217282A1 (en) | Predicting cpu availability for short to medium time frames on time shared systems | |
| CN117909044A (en) | Heterogeneous computing resource-oriented deep reinforcement learning cooperative scheduling method and device |
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 |