CN111695245A - Material irradiation damage spatial resolution random cluster dynamics parallel simulation method - Google Patents
Material irradiation damage spatial resolution random cluster dynamics parallel simulation method Download PDFInfo
- Publication number
- CN111695245A CN111695245A CN202010429538.XA CN202010429538A CN111695245A CN 111695245 A CN111695245 A CN 111695245A CN 202010429538 A CN202010429538 A CN 202010429538A CN 111695245 A CN111695245 A CN 111695245A
- Authority
- CN
- China
- Prior art keywords
- reaction
- updating
- defect
- sector
- communication data
- 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
Links
Images
Classifications
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a dynamic parallel simulation method for spatially resolved random clusters of material irradiation damage, which can realize large-scale SRSCD simulation. The method comprises the following steps: distributing the simulated volume of the space resolution random cluster dynamics to different processes, establishing a three-dimensional Cartesian topological structure of the processes, and dividing the process area into a plurality of sectors; determining a communication data structure; and traversing each sector in sequence according to the serial number sequence of the sectors, calculating the time threshold of the internal cycle of each sector, entering the internal cycle, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling a communication data list and updating the related reaction, and when the evolution time of the internal cycle is greater than the time threshold of the internal cycle, communicating with the neighbor process, synchronizing the corresponding defect information and updating the related reaction. The invention relates to the technical field of nuclear material irradiation damage simulation and the field of parallel computation.
Description
Technical Field
The invention relates to the technical field of nuclear material irradiation damage simulation and the field of parallel computation, in particular to a material irradiation damage space resolution random cluster dynamics parallel simulation method.
Background
The service performance of the structural material in the nuclear reactor mainly depends on the dynamic behaviors of defects generated by irradiation, such as diffusion, nucleation, growth, annihilation and the like of the defects, and the evolution behaviors in a long time scale can cause the size and density distribution of the defects to change, thereby causing the degradation of the mechanical properties of the structural material, such as radiation hardening/embrittlement, radiation swelling and the like. Since the evolution behavior of these long time scales often exceeds the capability range of micro scale simulation methods such as Molecular Dynamics (MD), Atomic Kinetic Monte Carlo (AKMC), etc., a mesoscale method is usually required for modeling and simulation.
The object kinetic Monte Carlo method (OKMC) and the Cluster kinetic method (Cluster Dynamics, CD) developed based on the mean field rate theory are two widely used mesoscale simulation methods. The former simulates the evolution of defects by tracking the random diffusion and interaction of each defect within the material; the latter assumes that the defects are in an isotropic homogeneous medium and follows the change in defect concentration over time by an equation of the rate of defect reaction. The OKMC can simulate complex behavior among various defects and can capture spatial correlation among defects, but is generally limited by the computation time and amount required for the simulation of the complex behavior among the defects. The CD method generally has high computational efficiency and can simulate high irradiation dose and time scale, however, the method is usually limited to simulate a system with few movable defects, and the defect evolution of a complex system is difficult to process.
Space Resolved random cluster dynamics (SRSCD) is a new method developed in recent years for simulating irradiation defect behavior of structural materials. It divides the simulation volume (which may also be called a simulation region) into a number of volume elements (cubes) within each of which defects can be aggregated and decomposed, assuming a uniform distribution of defects, while there are concentration differences and resulting diffusion between the volume elements. The reaction rates of the various reactions between defects are deduced from the classical CD method, and the choice of reaction and the time increment are determined by the classical KMC algorithm. The SRSCD method avoids the limitations of CD in defect type and behavior complexity on one hand, and reduces the computational requirements compared to OKMC on the other hand, but its simulation volume is limited by the computational effort brought by the KMC algorithm.
In order to enlarge the simulation volume of the SRSCD, the presenter of the SRSCD adopts a synchronous parallel KMC algorithm to realize the parallelization of the SRSCD, namely, the synchronous parallel SRSCD. However, this parallel approach is inefficient in parallelism and cannot be effectively extended to larger analog volumes, since it requires synchronization at every KMC step.
Disclosure of Invention
The invention aims to solve the technical problem of providing a dynamic parallel simulation method for spatially resolving random clusters of material irradiation damage, so as to solve the problems of low parallel efficiency and incapability of large-scale parallel simulation in the prior art.
In order to solve the above technical problem, an embodiment of the present invention provides a parallel simulation method for spatially resolved random cluster dynamics of material irradiation damage, including:
distributing the simulated volume of the space resolution random cluster dynamics to different processes, establishing a three-dimensional Cartesian topological structure of the processes, and dividing the process area into a plurality of sectors;
determining a communication data structure;
and traversing each sector in sequence according to the serial number sequence of the sectors, calculating the time threshold of the internal cycle of each sector, entering the internal cycle, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling a communication data list and updating the related reaction, and when the evolution time of the internal cycle is greater than the time threshold of the internal cycle, communicating with the neighbor process, synchronizing the corresponding defect information and updating the related reaction.
Further, the allocating the simulated volume of the spatially resolved random cluster dynamics to different processes and establishing a three-dimensional cartesian topology of the processes, and dividing the process area into a plurality of sectors includes:
distributing the simulated volume of the space resolution random cluster dynamics to different processes and establishing a three-dimensional Cartesian topological structure of the processes in a space division mode;
each process creates respective local volume element according to the given volume element number, and initializes the association relation between the volume element and the volume element on each process;
dividing the area of each process into a plurality of sectors and numbering;
and establishing a mapping relation between the volume element and the sector.
Further, each sector has at least two volume elements in each dimension, i.e., each process has at least four volume elements in each dimension.
Furthermore, the products of the injection reaction are cascade defects, and the reactants and products of the non-injection reaction are common defects;
the determining a communication data structure comprises:
if the defect is a common defect, constructing a communication data structure comDefectSwap;
and if the defect is a cascade defect, constructing a communication data structure cascadeDefect Swap.
Further, the communication data structure comdfectswap includes: sendProc, numSendCuff, sendBuff, recvProc, numRecv, and recvBuff;
wherein sendProc represents the number of the destination process to send, numSend represents the number of the data to send, sendBuff represents the transmit data buffer, recvProc represents the number of the source process to receive, numRecv represents the number of the data to receive and recvBuff represents the receive data buffer.
Further, the communication data structure cascadedeffectswap includes: cascadeCell, numCells, sendProc, numSend, sendInd, sendDindex, sendBuff, recvProc, numRecv, and recvBuff;
wherein cascadeCell indicates the volume element identification of the injection reaction, numCells indicates the volume element number of the injection reaction, sendProc indicates the destination process number of sending, numSend indicates the number of sending data, sendIndex indicates the index of the data sending buffer, sendBuff indicates the data sending buffer, recvProc indicates the source process number of receiving, numRecv indicates the number of receiving data, and recvBuff indicates the data receiving buffer.
Further, sequentially traversing each sector according to the serial number sequence of the sectors, calculating a time threshold of an inner loop of each sector, entering the inner loop, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling a communication data list and updating a related reaction, communicating with a neighbor process when the evolution time of the inner loop is greater than the time threshold of the inner loop, and synchronizing corresponding defect information and updating the related reaction comprise:
initializing the evolution time t of the outer loop and entering the outer loop:
sequentially traversing each sector according to the serial number sequence of the sectors, and calculating the time threshold tau of each internal cycle;
initializing the evolution time kmc _ time of the inner loop, entering the inner loop of the current sector:
calculating a time increment delta t according to the total reaction rate on the current sector;
randomly selecting a response in the current sector;
updating defects and filling a communication data list according to the types of reactants and products participating in the reaction, and updating corresponding reactions according to the defects related to the reactions;
accumulating the time increment delta t by the evolution time kmc _ time of the internal loop, and judging whether the current kmc _ time is greater than the time threshold tau of the internal iteration;
if yes, synchronizing the common defects and updating related reactions, and synchronizing the cascade defects and updating related reactions; if not, returning to execute the operation of calculating the time increment delta t according to the total reaction rate on the current sector;
accumulating tau by the evolution time T of the extrinsic cycle, and judging whether the current evolution time T is greater than an extrinsic cycle time threshold T or not;
if yes, finishing the outer circulation; otherwise, returning to execute the operation of sequentially traversing each sector according to the serial number sequence of the sectors.
Further, the updating the defect according to the reactant and product types participating in the reaction, populating the communication data list, and updating the corresponding reaction according to the defect involved in the reaction includes:
updating corresponding defects according to the types of reactants and products participating in the reaction;
if the reaction occurs in the boundary volume element, judging whether the selected reaction is an injection reaction;
if the reaction is injected, judging whether the product is in the Boundary area; the Boundary area is an area formed by volume elements which are positioned in a neighbor process and adjacent to the current process;
if yes, setting the value of the corresponding volume element number in a cascadeCell array of the communication data structure cascadeDefectSwap as '1';
if not, updating the reaction related to the defect according to the defect related to the reaction.
Further, the method further comprises:
if not, judging whether the reactant/product is in a Boundary/Ghost area; the Boundary area is an area formed by volume elements which are positioned in the current process and adjacent to the neighbor process; the Ghost area is an area formed by volume elements which are positioned in a neighbor process and adjacent to the current process;
if yes, filling the defects related to the reaction into a communication data structure comDefectSwap;
if not, updating the reaction related to the defect according to the defect related to the reaction.
Further, the synchronizing the common defects and updating the related reactions includes:
establishing non-blocking communication between the current process and the neighbor process, and exchanging defect information through a communication data structure commDefect Swap;
updating the received defect information into a corresponding volume element;
updating a reaction associated therewith based on the received defect information;
the synchronizing cascade defects and updating the related reactions comprises:
filling all defects in the volume element with the median value of '1' in the cascadeCell array into a communication data structure cascadeDefect Swap;
establishing non-blocking communication between the current process and the neighbor process, and exchanging defect information through a communication data structure cascadeDefect swap;
updating the received defect information into a corresponding volume element;
the response associated therewith is updated based on the received defect information.
The technical scheme of the invention has the following beneficial effects:
in the scheme, the simulation volume of the space resolution random cluster dynamics is allocated to different processes, a three-dimensional Cartesian topological structure of the processes is established, and the process area is divided into a plurality of sectors; determining a communication data structure; traversing each sector in sequence according to the serial number sequence of the sectors, calculating the time threshold of the internal cycle of each sector, entering the internal cycle, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling the communication data list and updating the related reaction, and when the evolution time of the internal cycle is greater than the time threshold of the internal cycle, communicating with the neighbor process, synchronizing the corresponding defect information and updating the related reaction; therefore, the SL algorithm is introduced into the SRSCD, the SRSCD parallelism is realized, the higher parallelism efficiency is achieved, the SRSCD can be effectively expanded into the large-volume simulation, and the large-scale SRSCD simulation is realized.
Drawings
Fig. 1 is a schematic flow chart of a method for parallel simulation of dynamics of spatially resolved random clusters of material irradiation damage according to an embodiment of the present invention;
fig. 2 is a detailed flowchart schematic diagram of a material irradiation damage spatial resolution random cluster dynamics parallel simulation method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of simulation volume division and sector division of SL _ SRSCD provided in the embodiment of the present invention under a two-dimensional condition;
FIG. 4 is a schematic diagram of 6 communication directions provided by an embodiment of the present invention;
fig. 5 is a schematic diagram of a communication data structure commDefectSwap in the SL _ SRSCD according to the embodiment of the present invention;
fig. 6 is a schematic diagram of a communication data structure cascadeffectswap in SL _ SRSCD according to an embodiment of the present invention;
fig. 7 is a communication schematic diagram of the SL _ SRSCD method provided in the embodiment of the present invention in a two-dimensional situation.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
The invention provides a dynamic parallel simulation method for spatially resolving random clusters of material irradiation damage, aiming at the problems that the existing parallel efficiency is low and large-scale parallel simulation cannot be carried out.
In order to better understand the material irradiation damage spatial resolution random cluster dynamics parallel simulation method according to the embodiment of the present invention, some terms are first described as follows:
for each type of reaction, the defects participating in the reaction are called reactants (reactants), the defects generated by the reaction are products (products), and the reactions among the defects mainly comprise five types:
(1) injection reaction (Implantation): indicating the generation of initial defects and no reactant. If electron irradiation, the product is two point defects (vacancies and self-interstitials), and if neutron irradiation, the product is a plurality of (more than 2) defects, including point defects and small clusters;
(2) decomposition reaction (Dissociation): the defective cluster emits out of a point defect. The reactant is one defect, the product is two defects, and one of the defects is a point defect;
(3) trap absorption reaction (Sink _ Removal): defects disappear by being absorbed by inherent defects (such as grain boundaries, dislocations and the like) in the material, and the inherent defects are called 'wells'; reactants are one defect and products are zero;
(4) aggregation reaction (Clustering): defect growth and recombination (recombination of vacancy defects and self-interstitial defects); the reactant is two defects, the product is generally one defect;
(5) diffusion reaction (Diffusion): defects migrate from one volume element to another; both reactants and products are a drawback;
wherein only diffusion reactions occur between two volume elements and the remaining reactions occur within the volume elements.
And secondly, a Synchronous Sublattice (SL) algorithm, which is used for avoiding collision among processors under the condition of allowing errors, dividing a region on each processor (corresponding to a process hereinafter) into sub-regions (corresponding to sectors hereinafter) and numbering the sub-regions, wherein each processor simultaneously and independently executes KMC events on the sub-regions corresponding to the same number, and communicating processors with boundary events until the execution time of the next event exceeds a time threshold tau. Thus, the parallel efficiency of SL algorithms is generally much higher than synchronous parallel KMC algorithms, which are synchronized at each step.
In this embodiment, an SL algorithm with higher parallelism efficiency is introduced into the SRSCD, so that the SRSCD massive parallel simulation method, SL _ SRSCD, for the material irradiation damage spatial resolution random cluster dynamics parallel simulation method according to the embodiment of the present invention is actually an SRSCD massive parallel simulation method based on the SL algorithm, so that the SRSCD can realize three-dimensional massive parallel simulation under the condition of introducing a certain error. This error in the SL algorithm is allowed and can be physically interpreted as the dependence of an event somewhere on a more distant location in the system is smaller, so that the events can be spatially decoupled. Therefore, it is feasible to extend the SL algorithm to SRSCD.
As shown in fig. 1, a method for parallel simulation of dynamics of spatially resolved random clusters of material irradiation damage provided by an embodiment of the present invention includes:
s101, distributing the simulated volume of the space resolution random cluster dynamics to different processes, establishing a three-dimensional Cartesian topological structure of the processes, and dividing the process area into a plurality of sectors;
s102, determining a communication data structure;
s103, traversing each sector in sequence according to the serial number sequence of the sectors, calculating the time threshold of the internal circulation of each sector, entering the internal circulation, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling the communication data list and updating the related reaction, communicating with the neighbor process when the evolution time of the internal circulation is greater than the time threshold of the internal circulation, synchronizing the corresponding defect information and updating the related reaction.
The material irradiation damage spatial resolution random cluster dynamics parallel simulation method disclosed by the embodiment of the invention allocates a simulation volume of spatial resolution random cluster dynamics to different processes, establishes a three-dimensional Cartesian topological structure of the processes, and divides the process area into a plurality of sectors; determining a communication data structure; traversing each sector in sequence according to the serial number sequence of the sectors, calculating the time threshold of the internal cycle of each sector, entering the internal cycle, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling the communication data list and updating the related reaction, and when the evolution time of the internal cycle is greater than the time threshold of the internal cycle, communicating with the neighbor process, synchronizing the corresponding defect information and updating the related reaction; therefore, the SL algorithm is introduced into the SRSCD, the SRSCD parallelism is realized, the higher parallelism efficiency is achieved, the SRSCD can be effectively expanded into the large-volume simulation, and the large-scale SRSCD simulation is realized.
In a specific embodiment of the foregoing method for parallel simulation of material irradiation damage spatially resolved random cluster dynamics, further, the assigning a simulated volume of spatially resolved random cluster dynamics to different processes and establishing a three-dimensional cartesian topology of the processes, and dividing a region of the processes into a plurality of sectors (S101) includes:
distributing the simulated volume of the space resolution random cluster dynamics to different processes and establishing a three-dimensional Cartesian topological structure of the processes in a space division mode;
each process creates respective local volume element according to the given volume element number, and initializes the association relation between the volume element and the volume element on each process;
dividing the area of each process into a plurality of sectors (sectors) and numbering; in general, the two-dimensional division is 4 sectors; three-dimensionally dividing the three-dimensional space into 8 sectors;
and establishing a mapping relation between the volume element and the sector.
In this embodiment, initializing the association relationship between the volume element and the volume element in each process refers to storing, for each volume element, the local number of its neighboring volume elements in the left, right, front, back, up, and down 6 directions.
In this embodiment, as shown in fig. 2, the purpose of S101 is to initialize a simulation volume, fig. 3 is a schematic diagram of region (simulation volume) partition and sector partition in a two-dimensional case of SL _ SRSCD, where P0 to P8 are process numbers; 1-4 are sector numbers; coarse implementation as a process boundary; the dotted line is the sector boundary; the thin implementation is a voxel boundary. Taking process P4 as an example, the dotted line is a communication area of P4, where light gray is a Ghost area (the Ghost area is an area formed by volume elements located in a neighboring process and adjacent to the current process (P4)), and dark gray is a Boundary area (the Boundary area is an area formed by volume elements located in the current process (P4) and adjacent to the neighboring process).
In this embodiment, an actual simulation space is three-dimensional, and therefore, a three-dimensional cartesian topology needs to be established for the process, an area in the process is further divided into 8 sectors, the number of each sector is 1 to 8, and two sectors are respectively provided in X, Y, Z directions, and then local numbers of neighbor processes, sectors, and volume elements in 6 communication directions shown in fig. 4 are stored for each process, sector, and volume element. Fig. 4 is a schematic diagram of 6 communication directions, which are also 6 directions of defect propagation. Since diffusion in SRSCD is caused by concentration gradients, only the direction of diffusion adjacent to a face is needed to communicate with the process adjacent to the face.
In this embodiment, in order to avoid the conflict caused by deleting and adding the same defect in the same volume element, when the SL _ SRSCD method is used, a constraint is required: it is ensured that each dimension of each sector contains at least two volume elements, i.e. each dimension of each process contains at least four volume elements.
In the specific implementation of the material irradiation damage spatial resolution random cluster dynamics parallel simulation method, further, the product of the injection reaction is a cascade defect, and the reactant and the product of the non-injection reaction are common defects;
the determining a communication data structure comprises:
if the defect is a common defect, constructing a communication data structure comDefectSwap;
and if the defect is a cascade defect, constructing a communication data structure cascadeDefect Swap.
In this embodiment, taking simulated neutron irradiation as an example, in the simulated neutron irradiation, the injection reaction may generate a plurality of defects in the SRSCD simulation system, and in order to distinguish these defects from defects participating in other reactions, the product of the injection reaction is referred to as "cascade defect" in SL _ SRSCD, and the reactant and product of other reactions are referred to as "normal defect".
In a specific embodiment of the foregoing method for spatially resolved stochastic cluster dynamics parallel simulation of material irradiation damage, further, the communication data structure comdfectswap includes: sendProc, numSendCuff, sendBuff, recvProc, numRecv, and recvBuff;
wherein sendProc represents the number of the destination process to send, numSend represents the number of the data to send, sendBuff represents the transmit data buffer, recvProc represents the number of the source process to receive, numRecv represents the number of the data to receive and recvBuff represents the receive data buffer.
FIG. 5 is a schematic diagram of a communication data structure comDefectSwap in SL _ SRSCD, wherein send 1-3 represent processes of sending data to 3 adjacent surfaces; recv 1-3 represent that data is received from 3 adjacent neighbor processes on the opposite direction; the defectType [ ] in sendBuff represents the defect type of the defect, num represents the number of the defects to be updated (num >0 represents that the defect is a product, num <0 represents that the defect is a reactant), cell represents the local number of the volume element where the defect is located, if the defect is in a Boundary area, dir and neighbor are both equal to 0, if the defect is in a Ghost area, the cell is indicated to be the volume element in the Ghost area, dir is equal to the direction of the Ghost area, and neighbor is equal to the local volume element number adjacent to the cell; the significance of the deffectType [ ], num and cell in recvBuff is the same as that of the sending buffer area, recvDir is the reverse direction of dir, and recvNeighbor is the volume element number in the Ghost area.
In a specific embodiment of the foregoing method for parallel simulation of spatially resolved random cluster dynamics of material irradiation damage, further, the communication data structure cascadeffectswap includes: cascadeCell, numCells, sendProc, numSend, sendInd, sendDindex, sendBuff, recvProc, numRecv, and recvBuff;
wherein, cascadeCell represents the volume element identification of the injection reaction (migration), numCells represents the volume element number of the injection reaction, sendProc represents the destination process number of the transmission, numend represents the number of the transmitted data, sendIndex represents the index of the data transmission buffer area, sendBuff represents the data transmission buffer area, recvProc represents the source process number of the reception, numRecv represents the number of the reception data and recvBuff represents the data reception buffer area.
FIG. 6 is a schematic diagram of a communication data structure cascadeDefectSwap in SL _ SRSCD, wherein send 1-3 represent sending data to 3 neighbor processes; recv 1-3 represents that data is received from 3 neighbor processes in the opposite direction; when data is sent, all defects in all volume elements in a Boundary region in a certain direction, which are subjected to injection reaction, need to be filled into sendBuff; for each volume element with the median value of '1' in the cascadeCell array, filling an index (index) of the volume element in sendBuff, filling the defects in the volume element in sequence, filling the next volume element in sequence and going on in sequence; the index includes: volume element number (cell), neighbor volume element number (neighbor cell) located in the Ghost area, total number of defect types (numDefects) in the volume element, and total number of boundary volume elements (numCells) where injection reaction occurs; after index is filled, the defect (defect) in the volume element is filled in sequence. When receiving data, the next index can be located according to numCells and numDefects in the previous index, and then the defect is taken out and updated to the defect with the number equal to the neighborCell.
In a specific embodiment of the foregoing material irradiation damage spatial resolution random cluster dynamics parallel simulation method, further, as shown in fig. 2, sequentially traversing each sector according to a sector number sequence, calculating a time threshold of an inner loop of each sector, entering an inner loop, randomly selecting a reaction in a current sector, updating a defect according to the selected reaction, filling a communication data list, and updating a related reaction, when an evolution time of the inner loop is greater than the time threshold of the inner loop, communicating with a neighboring process, and synchronizing corresponding defect information and updating the related reaction includes:
initializing the evolution time t of the outer loop, e.g., t ═ 0, and entering the outer loop:
sequentially traversing each sector according to the serial number sequence of the sectors, and calculating the time threshold tau of each internal cycle;
initializing the evolution time kmc _ time of the inner loop, for example, kmc _ time is 0, entering the inner loop of the current sector:
calculating a time increment delta t according to the total reaction rate on the current sector;
randomly selecting a response in the current sector;
updating defects and filling a communication data list according to the types of reactants and products participating in the reaction, and updating corresponding reactions according to the defects related to the reactions;
accumulating the time increment delta t by the evolution time kmc _ time of the internal loop, and judging whether the current kmc _ time is greater than the time threshold tau of the internal iteration;
if yes, synchronizing the common defects and updating related reactions, and synchronizing the cascade defects and updating related reactions; if not, returning to execute the operation of calculating the time increment delta t according to the total reaction rate on the current sector;
accumulating tau by the evolution time T of the extrinsic cycle, and judging whether the current evolution time T is greater than an extrinsic cycle time threshold T (namely the time of the whole simulation process);
if yes, finishing the outer circulation; otherwise, returning to execute the operation of sequentially traversing each sector according to the serial number sequence of the sectors.
In this embodiment, there is a corresponding average latency for each sector, and the time threshold τ is determined by the maximum average latency in all sectors over all processes.
In this embodiment, after entering the outer loop, the calculation on 8 sectors is sequentially performed according to the sector number sequence, one sector corresponds to one inner loop, if the time threshold T of the outer loop is not reached after the 8 sectors are completely performed, the 8 sectors are traversed again according to the sector number sequence, and if the inner loop of one sector is not finished yet and the current evolution time T reaches the time threshold T, the inner loop is skipped, the corresponding defect information is synchronized, and the whole loop process is ended.
In an embodiment of the foregoing method for parallel simulation of spatially resolved random cluster dynamics of material irradiation damage, further, the updating defects according to types of reactants and products participating in the reaction, populating a communication data list, and updating corresponding reactions according to defects involved in the reactions includes:
updating corresponding defects according to the types of reactants and products participating in the reaction;
if the reaction occurs in the boundary volume element, judging whether the selected reaction is an injection reaction;
if the reaction is injected, judging whether the product is in the Boundary area;
if so, setting the value of the corresponding volume element number in the cascadeCell array of the communication data structure cascadeDefectSwap to be 1, and recording the volume element number of the reaction;
if not, the reaction related to the reaction is updated according to the defect related to the reaction (the five reactions are referred to above).
In an embodiment of the foregoing method for spatially resolved stochastic cluster dynamics parallel simulation of material irradiation damage, the method further includes:
if not, judging whether the reactant/product is in a Boundary/Ghost area;
if yes, filling the defects related to the reaction into a communication data structure comDefectSwap;
if not, the reaction related to the reaction is updated according to the defect related to the reaction (the five reactions are referred to above).
In an embodiment of the foregoing method for parallel simulation of dynamics of spatially resolved random clusters of material irradiation damage, further, the synchronizing common defects and updating relevant reactions includes:
the method comprises the following steps that non-blocking communication is established between a current process and a neighbor process, and defect information is exchanged through a communication data structure comDefectSwap, specifically: establishing non-blocking communication with a neighbor process by using a non-blocking communication function MPI _ Isend/MPI _ Irecv of the MPI, and exchanging 'common defects' through a communication data structure commDefect swap, as shown in FIG. 7 (b);
updating the received defect information (i.e., the defect in the receive buffer) into the corresponding volume element;
updating the reaction (referring to the above five types of reactions) related to the received defect information according to the received defect information;
the synchronizing cascade defects and updating the related reactions comprises:
filling all defects in the volume element with the median value of "1" in the cascadeCell array into a communication data structure cascadeDefect Swap according to the structure of the graph 6;
the current process and the neighbor process establish non-blocking communication, and defect information is exchanged through a communication data structure cascadedefectSwap, specifically: establishing non-blocking communication with a neighbor process by using a non-blocking communication function MPI _ Isend/MPI _ Irecv of MPI, and exchanging cascade defects through a communication data structure cascadeDefect swap, as shown in FIG. 7 (c);
updating the received defect information into the corresponding volume element, specifically: according to the received cell number, deleting the existing defects, and then adding the received defects into the cell;
the responses associated with the received defect information (referred to as the five types of responses above) are updated based on the received defect information.
In this embodiment, fig. 7 is a communication schematic diagram of the SL _ SRSCD method under a two-dimensional condition, where P0 to P8 are process numbers, 1 to 4 are sector numbers, and a dotted line indicates an area requiring communication. Taking sector 1 communication on process P4 as an example, fig. 7(a) is a schematic diagram illustrating process P4 communicating with four neighbor processes; fig. 7(b) is a communication diagram of P4 exchanging a "common defect" with a neighbor process through a comdeffectswap, where data of a Ghost area and a Boundary area need to be communicated, and the two areas are merged for transmission/reception; fig. 7(c) is a communication diagram that P4 and a neighbor process exchange "cascade defect" through cascadedeffectswap, and only need to send boundry region data and receive Ghost region data.
Compared with the prior art, the invention has the beneficial effects that:
(1) the SL _ SRSCD method has higher parallel efficiency, can effectively expand the SRSCD into large-volume simulation, and realizes large-scale SRSCD simulation;
(2) by means of the constructed communication data structure and the communication mode (each sector is subjected to communication after calculation (referring to inner circulation, namely operations of selecting reaction, updating defects, filling a communication data list and updating related reaction) is finished, and is non-blocking communication), redundant communication is effectively reduced, calculation and communication are overlapped through the non-blocking communication, synchronous waiting time brought by communication is further reduced, and parallel efficiency is improved.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (10)
1. A parallel simulation method for spatially resolved random cluster dynamics of material irradiation damage is characterized by comprising the following steps:
distributing the simulated volume of the space resolution random cluster dynamics to different processes, establishing a three-dimensional Cartesian topological structure of the processes, and dividing the process area into a plurality of sectors;
determining a communication data structure;
and traversing each sector in sequence according to the serial number sequence of the sectors, calculating the time threshold of the internal cycle of each sector, entering the internal cycle, randomly selecting a reaction in the current sector, updating the defect according to the selected reaction, filling a communication data list and updating the related reaction, and when the evolution time of the internal cycle is greater than the time threshold of the internal cycle, communicating with the neighbor process, synchronizing the corresponding defect information and updating the related reaction.
2. The method of claim 1, wherein the assigning the simulated volume of spatially resolved random cluster dynamics to different processes and establishing a three-dimensional cartesian topology of the processes, and the dividing the process area into sectors comprises:
distributing the simulated volume of the space resolution random cluster dynamics to different processes and establishing a three-dimensional Cartesian topological structure of the processes in a space division mode;
each process creates respective local volume element according to the given volume element number, and initializes the association relation between the volume element and the volume element on each process;
dividing the area of each process into a plurality of sectors and numbering;
and establishing a mapping relation between the volume element and the sector.
3. The method of claim 1, wherein each sector has at least two volume elements per dimension, i.e. each dimension of each process has at least four volume elements.
4. The material irradiation damage spatial resolution random cluster dynamics parallel simulation method according to claim 1, wherein the product of the injection reaction is a cascade defect, and the reactant and the product of the non-injection reaction are common defects;
the determining a communication data structure comprises:
if the defect is a common defect, constructing a communication data structure comDefectSwap;
and if the defect is a cascade defect, constructing a communication data structure cascadeDefect Swap.
5. The method for spatially resolved stochastic cluster dynamics parallel simulation of material irradiation damage according to claim 4, wherein the communication data structure commDefectSwap comprises: sendProc, numSendCuff, sendBuff, recvProc, numRecv, and recvBuff;
wherein sendProc represents the number of the destination process to send, numSend represents the number of the data to send, sendBuff represents the transmit data buffer, recvProc represents the number of the source process to receive, numRecv represents the number of the data to receive and recvBuff represents the receive data buffer.
6. The method according to claim 4, wherein the communication data structure cascadeffectswap comprises: cascadeCell, numCells, sendProc, numSend, sendInd, sendDindex, sendBuff, recvProc, numRecv, and recvBuff;
wherein cascadeCell indicates the volume element identification of the injection reaction, numCells indicates the volume element number of the injection reaction, sendProc indicates the destination process number of sending, numSend indicates the number of sending data, sendIndex indicates the index of the data sending buffer, sendBuff indicates the data sending buffer, recvProc indicates the source process number of receiving, numRecv indicates the number of receiving data, and recvBuff indicates the data receiving buffer.
7. The material irradiation damage spatial resolution random cluster dynamics parallel simulation method of claim 1, wherein traversing each sector sequentially according to a sector number sequence, calculating a time threshold of an inner loop of each sector to enter the inner loop, randomly selecting a reaction in a current sector, updating a defect according to the selected reaction, filling a communication data list and updating a related reaction, communicating with a neighbor process when an evolution time of the inner loop is greater than the time threshold of the inner loop, and synchronizing corresponding defect information and updating the related reaction comprises:
initializing the evolution time t of the outer loop and entering the outer loop:
sequentially traversing each sector according to the serial number sequence of the sectors, and calculating the time threshold tau of each internal cycle;
initializing the evolution time kmc _ time of the inner loop, entering the inner loop of the current sector:
calculating a time increment delta t according to the total reaction rate on the current sector;
randomly selecting a response in the current sector;
updating defects and filling a communication data list according to the types of reactants and products participating in the reaction, and updating corresponding reactions according to the defects related to the reactions;
accumulating the time increment delta t by the evolution time kmc _ time of the internal loop, and judging whether the current kmc _ time is greater than the time threshold tau of the internal iteration;
if yes, synchronizing the common defects and updating related reactions, and synchronizing the cascade defects and updating related reactions; if not, returning to execute the operation of calculating the time increment delta t according to the total reaction rate on the current sector;
accumulating tau by the evolution time T of the extrinsic cycle, and judging whether the current evolution time T is greater than an extrinsic cycle time threshold T or not;
if yes, finishing the outer circulation; otherwise, returning to execute the operation of sequentially traversing each sector according to the serial number sequence of the sectors.
8. The method for spatially resolved stochastic cluster dynamics parallel simulation of material irradiation damage according to claim 7, wherein the updating the defect according to the types of reactants and products participating in the reaction, populating the communication data list, and updating the corresponding reaction according to the defect involved in the reaction comprises:
updating corresponding defects according to the types of reactants and products participating in the reaction;
if the reaction occurs in the boundary volume element, judging whether the selected reaction is an injection reaction;
if the reaction is injected, judging whether the product is in the Boundary area; the Boundary area is an area formed by volume elements which are positioned in a neighbor process and adjacent to the current process;
if yes, setting the value of the corresponding volume element number in a cascadeCell array of the communication data structure cascadeDefectSwap as '1';
if not, updating the reaction related to the defect according to the defect related to the reaction.
9. The method of parallel simulation of spatially resolved random cluster dynamics of material irradiation damage according to claim 8, further comprising:
if not, judging whether the reactant/product is in a Boundary/Ghost area; the Boundary area is an area formed by volume elements which are positioned in the current process and adjacent to the neighbor process; the Ghost area is an area formed by volume elements which are positioned in a neighbor process and adjacent to the current process;
if yes, filling the defects related to the reaction into a communication data structure comDefectSwap;
if not, updating the reaction related to the defect according to the defect related to the reaction.
10. The method of claim 7, wherein the synchronizing common defects and updating associated responses comprises:
establishing non-blocking communication between the current process and the neighbor process, and exchanging defect information through a communication data structure commDefect Swap;
updating the received defect information into a corresponding volume element;
updating a reaction associated therewith based on the received defect information;
the synchronizing cascade defects and updating the related reactions comprises:
filling all defects in the volume element with the median value of '1' in the cascadeCell array into a communication data structure cascadeDefect Swap;
establishing non-blocking communication between the current process and the neighbor process, and exchanging defect information through a communication data structure cascadeDefect swap;
updating the received defect information into a corresponding volume element;
the response associated therewith is updated based on the received defect information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010429538.XA CN111695245B (en) | 2020-05-20 | 2020-05-20 | Parallel simulation method for material irradiation damage space resolution random cluster dynamics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010429538.XA CN111695245B (en) | 2020-05-20 | 2020-05-20 | Parallel simulation method for material irradiation damage space resolution random cluster dynamics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111695245A true CN111695245A (en) | 2020-09-22 |
| CN111695245B CN111695245B (en) | 2023-04-28 |
Family
ID=72478037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010429538.XA Active CN111695245B (en) | 2020-05-20 | 2020-05-20 | Parallel simulation method for material irradiation damage space resolution random cluster dynamics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111695245B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668156A (en) * | 2020-12-14 | 2021-04-16 | 北京遥感设备研究所 | Method for calculating storage life of non-metal part |
| CN112885414A (en) * | 2021-03-31 | 2021-06-01 | 中国核动力研究设计院 | Irradiation damage simulation system and method based on rate theory and cluster dynamics |
| CN113239522A (en) * | 2021-04-20 | 2021-08-10 | 四川大学 | Atmospheric pollutant diffusion simulation method based on computer cluster |
| CN114490047A (en) * | 2022-01-12 | 2022-05-13 | 北京科技大学 | Heterogeneous data transmission method for nuclear fuel fission gas cluster dynamic simulation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180174695A1 (en) * | 2016-12-15 | 2018-06-21 | Westinghouse Electric Company, Llc | Integration of real-time measurements and atomistic modeling to license nuclear components |
| CN109727647A (en) * | 2018-12-29 | 2019-05-07 | 中国原子能科学研究院 | A kind of analogy method and system that irradiation causes material structure to develop |
| CN110459269A (en) * | 2019-08-07 | 2019-11-15 | 中国原子能科学研究院 | A kind of multi-scale coupling analogy method of nuclear reactor material irradiation damage |
| CN110457810A (en) * | 2019-08-07 | 2019-11-15 | 中国原子能科学研究院 | The Rate Theory Parallel Simulation method that reactor critical material vacancy, gap develop |
-
2020
- 2020-05-20 CN CN202010429538.XA patent/CN111695245B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180174695A1 (en) * | 2016-12-15 | 2018-06-21 | Westinghouse Electric Company, Llc | Integration of real-time measurements and atomistic modeling to license nuclear components |
| CN110024047A (en) * | 2016-12-15 | 2019-07-16 | 西屋电气有限责任公司 | Permit the real-time measurement of nuclear component and the integration of atom modeling |
| CN109727647A (en) * | 2018-12-29 | 2019-05-07 | 中国原子能科学研究院 | A kind of analogy method and system that irradiation causes material structure to develop |
| CN110459269A (en) * | 2019-08-07 | 2019-11-15 | 中国原子能科学研究院 | A kind of multi-scale coupling analogy method of nuclear reactor material irradiation damage |
| CN110457810A (en) * | 2019-08-07 | 2019-11-15 | 中国原子能科学研究院 | The Rate Theory Parallel Simulation method that reactor critical material vacancy, gap develop |
Non-Patent Citations (2)
| Title |
|---|
| DANDAN CHEN ET AL.: "An overview: multiscale simulation in understanding the radiation damage accumulation of reactor materials" * |
| JAIME MARIAN ET AL.: "Stochastic cluster dynamics method for simulations of multispecies irradiation damage accumulation" * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668156A (en) * | 2020-12-14 | 2021-04-16 | 北京遥感设备研究所 | Method for calculating storage life of non-metal part |
| CN112668156B (en) * | 2020-12-14 | 2023-06-20 | 北京遥感设备研究所 | Method for calculating storage life of nonmetallic part |
| CN112885414A (en) * | 2021-03-31 | 2021-06-01 | 中国核动力研究设计院 | Irradiation damage simulation system and method based on rate theory and cluster dynamics |
| CN113239522A (en) * | 2021-04-20 | 2021-08-10 | 四川大学 | Atmospheric pollutant diffusion simulation method based on computer cluster |
| CN113239522B (en) * | 2021-04-20 | 2022-06-28 | 四川大学 | Atmospheric pollutant diffusion simulation method based on computer cluster |
| CN114490047A (en) * | 2022-01-12 | 2022-05-13 | 北京科技大学 | Heterogeneous data transmission method for nuclear fuel fission gas cluster dynamic simulation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111695245B (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111695245B (en) | Parallel simulation method for material irradiation damage space resolution random cluster dynamics | |
| CN107688999B (en) | Block chain-based parallel transaction execution method | |
| US8160852B2 (en) | Selection-based criteria for computation of multiple body interactions | |
| US20040024578A1 (en) | Discrete event simulation system and method | |
| CN110543663A (en) | Coarse-grained MPI + OpenMP hybrid parallel-oriented structural grid area division method | |
| Shim et al. | Rigorous synchronous relaxation algorithm for parallel kinetic Monte Carlo simulations of thin film growth | |
| CN112149808A (en) | Method, system and medium for expanding stand-alone graph neural network training to distributed training | |
| CN113254241B (en) | Data cross-process transmission method applied to CFD high-performance computation | |
| Hasircioglu et al. | Bivariate polynomial coding for exploiting stragglers in heterogeneous coded computing systems | |
| Bandic et al. | Mapping quantum circuits to modular architectures with QUBO | |
| CN115878729A (en) | Node block storage allocation optimization method and system based on alliance chain | |
| CN111695243B (en) | Communication method for synchronous parallel space resolution random cluster dynamics | |
| CN111695244B (en) | Material irradiation defect storage method suitable for multi-element SRSCD simulation | |
| Mathis et al. | A general performance model of structured and unstructured mesh particle transport computations | |
| Chen et al. | A massively parallel spatially resolved stochastic cluster dynamics method for simulations of irradiated materials | |
| Chen | The merge phase of parallel divide-and-conquer scheme for 3D Delaunay triangulation | |
| CN112528456B (en) | Heterogeneous node computing system and method | |
| JPH07140289A (en) | Neutron transport simulator | |
| CN110704693A (en) | Distributed graph calculation system and distributed graph calculation method | |
| Kazemian et al. | High-Performance, Accurate Large-Scale Quantum Chemistry Calculations on GPU Supercomputers using Coulomb-Perturbed Fragmentation | |
| Macchiarulo | A massively parallel implementation of Gillespie algorithm on FPGAs | |
| Kalgin | Parallel implementation of asynchronous cellular automata on a 32-core computer | |
| Kunz | Parallel discrete event simulation | |
| Marshall et al. | Performance improvement of a two-dimensional flood simulation application in hybrid computing environments | |
| Pinsky | Canonical approximation in the performance analysis of distributed systems |
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 |