Cross-platform transplantation method and system for non-static hydrodynamic model

Abstract

The invention provides a non-static-pressure hydrodynamic model cross-platform transplantation method and system, and relates to the technical field of program transplantation, and the method comprises the steps: obtaining a dependence library needed by a non-static-pressure hydrodynamic model, and installing and configuring the dependence library to a target platform processor; modifying a compiling link file of the non-static hydrodynamic model to enable the compiling link file to pass through a compiling environment of a target platform processor; adding a target platform compiling option and a corresponding compiling rule in the compiling link file; modifying a link position of the dependency library, and executing a compiling command after related configuration is completed; after the compiling is successful, according to a hotspot analysis tool, determining a hotspot function during the operation of the non-static hydrodynamic model, and rewriting the hotspot function into a many-kernel function; and compiling an incremental compiling script, and linking the many-core program to the non-static-pressure hydrodynamic model. According to the method, cross-platform transplantation of the non-static-pressure hydrodynamic model is achieved, the program running speed is increased through many-core optimization, and rapid simulation and prediction of tide, storm surge and tsunami are facilitated.

Description

Technical Field

[0001] The present invention belongs to the technical field of program transplantation, and in particular relates to a cross-platform transplantation method and system for a non-hydrostatic hydrodynamic model. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Non-hydrostatic hydrodynamic models are mainly used to simulate wave and water flow phenomena in various hydrodynamic environments, including but not limited to coastal areas, estuaries, lakes and other inland waters. In addition to focusing on wave patterns, non-hydrostatic hydrodynamic models can also handle tidal currents, storm surges, tsunamis and complex fluid dynamics problems.

[0004] The Sunway supercomputer based on the Sunway heterogeneous multi-core processor has powerful computing power and efficient energy utilization. However, the current non-hydrostatic hydrodynamic model cannot be run on the Sunway supercomputer based on the Sunway heterogeneous multi-core processor. Therefore, a large amount of computing resources cannot be used, resulting in a long time to generate program results and slow software running speed during the application of the non-hydrostatic hydrodynamic model, and it is impossible to quickly simulate and predict tidal currents, storm surges and tsunamis.

[0005] However, at present, in the process of transplanting the non-hydrostatic hydrodynamic model, a series of technical challenges still need to be overcome to enable the non-hydrostatic hydrodynamic model to run on the Shenwei supercomputer based on the Shenwei heterogeneous multi-core processor. These challenges include: Software compatibility: Since the Sunway supercomputer based on the Sunway heterogeneous multi-core processor adopts a specific hardware architecture and operating system, it is necessary to ensure that the non-hydrostatic hydrodynamic model can run normally in these environments; this involves modifying the configuration files in the non-hydrostatic hydrodynamic model to adapt to the compiler, library and system of the Sunway supercomputer.

[0006] Many-core optimization: To run the non-hydrostatic hydrodynamic model on the Sunway supercomputer, it is necessary to ensure that the software has good adaptability, which includes adapting to the Sunway heterogeneous many-core architecture so that simulation tasks can be efficiently performed on multiple computing nodes; this involves improving the original code of the non-hydrostatic hydrodynamic model so that it can utilize the computing power of many cores.

[0007] Therefore, the current non-hydrostatic hydrodynamic model program has problems with adaptability to Shenwei's heterogeneous many-core processors and parallel strategy selection on the Shenwei supercomputer. It cannot be run on the Shenwei supercomputer in its own way. At the same time, if it is not compatible with the many-core architecture, it will affect the high-performance computing power of the Shenwei supercomputer provided by the field using the non-hydrostatic hydrodynamic model software, and it will not be able to quickly simulate and predict tides, storm surges and tsunamis. Summary of the invention

[0008] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a cross-platform transplantation method and system for a non-hydrostatic hydrodynamic model, which compiles and modifies the characteristics of the non-hydrostatic hydrodynamic model so that it can be transplanted to the Sunway supercomputer. At the same time, the program is multi-core optimized so that it can give full play to the performance of the Sunway supercomputer, speed up the program running speed, and can quickly simulate and predict tides, storm surges and tsunamis, fully promoting its application in numerical simulation fields such as waves, while providing strong technical support for simulation and prediction of tides, storm surges, tsunamis and other fields.

[0009] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: A first aspect of the present invention provides a cross-platform transplantation method for a non-hydrostatic hydrodynamic model.

[0010] A cross-platform transplantation method for a non-static pressure hydrodynamic model, comprising: Obtain the dependency libraries required for the non-hydrostatic hydrodynamic model, and install and configure them on the target platform processor; Modify the compilation link file of the non-static pressure hydrodynamic model: comment out the default original processor information, modify it to the control options of the compiler applicable to the target platform processor, specify it as the target platform language compilation command, and make it pass the compilation environment of the target platform processor; In the compilation link file, add the target platform compilation options and corresponding compilation rules; Modify the link location of the dependent library and execute the compilation command after completing the relevant configuration; After successful compilation, the hotspot function of the non-hydrostatic hydrodynamic model during operation is determined using a hotspot analysis tool, and the hotspot function is rewritten into a multi-core function. Based on the multi-core function, a multi-core program is developed; An incremental compilation script is written to link the many-core program to the non-hydrostatic hydrodynamic model for performing many-core optimization on the non-hydrostatic hydrodynamic model.

[0011] A second aspect of the present invention provides a non-hydrostatic pressure hydrodynamic model cross-platform transplantation system.

[0012] A non-static hydrodynamic model cross-platform transplantation system, including: Modify the compilation module and configure it to obtain the dependent library files required by the non-static pressure hydrodynamic model, and install and configure them to the target platform processor; Modify the compilation link file of the non-static pressure hydrodynamic model: comment out the default original processor information, modify it to the control options of the compiler applicable to the target platform processor, specify it as the target platform language compilation command, and make it pass the compilation environment of the target platform processor; In the compilation and linking rule file, add the target platform compilation options and the corresponding compilation rules; Modify the link location of the dependent library and execute the compilation command after completing the relevant configuration; The many-core optimization module is configured to, after successful compilation, determine the hotspot function of the non-hydrostatic hydrodynamic model during operation according to the hotspot analysis tool, rewrite the hotspot function into a many-core function, and develop a many-core program based on the many-core function; An incremental compilation script is written to link the many-core program to the non-hydrostatic hydrodynamic model for performing many-core optimization on the non-hydrostatic hydrodynamic model.

[0013] The third aspect of the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a method as described in the second aspect of the present invention are implemented.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in a method as described in the second aspect of the present invention.

[0015] A fifth aspect of the present invention provides a computer program product comprising instructions, which, when running on a computer, enables the computer program to implement the steps in a method as described in the second aspect of the present invention when executed by a processor.

[0016] One or more of the above technical solutions have the following beneficial effects: The present invention modifies the compile link file and the control options of the compiler applicable to the target platform processor, so that the non-hydrostatic pressure hydrodynamic model can better pass through the compilation environment of the target platform processor, and can successfully compile and run the non-hydrostatic pressure hydrodynamic model on the Shenwei heterogeneous multi-core processor; and uses a large amount of processor computing resources to reduce the program result generation time, speed up the software running speed, and realize rapid simulation and prediction of tidal currents, storm surges and tsunamis. Promote its application in the field of numerical simulation such as ocean waves, and provide strong technical support for simulation and prediction of tidal currents, storm surges, tsunamis and other fields.

[0017] The present invention transplants the non-hydrostatic hydrodynamic model to the Sunway supercomputer and performs multi-core optimization on the program. After successful compilation, the hotspot function is rewritten as a multi-core function, and a multi-core program is developed to link with the non-hydrostatic hydrodynamics, so that it can give full play to the performance of the Sunway supercomputer and ensure that the software has good adaptability, including adapting to the Sunway heterogeneous multi-core architecture, so as to efficiently execute simulation tasks on multiple computing nodes, so that it can utilize the computing power of the many cores.

[0018] At the same time, the present invention provides a convenient installation solution for installing and using or optimizing programs on the Shenwei processor, thereby greatly reducing the installation time, and performs multi-core optimization, providing a solid foundation for accelerating the operation of the program and improving the work efficiency of software users.

[0019] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0021] Figure 1 This is a flow chart of a cross-platform transplantation method of a non-hydrostatic pressure hydrodynamic model according to the first embodiment of the present invention; Figure 2 This is a schematic diagram of modules running on multiple cores in a cross-platform transplantation method for a non-hydrostatic hydrodynamic model according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0022] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0023] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0024] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0025] The overall idea proposed in the present invention is: first, by building the configuration environment required for transplantation, adapting and modifying the characteristics of the non-hydrostatic pressure hydrodynamic model software and its required dependent libraries, so that it can run normally in the compilation environment of the target platform processor, and solving the software compatibility problem; then modifying the source code of the non-hydrostatic pressure hydrodynamic model software, writing an incremental compilation script, developing a multi-core program, linking the multi-core program to the non-hydrostatic pressure hydrodynamic model software, realizing multi-core optimization, giving full play to the computing performance of the target platform processor, and accelerating the program running speed.

[0026] The non-hydrostatic pressure model compilation method of the present invention is different from the existing compilation method in that: 1. The non-hydrostatic hydrodynamic model is a general software and needs to be modified to adapt and port through the target platform compiler.

[0027] 2. The specific compilation options need to be modified in the program compilation rule file or method to adapt the compilation rules of the target platform processor architecture.

[0028] 3. An optimization method for many-core program development based on many-core functions is proposed, so that it can fully utilize the performance of the target platform processor and speed up the program running speed.

[0029] Terminology explanation: HYPRE: HYPRE (High Performance Preconditioners) is a high-performance preconditioner library that is mainly used to solve problems of large-scale sparse linear systems. It provides a series of efficient parallel algorithms and preconditioning techniques that are particularly suitable for parallel computer systems.

[0030] Many-core: A processor architecture with a large number of cores. That is, a large number of computing cores are integrated on a chip, and these cores can execute multiple tasks or instructions at the same time to achieve high-performance parallel computing. This architecture is particularly suitable for application scenarios that require a large amount of parallel processing, such as high-performance computing, graphics processing, and artificial intelligence.

[0031] Embodiment 1 This embodiment discloses a cross-platform transplantation method for a non-hydrostatic hydrodynamic model.

[0032] In one embodiment of the present disclosure, a non-hydrostatic pressure model cross-platform transplantation method is provided. Figure 1 As shown, the steps include: Step 1: Obtain the dependency libraries required for the non-hydrostatic hydrodynamic model, and install and configure them on the target platform processor; Step 2: Modify the compilation link file of the non-static pressure hydrodynamic model: comment out the default original processor information, and modify it to the control options of the compiler applicable to the target platform processor, specify it as the target platform language compilation command, so that it passes the compilation environment of the target platform processor; Step 3: Add the target platform compilation options and corresponding compilation rules in the compilation link file; Step 4: Modify the link location of the dependent library and execute the compile command after completing the relevant configuration; Step 5: After successful compilation, the hotspot function of the non-hydrostatic hydrodynamic model during operation is determined using a hotspot analysis tool, and the hotspot function is rewritten into a multi-core function. Based on the multi-core function, a multi-core program is developed; Step 6: Write an incremental compilation script to link the many-core program to the non-hydrostatic hydrodynamic model for many-core optimization of the non-hydrostatic hydrodynamic model.

[0033] In order to more clearly illustrate this embodiment, a method for transplanting a non-hydrostatic hydrodynamic model across platforms can be specifically described as follows: Step 1: Obtain the dependent library files required for the non-hydrostatic pressure model, and install and configure them on the target platform processor; In this embodiment, the non-hydrostatic hydrodynamic model is the NHWAVE program; the target platform processor is the Shenwei heterogeneous multi-core processor.

[0034] Specifically, obtain the dependent library file HYPRE required by the NHWAVE program, and install and configure it to the Shenwei heterogeneous multi-core processor.

[0035] The specific steps for installing and configuring the target platform processor are as follows: Set the parameter command configure (configuration), and pay special attention to specifying the cross-compilation platform as alpha.

[0036] The specific operation is to execute the following command in the folder containing the configure file to compile: Step 1: Prepare the project's build environment, specify the cross-compilation platform as alpha, specify the installation path, and specify the Shenwei compiler command; Step 2: Execute the compilation command: In the program, execute the make command; Step 3: Execute the installation command: execute make install; Through the above command, the library is installed in the specified path / home / xxx / hyper-install.

[0037] HYPRE dependency library is mainly used to solve the problem of large-scale sparse linear systems. It can provide a series of efficient parallel algorithms and preprocessing techniques, which can be better applied to parallel computer systems. And the dependency library provides predefined functions, which can quickly implement specific functions; by using the dependency library, you can avoid repeatedly writing the same code in multiple projects, improve the reusability of the code, simplify the development process, promote the standardization and normalization of the code, and help improve the quality and maintainability of the code.

[0038] Here, the dependent library files required for the non-hydrostatic hydrodynamic model are obtained, installed, and configured to the target platform processor, which can achieve faster construction of the configuration environment required for transplantation and better adaptation to the target platform processor.

[0039] Step 2: Modify the compilation link file of the non-hydrostatic model: comment out the default original processor information, and modify it to the control options of the compiler suitable for the target platform processor, specify it as the target platform language compilation command, so that it passes the compilation environment of the target platform processor.

[0040] The specific steps include: Step 2-1, modify the compilation link file of the non-static hydrodynamic model.

[0041] Specifically, modify the Makefile of the NHWAVE program so that it can pass the compilation environment of the Shenwei compiler: Unzip the NHWAVE program source code package and enter the src folder; Copy the compilation link file Makefile based on the settings of the compilation command mpif90 as the compilation link file Makefile of the NHWAVE program.

[0042] The specific command line can be written as: cp Makefile.supermic.mpif90 Makefile.

[0043] Step 2-2, comment out the default original processor information.

[0044] Specifically, open the compilation link file Makefile, modify the parameter DEF_FLAGS, and remove the parameter instruction "-C"; removing "-C" is to delete comments and other information during preprocessing to ensure accuracy during preprocessing; Comment parameter FLAG_9=-DINTEL; the line FLAG_9=-DINTEL indicates that Intel's processor is to be used. If selected, some Intel function libraries will be used by default.

[0045] Step 2-3, modify the compiler control options to suit the target platform processor.

[0046] Specifically, modify the compiler FC, determine the location of the compiler through the compilation command which mpif90, and replace the compiler location into the compilation link file Makefile.

[0047] By modifying the compilation link file and annotating the default original processor information, the accuracy of preprocessing can be guaranteed and the compilation can be made more accurate.

[0048] Step 3: In the compilation link file, add the target platform compilation options and corresponding compilation rules.

[0049] Specifically, add the static compilation option -static; since the Shenwei system supports static libraries better, static compilation is adopted; Create or change the static library generation index command ranlib, and use the target platform system Shenwei system corresponding command swranlib.

[0050] By adding compilation options and compilation rules that are more suitable for the target platform, the compilation efficiency can be improved.

[0051] Step 4: Modify the link location of the dependent library and execute the compilation command after completing the relevant configuration.

[0052] The specific steps include: Step 4-1, modify the location of the dependency library link and complete the relevant configuration.

[0053] The specific command is to specify the library file path of the dependent library HYPRE and the HYPRE header file path, and modify them to the installation path of the dependent library HYPRE installed in step 1.

[0054] The specific command line is: LIBS= -L / home / xxx / hyper-install / lib -lHYPER; INCS= -I / home / xxx / hyper-install / include.

[0055] Step 4-2, execute the compile command.

[0056] Specifically, after completing the relevant basic configuration required for the transplant, the software compatibility problem can be solved, and the compile and run can be performed by executing the make compilation command; After the compilation is completed, an executable file named nhwave will be generated in the current directory, indicating that the compilation is successful.

[0057] The non-hydrostatic hydrodynamic model was successfully compiled and run on the Shenwei heterogeneous multi-core processor, which facilitated the use of a large amount of processor computing resources, reduced the time for program result generation, accelerated the software running speed, and realized rapid simulation and prediction of tidal currents, storm surges, tsunamis, etc. It promoted its application in the field of numerical simulation such as ocean waves, and provided strong technical support for simulation and prediction of tidal currents, storm surges, tsunamis, etc.

[0058] However, in order to fully utilize the computing performance of Shenwei's heterogeneous multi-core processing and speed up the overall running speed of the program, it is still necessary to complete the NHWAVE multi-core optimization based on Shenwei's heterogeneous multi-core processor.

[0059] Step 5: After successful compilation, the hotspot function of the non-hydrostatic hydrodynamic model during operation is determined using a hotspot analysis tool, and the hotspot function is rewritten into a many-core function. Based on the many-core function, a many-core program is developed.

[0060] Specifically, the hotspot functions are determined by using the Sunway system hotspot analysis tool SWPROF. That is, the Sunway system hotspot analysis tool is used to dynamically analyze the program based on test cases, and the functions that take up a relatively high proportion of the time in the entire program running process are selected as hotspot functions based on the time consumption of the program functions.

[0061] Based on C language, the many-core interface function library Athread is used to rewrite hot functions into many-core functions, mainly including four functions: delyfun1_3d, delxfun1_3d, delyfun_3d, and delxfun_3d. Based on the many-core functions, code development is carried out in the many-core code files, such as the slave.c file.

[0062] Step 6: Write an incremental compilation script to link the many-core program to the non-hydrostatic hydrodynamic model for many-core optimization of the non-hydrostatic hydrodynamic model.

[0063] Specifically, write the incremental compilation script as follows: Step 6-1, rules for compiling many-core files: compile the code C file adapted to the Shenwei many-core processor into a many-core target file.

[0064] Step 6-2, linking to the rules of the overall program executable file: the many-core target file generated in step 6-1 is combined with other target files of the generated program to generate the final executable file.

[0065] When compiling the multi-core file in step 6-1, swgcc is used to compile the C language master core code and the C language slave core code, the -mhost parameter is used to compile the operation control core code, and the -mslvae parameter is used to compile the operation core code; In step 6-2, the parallel compilation command mpif90 is used to link the previously successfully compiled .o files. Here, the hypre dynamic library and the -mhybrid parameter are required. The -mhybrid parameter is used to link the operation control core code and the operation core code into an executable file.

[0066] Finally, the delyfun1_3d, delxfun1_3d, delyfun_3d, and delxfun_3d functions are operated on many cores, and other modules are operated on the main core. Figure 2 The process shown takes the delyfun1_3d function as an example. After modifying the function to be multi-core, the function is placed in the multi-core for execution.

[0067] By performing many-core optimization on the program, the hotspot functions are rewritten as many-core functions after successful compilation, and the many-core program is developed to link with the non-hydrostatic hydrodynamics to ensure that the model has good adaptability on the target platform processor so that simulation tasks can be efficiently executed on multiple computing nodes.

[0068] The model can fully utilize the performance of the Sunway supercomputer, use a large amount of computing resources, reduce the time to generate program results, speed up the software operation, and achieve rapid simulation and prediction of tidal currents, storm surges, tsunamis, etc. It can promote its application in the field of numerical simulation of waves, and provide strong technical support for simulation and prediction of tidal currents, storm surges, tsunamis, etc.

[0069] Embodiment 2 The purpose of this embodiment is to provide a non-static pressure hydrodynamic model cross-platform transplantation system, including: Modify the compilation module and configure it to obtain the dependency library required by the non-static hydrodynamic model, and install and configure it to the target platform processor; Modify the compilation link file of the non-static pressure hydrodynamic model: comment out the default original processor information, modify it to the control options of the compiler applicable to the target platform processor, specify it as the target platform language compilation command, and make it pass the compilation environment of the target platform processor; In the compilation link file, add the target platform compilation options and corresponding compilation rules; Modify the link location of the dependent library and execute the compilation command after completing the relevant configuration; The many-core optimization module is configured to, after successful compilation, determine the hotspot function of the non-hydrostatic hydrodynamic model during operation according to the hotspot analysis tool, rewrite the hotspot function into a many-core function, and develop a many-core program based on the many-core function; An incremental compilation script is written to link the many-core program to the non-hydrostatic hydrodynamic model for performing many-core optimization on the non-hydrostatic hydrodynamic model.

[0070] Embodiment 3 The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0071] Embodiment 4 The purpose of this embodiment is to provide a computer-readable storage medium.

[0072] A computer-readable storage medium stores a computer program, which executes the steps of the above method when executed by a processor.

[0073] Embodiment 5 The purpose of this embodiment is to provide a computer program product containing instructions, which, when running on a computer, enables the computer to execute the methods and functions involved in any of the above embodiments. The steps involved in the apparatus of the above embodiment correspond to the method embodiment 1, and the specific implementation method can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0074] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0075] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A cross-platform transplantation method for a non-hydrostatic pressure model, characterized in that: include: Obtain the dependency libraries required for the non-hydrostatic hydrodynamic model, and install and configure them on the target platform processor; Modify the compilation link file of the non-static pressure hydrodynamic model: comment out the default original processor information, modify it to the control options of the compiler applicable to the target platform processor, specify it as the target platform language compilation command, and make it pass the compilation environment of the target platform processor; In the compilation link file, add the target platform compilation options and corresponding compilation rules; Modify the link location of the dependent library and execute the compilation command after completing the relevant configuration; After successful compilation, the hotspot function of the non-hydrostatic hydrodynamic model during operation is determined using a hotspot analysis tool, and the hotspot function is rewritten into a multi-core function. Based on the multi-core function, a multi-core program is developed; An incremental compilation script is written to link the many-core program to the non-hydrostatic hydrodynamic model for performing many-core optimization on the non-hydrostatic hydrodynamic model.

2. A cross-platform transplantation method for a non-hydrostatic pressure model according to claim 1, characterized in that: The modification is to control options of the compiler applicable to the target platform processor, specifically, to determine the compiler position through the compilation command, and to replace the compiler position into the compilation link file.

3. A cross-platform transplantation method for a non-hydrostatic pressure model according to claim 1, characterized in that: The adding of target platform compilation options and corresponding compilation rules specifically includes: adding static compilation options, creating or changing static library generation index commands, and using target platform system corresponding commands.

4. A cross-platform transplantation method for a non-hydrostatic pressure model according to claim 1, characterized in that: Modify the location of the dependency library link and complete the relevant configuration, specifically: specify the dependency library file path and header file path, and modify them to the installation path when installing the dependency library.

5. The cross-platform transplantation method of a non-hydrostatic pressure model according to claim 1, characterized in that: Hotspot analysis tools are used to dynamically analyze the program based on test cases. Functions that take up a high proportion of the entire program execution time are selected as hotspot functions based on the time consumption of program functions.

6. A cross-platform transplantation method for a non-hydrostatic pressure model according to claim 1, characterized in that: The incremental compilation script is specifically written as follows: compiling the code file adapted to the target platform processor into a multi-core target file, and linking the multi-core target file to the overall program to generate a final executable file.

7. A non-hydrostatic pressure model cross-platform transplantation system, characterized in that: include: Modify the compilation module and configure it to obtain the dependency library required by the non-static hydrodynamic model, and install and configure it to the target platform processor; Modify the compilation link file of the non-static pressure hydrodynamic model: comment out the default original processor information, modify it to the control options of the compiler applicable to the target platform processor, specify it as the target platform language compilation command, and make it pass the compilation environment of the target platform processor; In the compilation link file, add the target platform compilation options and corresponding compilation rules; Modify the link location of the dependent library and execute the compilation command after completing the relevant configuration; The many-core optimization module is configured to, after successful compilation, determine the hotspot function of the non-hydrostatic hydrodynamic model during operation according to the hotspot analysis tool, rewrite the hotspot function into a many-core function, and develop a many-core program based on the many-core function; An incremental compilation script is written to link the many-core program to the non-hydrostatic hydrodynamic model for performing many-core optimization on the non-hydrostatic hydrodynamic model.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are performed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images