Statistical analysis method and system for territorial survey results

Abstract

The invention discloses a statistical analysis method and system for territorial survey achievements, and the method comprises the steps: constructing a task analysis scheme corresponding to a target analysis task, splitting the target analysis task into sub-analysis tasks corresponding to each task scene, and splitting each sub-analysis task into a plurality of analysis tasks; allocating computing resources and territorial survey result data based on the task analysis scheme; controlling the plurality of sub-analysis tasks to carry out parallel calculation, and controlling the plurality of analysis tasks to carry out distributed calculation; dividing the territorial survey result data into analysis territorial survey result data corresponding to each analysis task, and calling the spatial analysis model to the computing node to obtain an analysis result corresponding to each analysis task; and collecting a plurality of analysis results corresponding to each sub-analysis task in the plurality of sub-analysis tasks to obtain a target analysis result of the target analysis task so as to improve the efficiency and accuracy of statistical analysis.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to a statistical analysis method and system for land survey results. Background Art

[0002] Obtaining land survey results and promptly understanding annual land use changes is a crucial step in protecting land resources. Existing techniques often use pre-saved standard summary statistical task templates to analyze the corresponding results data during statistical analysis of land survey results. However, with the gradual accumulation of land survey results and the ever-changing statistical needs of land use application analysis and analysis scenarios for natural resource management, existing statistical analysis methods are unable to adapt to the flexible, complex, and ever-changing customized summary analysis requirements.

[0003] On the other hand, the results of land surveys accumulate year by year, and the resulting summary statistics involve data from multiple survey phases and tens of millions of elements across hundreds of county-level administrative regions, as well as data from multiple years of repeated surveys and altered surveys. The data collection process involves a multitude of data sources, large volumes of data, and complex statistical analysis logic, leading to difficulties in verifying the accuracy of results, poor processing timeliness, and a lengthy output process. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention discloses a statistical analysis method and system for land survey results, which are used to improve the efficiency and accuracy of statistical analysis.

[0005] In order to achieve the above-mentioned object, the present invention discloses a statistical analysis method for land survey results, comprising: Configuring task analysis parameters of the target analysis task to construct a task analysis plan corresponding to the target analysis task; Splitting the target analysis task into sub-analysis tasks corresponding to each task scenario, and splitting each sub-analysis task into several analysis tasks according to the task analysis plan; Allocate computing resources and land survey results data required for each of the sub-analysis tasks based on the task analysis plan; Controlling a plurality of the sub-analysis tasks to perform parallel computing based on the computing resources and the land survey results data, and allocating the plurality of the analysis tasks to various computing nodes for distributed computing based on the computing resources; Dividing the land survey result data into analysis land survey result data corresponding to each analysis task, and calling the spatial analysis model corresponding to the analysis task to obtain analysis results corresponding to each analysis task based on the analysis land survey result data, the spatial analysis model and the computing node; The target analysis result of the target analysis task is obtained by gathering the analysis results corresponding to each of the sub-analysis tasks.

[0006] The present invention discloses a statistical analysis method for land survey results. The statistical analysis method first configures the task analysis parameters of the target analysis task to construct a task analysis scheme corresponding to the target analysis task, etc., so as to simplify and accurately construct the analysis process of the target analysis task according to the analysis scheme to ensure the accuracy of the statistical analysis. Secondly, the complex target analysis task is split into sub-analysis tasks corresponding to each task scenario, and then the sub-analysis tasks are split into different types of analysis tasks to reduce the amount of data processed by each analysis task by task splitting, thereby improving the efficiency of statistical analysis. Then, corresponding computing resources and land survey result data are allocated to each analysis task, and parallel computing is performed between sub-analysis tasks, while distributed computing is performed between analysis tasks to improve the efficiency of task execution, thereby improving the efficiency of statistical analysis. Finally, a number of analysis results corresponding to each sub-analysis task in each sub-analysis task are summarized to quickly and accurately obtain the target analysis results of the target analysis task.

[0007] As a preferred example, configuring the task analysis parameters of the target analysis task to construct a task analysis solution corresponding to the target analysis task includes: Retrieving a corresponding statistical process template from a pre-built summary statistical process library based on the analysis requirements of the target analysis task; Determine the target statistical data type of the target analysis task based on the analysis requirement, and retrieve the corresponding data analysis logic from the pre-built basic spatial operator library based on the target statistical data type and the statistical process template; The task analysis rules of the target analysis task are called from a pre-built summary statistical rule library according to the data analysis logic to construct the task analysis plan based on the statistical process template, the target statistical data type, the data analysis logic and the task analysis rules.

[0008] In the above scheme, in order to solve the impact of the complexity of statistical logic analysis on the efficiency of statistical analysis, a database for storing data analysis logic, task analysis rules, etc. is pre-set, and the analysis logic of the target analysis task is simplified by calling the various statistical logics pre-saved in the database, so as to improve the efficiency of logical analysis by reducing the difficulty of logical analysis.

[0009] As a preferred example, configuring the task analysis parameters of the target analysis task to construct a task analysis solution corresponding to the target analysis task includes: Retrieving a completed analysis task with the highest similarity to the task analysis plan from a preset completed analysis task library; Determine whether the completed analysis task and the target analysis task have reused analysis results; When it is determined that the reuse analysis result does not exist, the spatial range of the data is configured to the task analysis plan to obtain the task analysis plan configured with the spatial range.

[0010] In the above scheme, based on the overlap of statistical analysis requirements, when the task analysis plan corresponding to the current target analysis task is constructed, the database that saves the historical completed analysis plans can be queried to identify whether there is a historical analysis task that is consistent with the current target analysis task. When the query finds that there is one, the analysis result of the historical analysis task is called as the reused analysis result, and there is no need to perform statistical analysis of new result data, thereby improving the efficiency of statistical analysis.

[0011] As a preferred example, the target analysis task is split into sub-analysis tasks corresponding to each task scenario, and each sub-analysis task is split into several analysis tasks according to the task analysis plan, including: Determine several business scenarios involved in the target analysis task according to the spatial scope, and generate sub-analysis tasks corresponding to each of the business scenarios based on the task analysis plan; determining the types of different analysis tasks in the sub-analysis tasks according to the task analysis scheme, so as to split each of the sub-analysis tasks into a plurality of analysis tasks according to the types; A task priority corresponding to each of the analysis tasks is formulated based on the task analysis plan.

[0012] In the above scheme, when performing statistics on the results data, based on the synchronous data calls involving multiple business scenarios, in order to avoid the impact of calls between data in different areas and to more accurately and specifically obtain the results data uploaded by each business scenario, a sub-analysis task corresponding to each business scenario can be generated, thereby improving the accuracy of the statistical analysis. Furthermore, the sub-analysis tasks are split into different types of analysis tasks, and then the priorities of different analysis tasks are determined. Through the priorities and the splitting, the analysis logic of the sub-analysis tasks can be accurately executed, thereby improving the accuracy of the statistical analysis.

[0013] As a preferred example, the allocation of computing resources and land survey results data required for each sub-analysis task based on the task analysis plan includes: Based on the task analysis plan and the preset data retrieval mechanism, the land survey result data required for each sub-analysis task is retrieved from the preset resource library to the preset large land survey result data pool; The remaining computing resources stored in the resource library are obtained and the occupied computing resources corresponding to each of the analysis tasks are evaluated to determine the computing resources allocated to each of the sub-analysis tasks.

[0014] In the above scheme, to ensure accurate analysis results for each sub-analysis task, the corresponding land survey results data and computing resources are first called up to execute the corresponding analysis process. When calling the land survey results data, the land survey results data required for each sub-analysis task is added to a preset large land survey results data pool. This allows different sub-analysis tasks to access the same land survey results data, thereby reducing the amount of land survey results data used and improving the accuracy and efficiency of statistical analysis.

[0015] As a preferred example, the allocating several analysis tasks to various computing nodes for distributed computing based on the computing resources includes: Collecting the operating status of each computing node in real time to determine the multiple task nodes corresponding to each of the sub-analysis tasks; Determining a calculation order corresponding to a plurality of the analysis tasks according to a task priority corresponding to each of the analysis tasks; According to the occupied computing resources, the computing resources, and the computing order, several analysis tasks are sequentially allocated to the corresponding computing nodes for distributed computing.

[0016] In this approach, when performing the calculation of a single sub-analysis task, multiple analysis tasks are synchronized by calling different computing nodes, improving the efficiency of statistical analysis. Furthermore, multiple analysis tasks are distributed across different computing nodes using task priorities, allowing each analysis task to be accurately completed based on the priority level, thereby improving the accuracy of statistical analysis.

[0017] As a preferred example, the allocating the plurality of analysis tasks to the corresponding computing nodes for distributed computing according to the occupied computing resources, the computing resources, and the computing order includes: Determining a plurality of the analysis tasks for first-round allocation according to the occupied computing resources, the computing resources, and the computing ranking; polling the operating status of each computing node in real time, and when determining that the computing node currently being polled has completed the calculation of the analysis task, releasing the computing resources occupied by the computing node currently being polled and corresponding to the analysis task being processed, and determining that the computing node is an idle node; Determining the analysis task to be assigned according to the calculation ranking, and assigning the analysis task to be assigned to the idle node when it is determined that the occupied computing resources released by the idle node are greater than or equal to the occupied computing resources required by the analysis task to be assigned; When the analysis task to be assigned is assigned to the idle node, the occupied computing resources required by the analysis task to be assigned are assigned to the idle node, so that the idle node executes the analysis task to be assigned.

[0018] In the above scheme, by monitoring the completion status of each computing node in real time, the computing resources occupied by the current computing node are released in time, and the calculation of the next sorted analysis task is performed based on the released computing resources and the computing nodes in the idle state. Then, by monitoring the computing nodes and releasing the computing resources, the occupation of computing resources can be reduced, and the distributed computing of multiple analysis tasks can be realized with fewer computing resources. Then, when allocating computing resources, multiple sub-analysis tasks can be provided for parallel computing, thereby improving the efficiency of statistical analysis.

[0019] As a preferred example, the land survey result data is divided into analysis land survey result data corresponding to each analysis task, and the spatial analysis model corresponding to the analysis task is called to obtain the analysis result corresponding to each analysis task according to the analysis land survey result data, the spatial analysis model and the computing node, including: Retrieving the analyzed land survey result data corresponding to the analysis task from the large land survey result data pool; Obtaining a spatial analysis model corresponding to the land survey result data; The analysis land survey result data and the spatial analysis model are distributed to the computing node where the analysis task is located, so as to output the analysis result corresponding to the analysis task through the computing node.

[0020] In the above scheme, the land survey results data based on the call analysis is the land survey results data, and the land survey results data includes geographic data with spatial characteristics. Therefore, in order to improve the accuracy of statistical analysis, calling the corresponding spatial analysis model to perform corresponding data analysis can improve the accuracy of statistical analysis.

[0021] As a preferred example, the aggregating of the analysis results corresponding to each of the plurality of sub-analysis tasks to obtain the target analysis result of the target analysis task includes: Retrieving the verification rule corresponding to the task analysis rule from the pre-built summary statistical rule library; When each of the analysis results meets the verification rules, the achievement statistical template corresponding to the target analysis task is retrieved from the pre-built achievement template library; Fill the result statistics template with a number of the analysis results to obtain the target analysis results of the target analysis task.

[0022] In this approach, the verification rules corresponding to the current target analysis task are first applied to verify the analysis results of each task, ensuring accuracy. Secondly, when all analysis results pass verification, the statistical template corresponding to the target analysis task is retrieved from the pre-built result template library to display the analysis results in a simple and clear manner, adapting to different statistical analysis needs.

[0023] On the other hand, the present invention also discloses a statistical analysis system for land survey results, including an analysis and construction module, a task splitting module, a resource scheduling module, a hybrid computing module, a task analysis module and a result display module; The analysis building module is used to configure the task analysis parameters of the target analysis task to build a task analysis plan corresponding to the target analysis task; The task splitting module is used to split the target analysis task into sub-analysis tasks corresponding to each task scenario, and split each sub-analysis task into several analysis tasks according to the task analysis plan; The resource scheduling module is used to allocate the computing resources and land survey results data required for each sub-analysis task based on the task analysis plan; The hybrid computing module is used to control a plurality of the sub-analysis tasks to perform parallel computing based on the computing resources and the land survey results data, and to distribute the plurality of the analysis tasks to each computing node for distributed computing based on the computing resources; The task analysis module is used to divide the land survey result data into analysis land survey result data corresponding to each analysis task, and call the spatial analysis model corresponding to the analysis task to obtain the analysis result corresponding to each analysis task based on the analysis land survey result data, the spatial analysis model and the computing node; The result display module is used to collect a number of analysis results corresponding to each of the sub-analysis tasks in a number of sub-analysis tasks to obtain the target analysis result of the target analysis task.

[0024] The present invention discloses a statistical analysis system for land survey results. The statistical analysis system first configures the task analysis parameters of the target analysis task to construct the task analysis scheme corresponding to the target analysis task, so as to simplify and accurately construct the analysis process of the target analysis task according to the analysis scheme to ensure the accuracy of the statistical analysis. Secondly, the complex target analysis task is split into sub-analysis tasks corresponding to each task scenario, and then the sub-analysis tasks are split into different types of analysis tasks to reduce the amount of data processed by each analysis task by task splitting, thereby improving the efficiency of statistical analysis. Then, corresponding computing resources and land survey result data are allocated to each analysis task, and parallel computing is performed between sub-analysis tasks, while distributed computing is performed between analysis tasks to improve the efficiency of task execution, thereby improving the efficiency of statistical analysis. Finally, a number of analysis results corresponding to each sub-analysis task in each sub-analysis task are summarized to quickly and accurately obtain the target analysis results of the target analysis task. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 : A flow chart of a statistical analysis method for land survey results provided by an embodiment of the present invention; Figure 2 : A schematic diagram of the structure of a statistical analysis system for land survey results provided by an embodiment of the present invention; Figure 3 : A schematic diagram of a statistical analysis framework for land survey results based on multi-mode heterogeneous computing technology provided by another embodiment of the present invention; Figure 4 : A schematic diagram of a computing framework provided in yet another embodiment of the present invention; Figure 5 : A schematic diagram of data circulation based on a computing framework provided in yet another embodiment of the present invention; Figure 6 : A schematic diagram of scheduling management from tasks to resources provided by another embodiment of the present invention; Figure 7 : A task scheduling management diagram provided by another embodiment of the present invention; Figure 8 : A schematic diagram of a task scheduling process provided by another embodiment of the present invention; Figure 9 : A schematic diagram of model management based on a summary statistical knowledge base provided in yet another embodiment of the present invention; Figure 10 : A schematic flow chart of a statistical analysis method for land survey results provided in another embodiment of the present invention; Figure 11: A schematic diagram of an analysis solution provided in another embodiment of the present invention; Figure 12 : A schematic diagram of constructing an analysis service provided by another embodiment of the present invention; Figure 13 : A schematic diagram of analyzing business splitting provided by another embodiment of the present invention; Figure 14 : A schematic diagram of analysis and calculation provided for another embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] Example 1 This embodiment provides a statistical analysis method for land survey results, which is used to improve the efficiency and accuracy of statistical analysis. Figure 1 , mainly including steps 101 to 106, wherein the steps are mainly: Step 101: configuring task analysis parameters of a target analysis task to construct a task analysis plan corresponding to the target analysis task; Step 102: Split the target analysis task into sub-analysis tasks corresponding to each task scenario, and split each sub-analysis task into several analysis tasks according to the task analysis plan; Step 103: allocating computing resources and land survey result data required for each sub-analysis task based on the task analysis plan; Step 104: controlling a plurality of the sub-analysis tasks to perform parallel computing based on the computing resources and the land survey result data, and allocating the plurality of the sub-analysis tasks to the respective computing nodes for distributed computing based on the computing resources; Step 105: Divide the land survey result data into analyzed land survey result data corresponding to each analysis task, and call the spatial analysis model corresponding to the analysis task to obtain analysis results corresponding to each analysis task based on the analyzed land survey result data, the spatial analysis model, and the computing node; Step 106: Gathering a number of analysis results corresponding to each of the sub-analysis tasks to obtain a target analysis result of the target analysis task.

[0028] In some implementations of this embodiment, the step 101 may preferably be performed by performing the following steps to build the task analysis solution, so as to improve the efficiency of the construction and the accuracy of the task analysis solution. Specifically, the steps are: Step 1011: Retrieving a corresponding statistical process template from a pre-built summary statistical process library based on the analysis requirements of the target analysis task; Step 1012: Determine the target statistical data type of the target analysis task based on the analysis requirement, and retrieve corresponding data analysis logic from a pre-built basic spatial operator library based on the target statistical data type and the statistical process template; Step 1013: Call the task analysis rules of the target analysis task from the pre-built summary statistical rule library according to the data analysis logic to build the task analysis plan based on the statistical process template, the target statistical data type, the data analysis logic and the task analysis rules.

[0029] Step 1014: Retrieving the completed analysis task with the highest similarity to the task analysis solution from a preset completed analysis task library; Step 1015: Determine whether the completed analysis task and the target analysis task have reused analysis results; Step 1016: When it is determined that the reuse analysis result does not exist, the spatial range of the data is configured to the task analysis plan to obtain the task analysis plan configured with the spatial range.

[0030] In this embodiment, in order to solve the problem of the influence of the complexity of statistical logic analysis on the efficiency of statistical analysis, the above steps pre-set a database for storing data analysis logic, task analysis rules, etc., and simplify the analysis logic of the target analysis task by calling the various statistical logics pre-stored in the database, so as to improve the efficiency of logical analysis by reducing the difficulty of logical analysis. Among them, based on the overlap of statistical analysis requirements, after constructing the task analysis plan corresponding to the current target analysis task, the database that stores historical completed analysis plans can be queried to identify whether there is a historical analysis task that is consistent with the current target analysis task. Then, when the query finds that there is one, the analysis result of the historical analysis task is called as the reused analysis result, so that there is no need to perform new statistical analysis of the result data, thereby improving the efficiency of statistical analysis.

[0031] In certain implementations of this embodiment, step 102 preferably splits the target analysis task into the following steps to improve the efficiency and accuracy of statistical analysis. Specifically, the steps are: Step 1021: determining several business scenarios involved in the target analysis task according to the spatial scope, and generating sub-analysis tasks corresponding to each business scenario based on the task analysis plan; Step 1022: determining the types of different analysis tasks in the sub-analysis tasks according to the task analysis plan, so as to split each sub-analysis task into a plurality of analysis tasks according to the types; Step 1023: Formulate a task priority corresponding to each analysis task based on the task analysis plan.

[0032] In this embodiment, when performing statistics on the results data, the above steps involve synchronous data calls for multiple business scenarios. To avoid the impact of calls between data in different regions and to more accurately and specifically obtain the results data uploaded by each business scenario, a sub-analysis task corresponding to each business scenario can be generated, thereby improving the accuracy of the statistical analysis. Furthermore, the sub-analysis tasks are split into different types of analysis tasks, and then the priorities of the different analysis tasks are determined. This allows the analysis logic of the sub-analysis tasks to be accurately executed through the priorities and splitting, thereby improving the accuracy of the statistical analysis.

[0033] In certain implementations of this embodiment, step 103 preferably allocates computing resources and data resources in the following steps to ensure that each sub-analysis task is accurately executed. Specifically, the steps are: Step 1031: Based on the task analysis plan and the preset data retrieval mechanism, retrieve the land survey result data required for each sub-analysis task from the preset resource library to the preset large land survey result data pool; Step 1032: Obtain the remaining computing resources stored in the resource library and evaluate the occupied computing resources corresponding to each of the analysis tasks to determine the computing resources allocated to each of the sub-analysis tasks.

[0034] In this embodiment, to ensure accurate analysis results for each sub-analysis task, the above steps first call the corresponding land survey results data and computing resources to execute the corresponding analysis process. When calling the land survey results data, the land survey results data required for each sub-analysis task is added to a preset large land survey results data pool. This allows different sub-analysis tasks to access the same land survey results data, thereby reducing the amount of land survey results data used and improving the accuracy and efficiency of statistical analysis.

[0035] In some implementations of this embodiment, step 104 preferably performs distributed computing of several analysis tasks to improve the efficiency and accuracy of statistical analysis. Specifically, the steps are: Step 1041: collecting the operating status of each computing node in real time to determine the multiple task nodes corresponding to each sub-analysis task; Step 1042: determining a calculation order corresponding to a plurality of the analysis tasks according to the task priority corresponding to each of the analysis tasks; Step 1043: Allocate the plurality of analysis tasks to the corresponding computing nodes in sequence for distributed computing according to the occupied computing resources, the computing resources, and the computing order.

[0036] In this embodiment, the above steps, when performing the calculation of a single sub-analysis task, achieve simultaneous calculation of multiple analysis tasks by calling different computing nodes, thereby improving the efficiency of statistical analysis. At the same time, multiple analysis tasks are distributed to different computing nodes using task priorities, so that the calculation of each analysis task is accurately completed according to the priority, thereby improving the accuracy of statistical analysis.

[0037] As a preferred solution, step 1043 can distribute the analysis tasks through the following steps to reduce the usage of computing resources and thus improve the efficiency of statistical analysis. Specifically, the steps are: Step 10431: Determine a plurality of analysis tasks for first-round allocation based on the occupied computing resources, the computing resources, and the computing ranking; Step 10432: polling the operating status of each computing node in real time, and when it is determined that the computing node currently being polled has completed the calculation of the analysis task, releasing the computing resources occupied by the computing node currently being polled and processing the corresponding analysis task, and determining that the computing node is an idle node; Step 10433: Determine the analysis task to be assigned according to the calculation ranking, and when it is determined that the occupied computing resources released by the idle node are greater than or equal to the occupied computing resources required by the analysis task to be assigned, assign the analysis task to the idle node; Step 10434: When the analysis task to be assigned is assigned to the idle node, the occupied computing resources required by the analysis task to be assigned are assigned to the idle node, so that the idle node executes the analysis task to be assigned.

[0038] In this embodiment, the above steps monitor the completion status of each computing node in real time, and release the computing resources occupied by the current computing node in a timely manner, and perform calculations on the next sorted analysis tasks based on the released computing resources and the computing nodes in an idle state. By monitoring the computing nodes and releasing the computing resources, the occupation of computing resources can be reduced, and the distributed computing of multiple analysis tasks can be achieved with fewer computing resources. When allocating computing resources, multiple sub-analysis tasks can be provided for parallel computing, thereby improving the efficiency of statistical analysis.

[0039] In some implementations of this embodiment, the step 105 preferably performs the following steps to execute each analysis task, thereby obtaining the analysis results corresponding to each analysis task to improve the accuracy of the analysis results. Specifically, the steps are: Step 1051: Retrieving the analyzed land survey result data corresponding to the analysis task from the large land survey result data pool; Step 1052: Obtaining a spatial analysis model corresponding to the land survey result data; Step 1053: Allocate the analyzed land survey result data and the spatial analysis model to the computing node where the analysis task is located, so as to output the analysis result corresponding to the analysis task through the computing node.

[0040] In this embodiment, the steps are based on the called analysis of land survey results data, which are land survey results data, and the land survey results data include geographic data with spatial characteristics. Therefore, in order to improve the accuracy of statistical analysis, calling the corresponding spatial analysis model to perform corresponding data analysis can improve the accuracy of statistical analysis.

[0041] In some implementations of this embodiment, step 106 preferably summarizes the analysis results in the following steps to improve the accuracy of the statistical analysis and adapt to different statistical analysis requirements. Specifically, the steps are: Step 1061: Retrieve the verification rule corresponding to the task analysis rule from the pre-built summary statistics rule library; Step 1062: When each of the analysis results meets the verification rule, retrieve the achievement statistical template corresponding to the target analysis task from the pre-built achievement template library; Step 1063: Fill the plurality of analysis results into the achievement statistics template to obtain the target analysis results of the target analysis task.

[0042] In this embodiment, the steps first call the verification rules corresponding to the current target analysis task to verify the analysis results of each analysis task, ensuring the accuracy of the analysis results. Secondly, when all analysis results pass the verification, the statistical template corresponding to the target analysis task is retrieved from the pre-built result template library to display the analysis results in a simple and clear manner to meet different statistical analysis needs.

[0043] On the other hand, this embodiment also discloses a statistical analysis system for land survey results. The specific structure of the statistical analysis system can be referred to Figure 2 , including an analysis and construction module 201, a task splitting module 202, a resource scheduling module 203, a hybrid computing module 204, a task analysis module 205 and an achievement display module 206.

[0044] The analysis building module 201 is used to configure task analysis parameters of the target analysis task to build a task analysis solution corresponding to the target analysis task.

[0045] The task splitting module 202 is used to split the target analysis task into sub-analysis tasks corresponding to each task scenario, and split each sub-analysis task into several analysis tasks according to the task analysis plan.

[0046] The resource scheduling module 203 is used to allocate the computing resources and land survey results data required for each sub-analysis task based on the task analysis plan.

[0047] The hybrid computing module 204 is used to control several of the sub-analysis tasks to perform parallel computing based on the computing resources and the land survey results data, and to distribute several of the analysis tasks to various computing nodes for distributed computing based on the computing resources.

[0048] The task analysis module 205 is used to divide the land survey results data into analysis land survey results data corresponding to each analysis task, and call the spatial analysis model corresponding to the analysis task to obtain the analysis results corresponding to each analysis task based on the analysis land survey results data, the spatial analysis model and the computing node.

[0049] The result display module 206 is used to collect a number of analysis results corresponding to each of the sub-analysis tasks in a number of sub-analysis tasks to obtain the target analysis result of the target analysis task.

[0050] This embodiment discloses a statistical analysis method and system for land survey results. First, the task analysis parameters of the target analysis task are configured to construct a task analysis scheme corresponding to the target analysis task, so as to simplify and accurately construct the analysis process of the target analysis task according to the analysis scheme to ensure the accuracy of the statistical analysis. Secondly, the complex target analysis task is split into sub-analysis tasks corresponding to each task scenario, and then the sub-analysis tasks are split into different types of analysis tasks to reduce the amount of data processed by each analysis task by task splitting, thereby improving the efficiency of statistical analysis. Then, corresponding computing resources and land survey result data are allocated to each analysis task, and parallel computing is performed between sub-analysis tasks, while distributed computing is performed between analysis tasks to improve the efficiency of task execution and thereby improve the efficiency of statistical analysis. Finally, a number of analysis results corresponding to each sub-analysis task in each sub-analysis task are summarized to quickly and accurately obtain the target analysis results of the target analysis task.

[0051] Example 2 During the statistical analysis of land survey results, to meet the ever-changing needs of flexible analysis, analysts often need to manually customize different analysis tasks to perform the corresponding statistical analysis. However, this statistical analysis process not only relies on the user's professional knowledge, resulting in a lack of statistical accuracy, but also is limited by manual operation, making it impossible to meet the time-sensitive requirements of summary analysis tasks. Secondly, the statistical analysis process requires a large workload of survey overlay computations. For example, the summary statistics of provincial land survey results involve multiple survey phases, tens of millions of feature versions, and multiple years of land survey data from the second national land survey and historical change surveys. During the data collection process, at least thousands of county-level overlays are required. The computational workload of generating these overlays and performing statistical analysis based on them is enormous, and traditional land summary statistical techniques cannot efficiently complete this data analysis task.

[0052] In order to solve the technical problems existing in the above-mentioned existing technologies and to realize efficient statistical analysis of massive land survey data, this embodiment provides a land survey results statistical analysis framework based on multi-mode heterogeneous computing technology, so as to perform statistical analysis of land survey results according to the framework, thereby improving the efficiency and accuracy of statistical analysis.

[0053] Specifically, the statistical analysis framework can refer to Figure 3, mainly including spatial big data computing technology, task-driven intelligent scheduling technology, and a summary statistical knowledge base. The spatial big data computing technology includes a big data computing framework, an algorithm model library, and a high-performance computing cluster; the task-driven intelligent scheduling technology includes a unified resource directory, unified computing resource allocation, a unified data access mechanism, unified authority control, and unified task management and scheduling; the unified task management and scheduling includes task decomposition, task scheduling, and task monitoring; and the summary statistical knowledge base includes a basic operator library, a statistical rule library, and a business solution library.

[0054] Among them, the task-driven intelligent scheduling technology involves the intelligent scheduling of data resources and computing resources, and uniformly schedules and manages the corresponding spatial computing analysis models or processes for different analysis tasks, so that the submission, monitoring and management of computing and analysis tasks of spatial big data can be realized through intelligent scheduling technology, thereby improving the accuracy of statistical analysis.

[0055] The spatial big data computing technology is used to split the total analysis task into sub-analysis tasks corresponding to different business scenarios based on the intelligent scheduling technology, and calculate each of the sub-analysis tasks to generate corresponding analysis results.

[0056] The summary statistical knowledge base is used for the constructed overall analysis task of land survey results and the sub-analysis tasks corresponding to different business scenarios formed after the overall analysis task of land survey results is split. Based on the pre-built business rule library, application analysis process library, and result report template library, it realizes the deep integration of general spatial algorithms and land business rules, and accumulates and learns the application analysis scenarios of land survey results through machine learning algorithms, realizes the intelligent scheduling iterative cycle of analysis models and business rules in the business operation process, and realizes the diversified scenario adaptive service mechanism according to the business model, regional scope, result output, user timeliness, cost and other factors in the analysis application scenario, which can intelligently select data storage form, business model (technical route), etc., and effectively meet the diversified application scenario requirements of land big data.

[0057] As a preferred solution, in certain implementations of this embodiment, the spatial big data computing technology is used to improve the efficiency of statistical analysis of survey results and meet the timeliness requirements of statistical analysis of diversified and multi-scale scenarios. Figure 3 The spatial big data computing technology framework shown introduces parallel computing technology and distributed computing technology to select different computing frameworks for summary statistics and application analysis scenarios with different spatial scales and computing characteristics. At the same time, through unified scheduling services for heterogeneous computing, the problems of data skew or unbalanced computing load in the computational analysis of spatial big data are solved.

[0058] The parallel computing technology and distributed computing technology correspond to two computing modes: the first is distributed memory computing, which focuses on solving the efficiency problem of large-scale computing. It uses a distributed storage resource pool (storing data sets used for distributed computing) to support distributed memory computing to solve the high throughput bottleneck of large-scale computing. The second is parallel computing technology, which focuses on solving the problem of batch processing of small and medium-scale data (sub-analysis tasks formed after the task division of the native coupled land by administrative units). It uses a shared storage resource pool to solve the bottleneck of task concurrency in the processing of large-scale data. When using parallel computing technology and distributed computing technology for hybrid computing, parallelism is adopted between different tasks, and distributed memory computing is used for specific tasks.

[0059] Furthermore, on the basis of the hybrid computing model, the high-performance computing cluster established in combination with the hardware layer and the algorithm model library constructed in the software layer provide computing services for the application layer to meet the computing and analysis needs of the total task of statistical analysis of land survey results at different stages or in different regions. Among them, in the technology of the above-mentioned hybrid computing model, the service mode of combining the algorithm model library and the high-performance computing cluster in memory computing is mainly to provide services for data resident in memory and batch processing services for data. Specifically, since distributed memory and parallel computing face different scenarios, or when the scenarios require multiple computing methods to meet them, it is necessary to open up an effective connection between the computing results of distributed memory and parallel computing. Specifically, when opening up the connection to realize the circulation of computing results between different computing modes, the constructed computing framework is as follows: Figure 4 As shown, to support hybrid computing, a unified data model, unified directory structure, and unified regional division are set up in the input and output services to achieve the simultaneous execution of distributed and parallel computing based on the input and output services. Furthermore, after the corresponding computing resources and service resources are provided for distributed computing and parallel computing respectively based on the input data service and resource pool, the distributed computing and parallel computing use their corresponding distributed file systems or shared file systems to store the corresponding intermediate result caches. Then, a preset relational database is used to facilitate bidirectional circulation of the intermediate result caches in the distributed computing and parallel computing.

[0060] Reference Figure 4 In the computing framework shown, in the in-memory data service mode, different models exchange data through memory or disk. In the second service mode, different models exchange data primarily through disk. Parallel computing mainly uses batch processing as a service mode, and data sharing or exchange between different models is done through disk cache.

[0061] Further, refer to Figure 4The calculation framework shown in the figure, the process of data circulation of land survey data based on the calculation framework is as follows Figure 5 Specifically, after building a file library and a spatial library based on land survey data, the file library can directly provide data for computation to the parallel computing cluster. Based on the file and spatial libraries, a unified data retrieval mechanism extracts the data required for each task into a big data resource pool based on the data version, scope, database connection parameters, and other parameters specified during task creation. The data is then computed in a distributed memory computing cluster, thus achieving the fusion of the two computing frameworks.

[0062] As a preferred solution, in certain implementations of this embodiment, in order to ensure the accuracy of statistical summary and analysis results, ensure their accuracy, and at the same time guarantee the efficient operation of statistical summary work, the task-driven intelligent scheduling technology adopts a task-driven data management mechanism and intelligent scheduling technology. According to the work requirements of basic statistics of land surveys, land use change analysis, and comprehensive statistical summary analysis, it realizes the integrated scheduling of statistical summary data flow analysis, original data materials, intermediate results, and statistical results, as well as the automatic splitting of computing tasks and the automatic allocation of computing resources, so as to achieve full utilization of data resources and computing resources through intelligent scheduling of data resources and computing resources.

[0063] Specifically, the task-driven intelligent scheduling technology includes resource management, task-to-resource scheduling management, and task management.

[0064] First, resource management ensures the normal operation of high-performance computing and analysis of spatial big data. It monitors the operating status of computing nodes and business execution, while also displaying comprehensive information such as the operational status of business processes and plan execution. During resource management, the platform and device operating status of each subsystem (including services) of the application system are monitored, and a variety of methods, such as graphics and lists, are used to provide a detailed and vivid display of the operating status of each subsystem and device. It also collects real-time information on the operating status and usage of computing nodes, and is responsible for the application and release of computing resources.

[0065] Secondly, the scheduling of tasks to computing resources is an NP-hard (non-deterministic polynomial problem). With the increase of tasks, large-scale task resource scheduling is a challenge faced by space high-performance computing. It requires a universal and flexible computing resource scheduling method to meet the needs of task diversity. The scheduling of tasks to resources not only focuses on unilateral performance, but also needs to take into account efficiency and fairness of resource allocation. In this regard, the management of tasks to resources in this embodiment can refer to Figure 6 .like Figure 6 As shown in the figure, the scheduling management of tasks to resources mainly includes three aspects: (1) Task priority classification: tasks are queued and sorted according to their priority level and the time when they are submitted, and the order in which they are executed is determined.

[0066] (2) Task scheduling algorithm: According to the order of the task queue and the resource parameters occupied by the task (including the number of processes, memory, etc.), apply for the corresponding computing resources and allocate them to the tasks in turn; if the computing resources are insufficient to allocate, wait for the task that is currently performing the calculation to end, and release and collect the resources. When enough computing resources are accumulated, allocate resources to the queued tasks again; in task scheduling, allocate computing resources to one or more tasks at the same time until the computing resources are allocated.

[0067] (3) Allocation of computing resources: Dynamically monitor the usage of computing resources, including used computing resources and remaining computing resources, monitor, release and recycle computing resources, and allocate idle computing resources to tasks.

[0068] Finally, task scheduling management is a key management measure for achieving load balancing. Load balancing encompasses both task load balancing and data load balancing. It involves using task mapping techniques to rationally allocate computing tasks to each process, reducing inter-process waiting time to achieve the fastest execution speed and shortest execution time. Load balancing techniques include static and dynamic allocation. Static allocation involves pre-determined allocation of computing tasks to each process during algorithm execution. This allocation method is influenced by various factors, such as task size, data volume, inter-process interactions, and the parallelization strategy used in programming. Dynamic allocation involves automatic allocation and adjustment of tasks to each process during algorithm execution. Static allocation is relatively easy to implement, while dynamic allocation is more complex. However, dynamic allocation offers significant advantages when the number of tasks exceeds the number of processors, or when the number of tasks is uncertain during the initial execution of a program or during execution. As the preferred solution, since the land survey results are mostly managed in county-level administrative units, the tasks are mainly divided into counties. After the first batch of tasks are statically allocated to each computing node in sequence and start parallel execution, the execution nodes are dynamically allocated from the tasks to be executed based on the completion status of each node.

[0069] Specifically, the process of combining static allocation and dynamic allocation to perform task scheduling management can be referred to Figure 7 .like Figure 7As shown in the figure, after the data and tasks are decomposed, the dynamic scheduling task tsi is assigned to different computing nodes during the parallel computing process. The tasks of each node are {ts1, ts11}, {ts2, ts8, ts14}, {ts3}, {ts4, ts12, ts16}, {ts5, ts10, ts15}, {ts6, ts9}, {ts7, ts13}, so that the computing tasks are completed dynamically and balanced within the time period (T1 - T0), thereby achieving the following: Figure 7 The tasks shown are parallel computing of Task 3, Task 1, and Task 11 on different computing nodes.

[0070] Furthermore, during task scheduling and management, since land survey results include vector data such as map patches, and vector data is spatial and unstructured, geographic spatial parallel computing differs from traditional parallel computing. Furthermore, vector data is irregular and discrete in spatial distribution or shape, resulting in significant differences in the amount of computation required for spatial analysis of various vector targets. Therefore, the spatial distribution characteristics of vector data and the computational effort required for spatial analysis must be considered when performing data and task partitioning. Therefore, during the parallelization of vector data-based spatial algorithms, spatial computation tasks are decomposed through spatial data partitioning. Based on this task decomposition, further task partitioning is performed to achieve computational and task load balancing for spatial parallel computing.

[0071] Based on the decomposition of spatial computing tasks based on spatial data division, the task scheduling process can refer to Figure 8 .like Figure 8 As shown, the task scheduling management process mainly includes steps S1 to S7. Specifically, the steps are: S1: Start available nodes to determine the legality of the analysis plan; wherein, start available computing nodes to verify the syntax and semantics of the issued analysis plan, determine the legality of the analysis plan according to the task parameters of the analysis plan, interpret the format of the legal analysis plan, determine the rationality based on the content of the analysis plan, feedback confirmation information for the executable analysis plan, and feedback illegal confirmation information for the illegal analysis plan.

[0072] S2: Analyze the dependencies between tasks; analyze whether there are dependencies between the input and output of each subtask, and manage the subtasks in layers according to the data flow.

[0073] S3: Task items are grouped by layers; the tasks to be executed are analyzed and decomposed, converted into internal tasks, and divided based on rule-based resource allocation. The divided subtask sets are moved into the waiting queue, and the six-stage scheduling management process of waiting queue update, scheduling queue update, processing queue update, completion queue management, cancellation queue management and exception queue management is carried out to group task items by layers.

[0074] S4: Add the first-level task items to the candidate queue and traverse the candidate queue to determine whether the dependent items are completed; in a cluster environment, the subtask set waiting to be run in the waiting queue is dynamically allocated to the parallel computing processing nodes in real time according to the process sequence, priority algorithm, resource usage status, etc., as well as the task running status.

[0075] S5: Determine whether the dependency has been completed, add the first-level task to the scheduling queue, dispatch the task items of the scheduling queue, and determine whether the task items of the scheduling queue have reached the maximum concurrency number.

[0076] S6: When it is determined that the maximum number of concurrent connections has been reached, the optimal computing node is selected for computing, and a message is sent to the client to execute the specified plug-in; when it is determined that the maximum number of concurrent connections has not been reached, the process continues to wait.

[0077] S7: When it is determined that the task item is completed, all subordinate task items are taken to the candidate queue, and the calculation is ended after it is determined that all task items are completed; among them, the parallel computing processing node can receive the assigned subtasks in time, start the task processing middleware to perform computational analysis on the task, and at the same time, during the task execution process, the computing node needs to maintain and manage the communication between the nodes, send the current operating status of the computing node in real time, and record the current operating log.

[0078] Among them, when executing steps S1 to S7, the execution process and current execution status of the business are displayed in a graphical visualization manner; alarms for abnormal situations, including alarms for abnormal operation of various subsystems, equipment, storage nodes, computing nodes, etc., are used to troubleshoot and ensure the smooth operation of system tasks by timely discovering and capturing abnormal status information; through comprehensive monitoring and display of logs such as operation, task flow, and subsystem operation of task-centered business processes, administrators can discover, analyze, and solve problems in a timely manner.

[0079] As a preferred solution, in certain implementations of this embodiment, the summary statistical knowledge base includes a rich basic spatial operator library, a summary statistical process model library, a summary statistical rule library, and an outcome template library, thereby realizing the effective integration of general spatial algorithms and land business rules.

[0080] Among them, the basic spatial operator library provides the basic spatial computing capabilities required for summary statistics, including spatial cropping, spatial overlay, vector fusion, key query, field calculation, etc.

[0081] The summary statistics process library uses visual process modeling to build a workflow based on basic spatial operators according to summary statistics processing, which can achieve rapid construction and effective calculation of different summary statistics processes.

[0082] The summary statistics rule base includes area control and adjustment rules, land classification deduction rules, field calculation rules, grouping and merging rules, etc. By combining the summary statistics process and rules, a complete calculation process for summary statistics is realized.

[0083] The achievement template library is oriented to the output of the final summary statistics results, and provides report templates, statement templates, and drawing templates. Based on the achievement templates, personalized output of analysis results can be achieved.

[0084] The design summarizes the statistical knowledge base, provides flexible application for the system, manages various application business knowledge components in a unified manner, can register new application businesses, and delete cancelled application businesses.

[0085] Specifically, the design summary statistical knowledge base includes several business models for unified management of various application business knowledge components. The process of managing the business models in the design summary statistical knowledge base is as follows: Figure 9 As shown, the specific model management process mainly includes steps S8 to S11.

[0086] Step S8: Model registration: When a user registers a new application service, the prototype system automatically backs up the newly registered application service to the database: first, a new application service record is added to the database table; second, the application service entity file is compressed and uploaded to the server disk.

[0087] Step S9: Model update: When the application business is updated, first check whether the model exists on the client. If so, update the client model. The update only requires replacing the original version of the model with the new version of the algorithm; then update the metadata, and compress the new version of the model to replace the original version of the model on the server.

[0088] Step S10: Model download: When a user uses an application business model, the client first searches the local file system for the model to see if it exists. If it does, the local model is directly called. If it does not exist, the client prototype system automatically queries the model from the database, downloads the algorithm, and decompresses it to the client machine for use.

[0089] Step S11: Model deletion: When the user wants to delete an existing environmental application business model, the prototype system will perform two operations: first, delete the selected model record (including metadata information) from the database table; second, delete the corresponding model entity file on the server disk.

[0090] Reference Figures 3 to 9 The statistical analysis framework shown here targets diverse application scenarios and selectively selects implementation technologies based on diverse spatial big data computation methods, including basic algorithm models, business rules, implementation approaches, and business models. Based on the upper-level business application scenarios, the spatial big data computation framework is integrated for business-oriented transformation and application. The basic algorithm models, business rules, implementation approaches, and business models are reorganized to form implementation logic solutions suitable for different business scenarios.

[0091] In one embodiment of this embodiment, refer to Figures 3 to 9 The statistical analysis framework shown in the figure provides a statistical analysis method for land survey results, which is used to improve the efficiency and accuracy of statistical analysis. Figure 10 , mainly including steps 1001 to 1005, specifically, the steps are: Step 1001: Construct an analysis plan for the target analysis task based on the summary statistical knowledge base.

[0092] In this embodiment, according to different analysis requirements and business scenarios, the summary statistical knowledge base provides support for the analysis process, analysis operators, operator parameters and results templates in the process of process building, assisting business personnel to quickly build analysis solutions that meet the scenario requirements.

[0093] Specifically, the construction process of the analysis scheme can refer to Figure 11 , mainly to select a new plan according to the analysis requirements of the target analysis task, then select and import an existing plan from the summary statistics process library of the summary statistics knowledge base, then select a data model for the plan and select an operator for the plan from the basic space operator library of the summary statistics knowledge base and configure operator parameters for the operator from the summary statistics rule library of the summary statistics knowledge base, further, configure the result template from the result template library of the summary statistics knowledge base, and finally check the parameter integrity to generate an analysis plan for the target analysis task.

[0094] Step 1002: Based on the analysis scheme and intelligent scheduling technology, the target analysis task is split into sub-analysis tasks corresponding to different business scenarios, and the sub-analysis tasks are split into different types of analysis tasks.

[0095] In this embodiment, multiple business scenarios for statistical analysis of land survey results data are first determined based on the analysis scope of the target analysis task. Sub-analysis tasks corresponding to different business scenarios are then created according to the analysis plan. Then, based on the different analysis task types, the subtasks of the statistical task are divided into multiple subtasks, including data reading tasks (such as data loading and data splicing), spatial data analysis tasks (such as overlay analysis and cropping analysis), and report data processing tasks (such as report parsing). Priority is assigned to each task based on the business processing sequence specified in the analysis plan.

[0096] Specifically, refer to Figure 12 , after determining to execute the target analysis task, first call the analysis plan corresponding to the target analysis task, and then set whether to use the reused results, that is, determine whether there is a historical analysis plan that is consistent with the current analysis plan, and then determine whether to use the analysis results of the historical analysis plan as the analysis results of the current analysis plan. If it is determined not to use it, set the scope, year and version of the statistical analysis of the land survey results data. Then, determine the multiple business scenarios participating in the statistical analysis based on the set scope to determine the analysis tasks corresponding to each business scenario. Specifically, taking the parameter business scenario including County A as an example, the sub-analysis task corresponding to County A is as follows Figure 13 As shown, this involves calling primary and auxiliary data resources. Operator 1 and the basic statistical operator configured in the analysis plan are then used to execute the corresponding analysis tasks, such as Operator 1 - Verification Operator, Basic Statistics Operator - Verification Operator, and report statistics. Primary data is primarily used to define the spatial scope of the analysis task. Generally, data with full spatial coverage is selected to solidify the spatial scope of a single analysis task. Auxiliary data is automatically filtered and processed within the spatial scope defined by the primary data. Report statistics primarily summarize the results of the task, summarizing various achievements.

[0097] Step 1003: Execute the analysis task based on the spatial big data computing technology, the scheduling technology and the analysis solution.

[0098] In this embodiment, the computing resources and land survey results data required for each of the sub-analysis tasks are allocated based on the task analysis plan, and several sub-analysis tasks are controlled for parallel computing based on the computing resources and the land survey results data. Based on the computing resources, several types of analysis tasks are allocated to various computing nodes for distributed computing, the land survey results data are divided into analysis land survey results data corresponding to each type of analysis task, and the spatial analysis model corresponding to the analysis task is called.

[0099] Specifically, Figure 13When performing analysis and calculation on the sub-analysis task corresponding to County A, the process of using the spatial big data computing technology and the scheduling technology to perform analysis and calculation is as follows: Figure 14 As shown. Figure 14 , task dispatching, task status polling, resource status polling and resource scheduling are performed based on the task scheduling mechanism. Wherein, when the task is dispatched, the task is split into several analysis tasks based on the scheduling calculation, and the task priority of each analysis task is specified, and the divided tasks, scheduled computing resources, and data resources are sent to the spatial big data computing technology framework for parallel computing and distributed computing. Wherein, when scheduling resources, resources are read from pre-built collection libraries, such as file libraries and space libraries, and big data resource pools. Furthermore, when performing the distributed computing, because the computing process needs to call the corresponding spatial analysis model or the corresponding business rules to limit the computing logic, it is also necessary to call the corresponding rules from the pre-built summary statistical knowledge base to the spatial big data computing technology framework.

[0100] Step 1004: Collect the execution information of various subtasks in the hybrid computing framework and the execution information of each computing node. Through the intelligent scheduling mechanism, schedule each subtask to the optimal computing node for processing and calculation, and automatically track the execution status of each analysis subtask. Analyze the analysis and processing results that do not meet the verification rules and provide users with decision-making assistance.

[0101] Step 1005: Synchronize the newly added statistical rules and analysis processes to the summary statistical knowledge base to provide knowledge support for subsequent analysis tasks.

[0102] The statistical analysis method of land survey results provided in this embodiment is targeted at the diversified and multi-scenario application analysis needs of land survey results. In order to solve the computing performance bottleneck of land survey results at the provincial scale and hundreds of millions of map patches, a distributed memory computing technology for land survey results has been developed, and an intelligent scheduling iterative cycle mechanism for business models has been established in combination with land survey business rules. This has formed a big data integration and business computing capability with land survey data as the core, and realized high-performance computing of ultra-large-scale space of land surveys. This provides supercomputing capability support for ensuring the analysis of change survey results, especially the spatial computing of different natural units and different indicators for the management needs of natural resources departments.

[0103] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A statistical analysis method for land survey results, characterized in that: include: Configuring task analysis parameters of the target analysis task to construct a task analysis plan corresponding to the target analysis task; Splitting the target analysis task into sub-analysis tasks corresponding to each task scenario, and splitting each sub-analysis task into several analysis tasks according to the task analysis plan; Allocate computing resources and land survey results data required for each of the sub-analysis tasks based on the task analysis plan; Controlling a plurality of the sub-analysis tasks to perform parallel computing based on the computing resources and the land survey results data, and allocating the plurality of the analysis tasks to various computing nodes for distributed computing based on the computing resources; Dividing the land survey result data into analysis land survey result data corresponding to each analysis task, and calling the spatial analysis model corresponding to the analysis task to obtain analysis results corresponding to each analysis task based on the analysis land survey result data, the spatial analysis model and the computing node; A plurality of analysis results corresponding to each of the plurality of sub-analysis tasks are collected to obtain a target analysis result of the target analysis task.

2. The statistical analysis method of land survey results according to claim 1, characterized in that: The task analysis parameters of the target analysis task are configured to construct a task analysis solution corresponding to the target analysis task, including: Retrieving a corresponding statistical process template from a pre-built summary statistical process library based on the analysis requirements of the target analysis task; Determine the target statistical data type of the target analysis task based on the analysis requirement, and retrieve the corresponding data analysis logic from the pre-built basic spatial operator library based on the target statistical data type and the statistical process template; The task analysis rules of the target analysis task are called from a pre-built summary statistical rule library according to the data analysis logic to construct the task analysis plan based on the statistical process template, the target statistical data type, the data analysis logic and the task analysis rules.

3. The statistical analysis method of land survey results according to claim 2, characterized in that: The task analysis parameters of the target analysis task are configured to construct a task analysis solution corresponding to the target analysis task, including: Retrieving a completed analysis task with the highest similarity to the task analysis plan from a preset completed analysis task library; Determine whether the completed analysis task and the target analysis task have reused analysis results; When it is determined that the reuse analysis result does not exist, the spatial range of the data is configured to the task analysis plan to obtain the task analysis plan configured with the spatial range.

4. The statistical analysis method of land survey results according to claim 1, characterized in that: The target analysis task is split into sub-analysis tasks corresponding to each task scenario, and each sub-analysis task is split into several analysis tasks according to the task analysis plan, including: Determine several business scenarios involved in the target analysis task according to the spatial scope, and generate sub-analysis tasks corresponding to each of the business scenarios based on the task analysis plan; Determining the types of different analysis tasks in the sub-analysis tasks according to the task analysis plan, and splitting each sub-analysis task into several analysis tasks according to the types; A task priority corresponding to each of the analysis tasks is formulated based on the task analysis plan.

5. The statistical analysis method of land survey results according to claim 4, characterized in that: The allocating computing resources and land survey results data required for each of the sub-analysis tasks based on the task analysis plan includes: Based on the task analysis plan and the preset data retrieval mechanism, the land survey result data required for each sub-analysis task is retrieved from the preset resource library to the preset large land survey result data pool; The remaining computing resources stored in the resource library are obtained and the occupied computing resources corresponding to each of the analysis tasks are evaluated to determine the computing resources allocated to each of the sub-analysis tasks.

6. The statistical analysis method of land survey results according to claim 5, characterized in that: The allocating the plurality of analysis tasks to the respective computing nodes for distributed computing based on the computing resources includes: Collecting the operating status of each computing node in real time to determine the multiple task nodes corresponding to each of the sub-analysis tasks; Determining a calculation order corresponding to a plurality of the analysis tasks according to a task priority corresponding to each of the analysis tasks; According to the occupied computing resources, the computing resources, and the computing order, several analysis tasks are sequentially allocated to the corresponding computing nodes for distributed computing.

7. The statistical analysis method of land survey results according to claim 6, characterized in that: The allocating the plurality of analysis tasks to the corresponding computing nodes in sequence for distributed computing according to the occupied computing resources, the computing resources, and the computing order includes: Determining a plurality of the analysis tasks for first-round allocation according to the occupied computing resources, the computing resources, and the computing ranking; polling the operating status of each computing node in real time, and when determining that the computing node currently being polled has completed the calculation of the analysis task, releasing the computing resources occupied by the computing node currently being polled and corresponding to the analysis task being processed, and determining that the computing node is an idle node; Determining the analysis task to be assigned according to the calculation ranking, and assigning the analysis task to be assigned to the idle node when it is determined that the occupied computing resources released by the idle node are greater than or equal to the occupied computing resources required by the analysis task to be assigned; When the analysis task to be assigned is assigned to the idle node, the occupied computing resources required by the analysis task to be assigned are assigned to the idle node, so that the idle node executes the analysis task to be assigned.

8. The statistical analysis method of land survey results according to claim 5, characterized in that: The dividing the land survey result data into analysis land survey result data corresponding to each analysis task, and calling the spatial analysis model corresponding to the analysis task to obtain the analysis result corresponding to each analysis task according to the analysis land survey result data, the spatial analysis model and the computing node, includes: Retrieving the analyzed land survey result data corresponding to the analysis task from the large land survey result data pool; Obtaining a spatial analysis model corresponding to the land survey result data; The analysis land survey result data and the spatial analysis model are distributed to the computing node where the analysis task is located, so as to output the analysis result corresponding to the analysis task through the computing node.

9. The statistical analysis method of land survey results according to claim 2, characterized in that: The step of aggregating the analysis results corresponding to each of the plurality of sub-analysis tasks to obtain the target analysis result of the target analysis task includes: Retrieving the verification rule corresponding to the task analysis rule from the pre-built summary statistical rule library; When each of the analysis results meets the verification rules, the achievement statistical template corresponding to the target analysis task is retrieved from the pre-built achievement template library; Fill the result statistics template with a number of the analysis results to obtain the target analysis results of the target analysis task.

10. A statistical analysis system for land survey results, characterized in that: It includes analysis and construction module, task splitting module, resource scheduling module, hybrid computing module, task analysis module and achievement display module; The analysis building module is used to configure the task analysis parameters of the target analysis task to build a task analysis plan corresponding to the target analysis task; The task splitting module is used to split the target analysis task into sub-analysis tasks corresponding to each task scenario, and split each sub-analysis task into several analysis tasks according to the task analysis plan; The resource scheduling module is used to allocate the computing resources and land survey results data required for each sub-analysis task based on the task analysis plan; The hybrid computing module is used to control a plurality of the sub-analysis tasks to perform parallel computing based on the computing resources and the land survey results data, and to distribute the plurality of the analysis tasks to each computing node for distributed computing based on the computing resources; The task analysis module is used to divide the land survey result data into analysis land survey result data corresponding to each analysis task, and call the spatial analysis model corresponding to the analysis task to obtain the analysis result corresponding to each analysis task based on the analysis land survey result data, the spatial analysis model and the computing node; The result display module is used to collect a number of analysis results corresponding to each of the sub-analysis tasks in a number of sub-analysis tasks to obtain the target analysis result of the target analysis task.

Images