CN117785332B - Virtual three-dimensional space dynamic resource loading and releasing method - Google Patents
Virtual three-dimensional space dynamic resource loading and releasing method Download PDFInfo
- Publication number
- CN117785332B CN117785332B CN202410221439.0A CN202410221439A CN117785332B CN 117785332 B CN117785332 B CN 117785332B CN 202410221439 A CN202410221439 A CN 202410221439A CN 117785332 B CN117785332 B CN 117785332B
- Authority
- CN
- China
- Prior art keywords
- data
- resource
- coefficient
- frequency
- constructing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012216 screening Methods 0.000 claims abstract description 10
- 238000012544 monitoring process Methods 0.000 claims description 39
- 238000012360 testing method Methods 0.000 claims description 22
- 238000010606 normalization Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000009877 rendering Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012407 engineering method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Memory System Of A Hierarchy Structure (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a virtual three-dimensional space dynamic resource loading and releasing method, which relates to the technical field of dynamic resource loading, wherein the resource priority of each data group is constructed by a data quality set, and each data group in each data group is loaded in sequence according to the resource priority; constructing an operation state coefficient by an operation state data set, if the operation state coefficient exceeds an operation threshold, observing the use frequency of each item of data in a data set and generating the use frequency of each data set, screening the data set to be cached, caching, identifying the data in a caching state, constructing a caching coefficient by the identified data caching set, and if the acquired caching coefficient exceeds the caching threshold, sending a resource release instruction; according to the resource release strategy corresponding to the data matching in the cache state with the cache characteristics, when the resources are required to be loaded, the loading sequence is predetermined, so that the data of the required resources can be released rapidly.
Description
Technical Field
The invention relates to the technical field of dynamic resource loading, in particular to a virtual three-dimensional space dynamic resource loading and releasing method.
Background
A resource load trigger is a mechanism for automatically triggering a resource load operation when certain conditions are met, the manner in which it is implemented depending on the particular scenario and requirements, for example, these conditions may include: when the application program is started, the required resources can be automatically loaded so as to ensure the normal running of the application program. When the user interacts, the following steps: some resources may need to be loaded when the user performs some action, such as opening a new window or loading a new page. It may be implemented using events or callback functions in the programming language; in other cases, it may be implemented using mechanisms provided by an operating system or framework.
A virtual three-dimensional space refers to a three-dimensional space created by computer technology that can be rendered in a stereoscopic manner on a computer screen. Such techniques may simulate objects, scenes, and environments in the real world and allow a user to interact therewith. Virtual three-dimensional space has applications in many areas, such as game design, architectural design, industrial design, medical simulation, and the like. Can provide richer and vivid visual effect and interactive experience, is beneficial to improving the production efficiency and promotes innovation and creation value.
In the chinese patent application publication No. CN113516769a, a method, an apparatus and a terminal device for loading and rendering a virtual reality three-dimensional scene are provided, where the method includes: acquiring a three-dimensional space coordinate of a user in a virtual reality interaction scene, and projecting the three-dimensional space coordinate on a planar terrain to obtain longitude and latitude coordinates of a user viewpoint; determining a target block three-dimensional scene where a user viewpoint is located according to the longitude and latitude coordinates, and selecting a field block three-dimensional scene according to a preset step length to obtain a candidate loading three-dimensional scene; performing edge scene clipping on the candidate loading three-dimensional scene according to a preset radius to obtain an actual loading three-dimensional scene; and loading and rendering the corresponding block three-dimensional geographic scene model from the three-dimensional scene model resource library in real time according to the actual loaded three-dimensional scene. The technical scheme ensures loading and rendering efficiency, and simultaneously ensures the sense of reality of three-dimensional scene roaming, thereby greatly improving user experience and the like.
Combining the above applications and prior art: when loading and releasing dynamic data in a virtual three-dimensional space, due to the large amount of data, the large amount of data and the poor synchronization between the use states of all data, the use frequency of partial data is high, the use frequency of the other partial data is low, and the partial data used at high frequency can be in a cache state, so that when the needed resource data is loaded and released, the priority of loading or releasing all data is difficult to determine, when a large amount of data is needed, the risk of loading error or releasing error is high, the loading and releasing efficiency is low, and even the partial data has the risk of losing.
Therefore, the invention provides a virtual three-dimensional space dynamic resource loading and releasing method.
Disclosure of Invention
(One) solving the technical problems
Aiming at the defects of the prior art, the invention provides a virtual three-dimensional space dynamic resource loading and releasing method, which comprises the steps of constructing an operation state coefficient by an operation state data set, observing the use frequency of each item of data in a data set and generating the use frequency of each data set if the operation state coefficient exceeds an operation threshold value, screening the data set to be cached and caching the data, identifying the data in a caching state, constructing a caching coefficient by an identified data caching set, and sending a resource releasing instruction if the acquired caching coefficient exceeds the caching threshold value; according to the resource release strategy corresponding to the data matching in the cache state with the cache characteristics, when the resources are required to be loaded, the loading sequence is determined in advance, so that the data of the required resources can be rapidly released, and the technical problem in the background technology is solved.
(II) technical scheme
In order to achieve the above purpose, the invention is realized by the following technical scheme: a dynamic resource loading and releasing method for virtual three-dimensional space includes classifying resource data, identifying the use state of each class of data, and constructing the use degree of each class of resource data by acquiring the use state dataAnd further construct busy coefficients/>If the obtained busy coefficientIf the threshold exceeds the busy threshold, an early warning instruction is sent;
dividing the current resource data into a plurality of data groups, analyzing the data states in the data groups and constructing corresponding data state sets; construction of resource priorities for individual data groups from a data quality set According to resource priority/>Loading each data group in sequence; wherein resource priority/>The acquisition mode of (a) is as follows: for urgency/>Correlation/>Performing linear normalization processing, and mapping corresponding data values to interval/>In, according to the following formula: weight coefficient: /(I) And/>；
Monitoring and acquiring running state data of a computer, constructing a running state data set, and constructing a running state coefficient from the running state data setIf the running state coefficient/>Triggering a resource release mechanism to send out a resource evaluation instruction when the operation threshold is exceeded;
Observing the frequency of use of individual items of data within a data set And generates the frequency of use/>, of each data setScreening and caching the data group to be cached, identifying the data in a cached state, and constructing a cache coefficient/>' by the data caching set obtained by identificationIf the obtained cache coefficient/>The method comprises the steps of exceeding a cache threshold value and sending out a resource release instruction;
predicting the execution of the resource release strategy according to the resource release strategy corresponding to the data matching in the cache state with the cache characteristics, and constructing the performance coefficient by the obtained predicted data Judging whether the current resource release strategy is feasible or not according to the method, and optimizing the resource release strategy if not.
Further, after the resource to be loaded is obtained, classifying the resource data according to the type of the data, obtaining a plurality of resource data types, and obtaining the call data volume of the resource data of each type at each monitoring node in the monitoring periodData reading speed/>After summarizing, constructing a resource reading data set;
obtaining the usage of each resource data category Build busyness coefficient/>The mode is as follows: wherein/> ，/>Weight coefficient is the number of resource data categories: /(I)And/>; Said/>In order to use the average value of the degrees,Is a qualified standard value of the usage degree.
Further, the current data use scene and the required resource data are identified, the data are classified according to the category of the data, a corresponding data set is constructed, the acquired data sets are summarized, and a resource data set to be used is constructed; performing correlation analysis on each data set in the resource data set to be used to obtain the correlation between the current data set and the work task; determining urgency of data call according to time node used by data group in work task and current time difference, and urgency in each array groupCorrelation/>Summarizing, and constructing a data state set of the data group.
Further, the running state of the computer is monitored, and the memory occupancy rate of the computer is monitored at each monitoring nodeAnd the video memory occupancy rate/>, of the GPUMonitoring, namely summarizing the acquired monitoring data and constructing an operation state data set of the computer; construction of the running State coefficient/>The way of (2) is as follows: for memory occupancy/>Display memory occupancy/>After linear normalization processing, mapping the corresponding data value to interval/>In, according to the following formula:
Wherein, Is weight,/>And/>N is the number of monitoring nodes,/>Is the average value of the memory occupancy rate,/>Is the average value of the occupancy rate of the video memory.
Further, after receiving the resource evaluation instruction, an observation period including a plurality of sub-periods is set, and the use frequency of the data group in each sub-period is obtainedSummarizing the acquired frequency data, and constructing a use frequency set of the data set; generating frequency of use/>, of individual data sets from a set of frequency of useIf the frequency of use/>And if the frequency threshold value is exceeded, determining the corresponding data set as the data set to be cached, and caching the data set in the memory.
Further, the frequency of useThe acquisition mode of (a) is as follows: for the frequency of use/>After the linear normalization process, mapping the corresponding data value to interval/>In, according to the following formula: Where n is the number of data,/> Is the difference between the frequency of use of the ith data and the jth data,/>To use the mean value of the frequency,/>Is the frequency of use of the ith data.
Further, the cached data is identified, and the frequency of use of the data in the caching period is obtainedCache duration/>, when cachedAfter summarizing, constructing a cache data set of the data group; construction of cache coefficient/>, from data cache setThe method is as follows: for the frequency of use/>Cache duration/>, when cachedAfter linear normalization processing, mapping the corresponding data value to the interval/>In, according to the following formula: /(I)Wherein, the frequency factor is used for generating the frequency signal,Duration factor,/>，/>Is a constant correction coefficient.
Further, after receiving the resource release instruction, identifying the resource data to be released and the cache state thereof, and obtaining corresponding cache characteristics; the resource release and related words thereof are used as target words, and a resource release knowledge graph is constructed in advance; and giving the resource release strategy by the resource release knowledge graph according to the correspondence between the cache characteristics and the resource release strategy.
Further, testing the resource release strategy by using the trained resource release model, obtaining corresponding test data and summarizing to construct a test data set; construction of coefficient of performance from test data setsIf the coefficient of performance/>And (3) when the release threshold value is lower than the release threshold value, optimizing the current resource release strategy by using the trained optimization model, acquiring the optimized resource release strategy, and executing the optimized resource release strategy to release the cached resource data.
Further, coefficient of performanceThe acquisition mode of (a) is as follows: acquiring the data release speed/>, on each test nodeAfter linear normalization processing is carried out on the data values, the corresponding data values are mapped to the interval/>In, according to the following formula: wherein/> The number of the test nodes; weight coefficient: /(I)And/>; Said/>Is the mean value of the data release speed,/>Is a qualified standard value of the data release speed.
(III) beneficial effects
The invention provides a virtual three-dimensional space dynamic resource loading and releasing method, which has the following beneficial effects:
1. Selecting data to be used and corresponding data groups from the existing resource data, monitoring the execution state of the resource data in each data group, acquiring the corresponding data state, and further constructing the resource priority By resource priority/>The priority degree of loading of each data group can be judged, so that when the resource needs to be loaded, the loading sequence is determined in advance, and the loading efficiency of the resource data is higher.
2. Construction of operational state coefficients from an operational state data setBy constructing the running state coefficient/>The current running state of the computer can be evaluated, if the running state of the computer is good, the resource data can still be in a continuous loading state, the resource data loading task is executed, and conversely, part of the resource data needs to be released, so that the running load is reduced; by the corresponding frequency of use/>Building a corresponding frequency of use/>; Thereby according to the individual frequency of use/>And screening each data group to determine the data group to be cached, so that when the data in the data group has higher use frequency, the data group is cached preferentially for improving the reading efficiency, and the data group can be loaded rapidly in the subsequent loading process.
3. When more data are in the cache state, part of the data group belonging to the cache state needs to be released, and the cache coefficient is constructedAccording to the cache coefficient/>And evaluating each data set so as to judge whether each data set needs to be released, thereby releasing part of the data sets after screening the data sets to be released, reducing the operating pressure of the computer and improving the operating state of the computer.
4. Through feature recognition and knowledge graph construction, corresponding resource release strategies can be quickly matched when resource data are required to be released, so that when the resource release strategies are executed, the required resource data can be quickly released after the resource loading is completed;
5. On the basis of obtaining all parameters of the resource release strategy, the trained optimization model is used for optimizing the resource release strategy, so that the optimized resource release strategy is obtained, and the efficiency of releasing the resource data is higher; by optimizing the resource release strategy, the current resource release strategy library can be enriched, and when the resource data is required to be released later, the diversity of the resource release strategy can be increased.
Drawings
FIG. 1 is a flow chart of a method for loading and releasing dynamic resources in a virtual three-dimensional space according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the present invention provides a method for loading and releasing dynamic resources in a virtual three-dimensional space, which includes the following steps:
Step one, after classifying the resource data, identifying the use state of each category data, and constructing the use degree of each resource data category by acquiring the use state data And further construct busy coefficients/>If the obtained busy coefficient/>If the threshold exceeds the busy threshold, an early warning instruction is sent;
The first step comprises the following steps:
Step 101, after obtaining the resource to be loaded, classifying the resource data according to the type of the data, including: scene, model, texture, sound and the like, obtaining a plurality of resource data categories, and adding a time stamp to each item of resource data according to a time node obtained by the resource data; the use state of each resource data is monitored in the following manner: setting a monitoring period, wherein a plurality of monitoring nodes with equal intervals exist in the monitoring period, and acquiring the call data volume of each category of resource data at each monitoring node Data reading speed/>；
Call data volume to be at each monitoring nodeData reading speed/>Summarizing, and constructing a resource reading data set;
Step 102, constructing the usage degree of each resource data category by the resource reading data set in the following manner:
Amount of scheduling data Data reading speed/>Performing linear normalization processing, and mapping corresponding data values to interval/>In, according to the following formula: /(I)Wherein, the weight coefficient: /(I)And/>To monitor the number of nodes,/>To invoke the mean of the data volume,/>To call the preset standard value of the data quantity,/>The method comprises the steps of monitoring call data volume on a node for an ith monitoring node; /(I)Is the average value of data reading speed,/>Is a preset standard value of data reading speed,/>The data reading speed on the ith monitoring node;
step 103, obtaining the usage degree of each resource data category Build busyness coefficient/>The mode is as follows: wherein/> ，/>Weight coefficient is the number of resource data categories: /(I)And/>；/>For the usage of the ith resource data class, the/>To be the average of the usage degrees,/>Is a qualified standard value of the usage degree.
Presetting a busy threshold according to the use state of the resource data and the data management expectation, if the busy coefficient is the busy coefficientIf the threshold exceeds the busy threshold, an early warning instruction is sent;
In use, the contents of steps 101 and 102 are combined:
When judging whether the resource data needs to be loaded or released, classifying each item of resource data, monitoring the use state of each item of resource data, and constructing the use degree of each item of resource data category by the monitoring data Therefore, the use state of each resource data category can be judged, and the busyness coefficient/>, based on the judgment, can be further constructedEvaluating the overall use state of a plurality of resource data types, if the use of the resource data is more frequent, the use frequency is higher, the data used at high frequency needs to be loaded in advance, otherwise, the data needs to be released in time so as to avoid occupying excessive space, and further, the use degree/>, is constructed in sequenceBusyness coefficient/>The method can screen various resource data, process and arrange the data in advance, process the data in a targeted way and improve the efficiency of resource release and loading.
Combining the above applications and prior art: when loading and releasing the dynamic data in the virtual three-dimensional space, because the data quantity involved is numerous, the data variety is numerous, if the synchronism among the use states of each item of data is poor, the use frequency of part of the data is higher, the use frequency of the other part of the data is lower, and because part of the data used at high frequency can be in a cache state, the priority of loading or releasing each item of data is difficult to determine when the needed resource data is loaded and released, when a large amount of data is needed, the risk of loading error or releasing error is high, the loading and releasing efficiency is lower, and part of the data even has the risk of losing.
Dividing the current resource data into a plurality of data groups, analyzing the data states in the data groups and constructing corresponding data state sets; construction of resource priorities for individual data groups from a data quality setAccording to resource priority/>Loading each data group in sequence;
The second step comprises the following steps:
Step 201, identifying a current data use scene and required resource data according to a work task according to a current work task obtained by inquiry, classifying the data according to the category of the data, constructing a corresponding data set, summarizing the obtained data set, and constructing a resource data set to be used;
Step 202, performing correlation analysis on each data set in a resource data set to be used, and obtaining the correlation between the current data set and a work task; determining urgency of data call according to time node used by data group in work task and current time difference, and urgency in each array group Correlation/>Summarizing, and constructing a data state set of a data group;
Step 203, constructing resource priority of each data group from the data quality set The mode is as follows: to the degree of urgencyCorrelation/>Performing linear normalization processing, and mapping corresponding data values to interval/>In, according to the following formula: weight coefficient: /(I) And/>; The weight coefficient can be obtained by referring to an analytic hierarchy process;
to obtain priority of resources Marking a resource data group in a resource data set to be used according to resource priority/>Sequencing all the data sets to obtain a resource loading sequence; presetting a resource loading triggering rule, and when the resource loading condition is met, according to the resource priority/>Loading each data group in each data group;
in use, the contents of steps 201 to 203 are combined:
If the task is in the execution state, selecting the data to be used and the corresponding data group from the existing resource data according to the task and the use scene thereof, at the moment, monitoring the execution state of the resource data in each data group, acquiring the corresponding data state, and further constructing the resource priority At this time, it is possible to pass resource priority/>The priority degree of loading of each data group can be judged, so that when the resource needs to be loaded, the loading sequence is determined in advance, and the loading efficiency of the resource data is higher.
Step three, monitoring and acquiring the running state data of the computer, constructing a running state data set, and constructing a running state coefficient by the running state data setIf the running state coefficient/>Triggering a resource release mechanism to send out a resource evaluation instruction when the operation threshold is exceeded;
the third step comprises the following steps:
step 301, when a computer is used to execute a task and each resource data set is loaded according to a resource priority, an operation state of the computer is monitored, a plurality of monitoring nodes with equal intervals are provided, and a memory occupancy rate of the computer is set at each monitoring node And the video memory occupancy rate/>, of the GPUMonitoring, namely summarizing the acquired monitoring data and constructing an operation state data set of the computer;
Step 302, constructing operation state coefficients from the operation state data set of the computer The mode is as follows: for memory occupancy/>Display memory occupancy/>After linear normalization processing, mapping the corresponding data value to interval/>In, according to the following formula: /(I)Wherein/>As the weight of the material to be weighed,And/>，/>For the memory occupancy at the ith monitoring node,/>For the video memory occupancy rate at the ith monitoring node,/>N is the number of monitoring nodes,/>Is the average value of the memory occupancy rate,/>Is the average value of the occupancy rate of the video memory;
Presetting an operation threshold according to management expectation of the computer operation state and historical data of executing work tasks, if the operation state coefficient is If the operation threshold is exceeded, triggering a resource release mechanism and sending out a resource evaluation instruction;
in use, the contents of steps 301 and 302 are combined:
When the computer execution is in an operation state, monitoring the operation state of the computer execution and constructing an operation state data set; construction of operational state coefficients from an operational state data set Thus, by constructing the running state coefficients/>The current running state of the computer can be evaluated, if the running state of the computer is good, the resource data can still be in a continuous loading state at the moment, the resource data loading task is executed, and on the contrary, part of the resource data needs to be released, so that the running load is reduced;
Further, by evaluating the running state of the computer, if the running state of the computer is worse in persistence, an operation optimization instruction can be sent out at this time to optimize the running of the computer, so that the performance of the computer is improved, and the loading or releasing efficiency of the subsequent resource data is improved.
Step four, observing the use frequency of each item of data in the data setAnd generates the frequency of use/>, of each data setScreening and caching the data group to be cached, identifying the data in a cached state, and constructing a cache coefficient/>' by the data caching set obtained by identificationIf the obtained cache coefficient/>The method comprises the steps of exceeding a cache threshold value and sending out a resource release instruction;
The fourth step comprises the following steps:
Step 401, after receiving a resource evaluation instruction, setting an observation period including a plurality of sub-periods, and obtaining the frequency of use of the data group in each sub-period Summarizing the acquired frequency data, and constructing a use frequency set of the data set; generating frequency of use/>, of individual data sets from a set of frequency of useThe method is as follows: for the frequency of use/>After the linear normalization process, mapping the corresponding data value to interval/>In, according to the following formula: Where n is the number of data,/> Is the difference between the frequency of use of the ith data and the jth data,/>To use the mean value of the frequency,/>The use frequency of the ith data;
presetting a frequency threshold according to historical data and data processing management expectations, and if the frequency of use is high If the frequency threshold value is exceeded, the corresponding data set is determined to be the data set to be cached, and the data set is cached in the memory;
Step 402, after data caching is completed, identifying cached data, and identifying and acquiring the frequency of use of the data in a caching period Cache duration/>, when cachedAfter summarizing, constructing a cache data set of the data group; the buffer coefficients are constructed from the data buffer set in the following manner: for the frequency of use/>Cache duration/>, when cachedAfter linear normalization processing, mapping the corresponding data value to the interval/>In, according to the following formula: /(I)The significance of the parameters is as follows: frequency factor,/>Duration factor,/>，/>Is a constant correction coefficient, and the value of the constant correction coefficient falls into/>；
It should be noted that: a person skilled in the art collects a plurality of groups of sample data and sets a corresponding preset scaling factor for each group of sample data; substituting the preset proportionality coefficient and the collected sample data into a formula, forming a binary one-time equation set by any two formulas, screening the calculated coefficient and taking an average value, thereby obtaining a frequency factorDuration factor/>Is a value of (a).
The size of the coefficient is a specific numerical value obtained by quantizing each parameter, so that the subsequent comparison is convenient, and the size of the coefficient depends on the number of sample data and the corresponding preset proportional coefficient preliminarily set by a person skilled in the art for each group of sample data, so long as the proportional relation between the parameter and the quantized numerical value is not influenced.
Presetting a cache threshold according to historical data and the expectation of data cache management; if the obtained buffer memory coefficientWhen the buffer threshold is exceeded, the opportunity of reusing the corresponding data set is relatively low, the need of continuing buffer is eliminated, the part of data needs to be released, the subsequent buffer pressure is reduced, and at the moment, a resource release instruction is sent to the outside;
In use, the contents of steps 401 to 402 are combined:
Identifying the use state in each data group and determining the corresponding use frequency Building a corresponding frequency of use/>; Thereby according to the individual frequency of use/>Screening each data group to determine the data group to be cached, so that when the use frequency of the data in the data group is higher, the data group is cached preferentially for improving the reading efficiency, and therefore, the data group can be loaded rapidly in the subsequent loading process;
and as further content: when more data are in the cache state, part of the data group belonging to the cache state needs to be released, and the cache coefficient is constructed According to the cache coefficient/>And evaluating each data set so as to judge whether each data set needs to be released, thereby releasing part of the data sets after screening the data sets to be released, reducing the operating pressure of the computer and improving the operating state of the computer.
Fifthly, predicting execution of the resource release strategy according to the resource release strategy corresponding to the data matching in the cache state with the cache characteristics, and constructing a performance coefficient according to the obtained predicted dataJudging whether the current resource release strategy is feasible or not according to the method, and optimizing the resource release strategy if the current resource release strategy is not feasible;
The fifth step comprises the following steps:
step 501, after receiving a resource release instruction, identifying resource data to be released and a cache state thereof, and obtaining corresponding cache characteristics; the resource release and related words thereof are used as target words, and a resource release knowledge graph is constructed in advance; using the trained matching model, and giving a resource release strategy by a resource release knowledge graph according to the correspondence between the cache characteristics and the resource release strategy;
When the method is used, after a resource release instruction of the resource data is received, the corresponding resource release strategy can be quickly matched when the resource data needs to be released by means of feature recognition and knowledge graph construction, so that when the resource release strategy is executed, the needed resource data can be quickly released after the resource loading is completed;
Step 502, collecting type and quantity data of each item of cache data, specification performance data of a data cache area and the like, extracting part of data from the collected data as sample data, constructing an initial model by a convolutional neural network, training and testing the initial model by the sample data, acquiring a trained resource release model, testing a resource release strategy by using the trained resource release model, acquiring corresponding test data, and summarizing and constructing a test data set;
Step 503, constructing coefficient of performance from the test data set The mode is as follows: acquiring the data release speed/>, on each test nodeAfter linear normalization processing is carried out on the data, the corresponding data value is mapped to the interval/>In, according to the following formula: /(I)Wherein/>The number of the test nodes; weight coefficient: /(I)And/>；/>For the data release rate on the ith test node, the/>Is the mean value of the data release speed,/>Is a qualified standard value of the data release speed.
Presetting a release threshold according to management expectation and historical data of resource release, and if the performance coefficient isWhen the release threshold is lower than the release threshold, the release efficiency and effect of the resource data can be difficult to reach the expectations when the current resource release strategy is executed, and further optimization is required to be carried out on the resource release scheme, and an optimization instruction is sent to the outside; conversely, if the expected situation can be reached, executing a resource release strategy;
When the method is used, after the corresponding resource release strategy is obtained, before execution, the resource release model obtained by training is used for testing the resource release model to judge whether the resource release strategy is reliable, if so, the strategy can be executed to finish the release of the resource, and the loading and release processes of the resource are finished, so that the efficiency and the reliability of the resource release process are higher;
step 504, after receiving the optimization instruction, constructing an initial model by using an ant colony algorithm, training and obtaining a strategy optimization model by using sample data, and optimizing the current resource release strategy by using the trained optimization model to obtain an optimized resource release strategy; executing the optimized resource release strategy to release the cached resource data;
In use, the contents of steps 501 to 504 are combined:
When the execution effect of the matched resource release strategy is difficult to reach the expected value, the resource release strategy is optimized by using the trained optimization model on the basis of acquiring various parameters of the resource release strategy, so that the optimized resource release strategy is acquired, and the efficiency of releasing the resource data is higher; as a further content, the current resource release policy library can be enriched by optimizing the resource release policy, and the diversity of the resource release policy can be increased when the resource data is required to be released subsequently.
The weight coefficient can be obtained by referring to a analytic hierarchy process, the analytic hierarchy process is a qualitative and quantitative combined analytic method, the complex problem can be decomposed into a plurality of layers, a decision maker can be helped to make a decision on the complex problem by comparing the importance of each layer factor, and a final decision scheme is determined.
It should be noted that, the knowledge graph may be constructed by referring to the following:
various data sources related to predictive model optimization are collected, including model training data, validation data, test data, and model performance metrics, feature engineering methods, and the like. Such data may come from a number of sources, such as public data sets, industry reports, academic papers, enterprise internal data, and the like.
And determining the structure of the knowledge graph, including entities, relationships, attributes and the like, according to the collected data and related documents. It is contemplated that the structure of the knowledge-graph may be organized and managed using a graph database or knowledge-graph modeling tool.
And filling the collected data and related information into a knowledge graph, and establishing the association between entities, wherein the association comprises information such as model parameters, an optimization method, feature engineering technology, model performance indexes and the like. This may be done manually or automatically by natural language processing and information extraction techniques.
By using the constructed knowledge graph, the applications such as recommendation of model optimization strategies, model performance analysis, model parameter adjustment and the like can be performed. Related information can be obtained from the knowledge graph by means of a knowledge graph query language (such as SPARQL) or a query function of a graph database, and decision making and optimization models are assisted.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the elements is merely a division of some logic functions, and there may be additional divisions in actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or other various media capable of storing program codes.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application.
Claims (1)
1. A virtual three-dimensional space dynamic resource loading and releasing method is characterized in that: comprising the steps of (a) a step of,
Step one, after classifying the resource data, identifying the use state of each category data, and constructing the use degree of each resource data category by acquiring the use state dataAnd further construct busy coefficients/>If the obtained busy coefficient/>If the threshold exceeds the busy threshold, an early warning instruction is sent; wherein,
Degree of useThe construction method of (2) is as follows: call data volume/>Data reading speed/>Performing linear normalization processing, and mapping corresponding data values to interval/>In, according to the following formula:
wherein, the weight coefficient: and/> In order to monitor the number of nodes,To invoke the mean of the data volume,/>To call the preset standard value of the data quantity,/>The method comprises the steps of monitoring call data volume on a node for an ith monitoring node; /(I)Is the average value of data reading speed,/>Is a preset standard value of data reading speed,/>The data reading speed on the ith monitoring node;
wherein, the usage degree of each resource data category is obtained Build busyness coefficient/>The mode is as follows:
Wherein, Weight coefficient is the number of resource data categories: /(I)And is also provided with；/>To be the average of the usage degrees,/>Is a qualified standard value of the usage degree;
Dividing the current resource data into a plurality of data groups, analyzing the data states in the data groups and constructing corresponding data state sets; construction of resource priorities for individual data groups from a data quality set According to resource priority/>Loading each data group in sequence;
The method comprises the steps of identifying current data use scenes and needed resource data, classifying the data according to the types of the data, constructing corresponding data sets, summarizing the acquired data sets, and constructing a resource data set to be used; performing correlation analysis on each data set in the resource data set to be used to obtain the correlation between the current data set and the work task;
Determining urgency of data call according to time node used by data group in work task and current time difference, and urgency in each array group Correlation/>Summarizing, and constructing a data state set of a data group;
Wherein the resource priority The acquisition mode of (a) is as follows: for urgency/>Correlation/>Performing linear normalization processing, and mapping corresponding data values to interval/>In, according to the following formula:
Weight coefficient: and/> ；
Step three, monitoring and acquiring the running state data of the computer, constructing a running state data set, and constructing a running state coefficient by the running state data setIf the running state coefficient/>Triggering a resource release mechanism to send out a resource evaluation instruction when the operation threshold is exceeded; wherein,
Monitoring the running state of the computer, and controlling the memory occupancy rate of the computer at each monitoring nodeAnd the video memory occupancy rate/>, of the GPUMonitoring, namely summarizing the acquired monitoring data and constructing an operation state data set of the computer;
Construction of the running State coefficient The way of (2) is as follows: for memory occupancy/>Display memory occupancy/>After linear normalization processing, mapping the corresponding data value to interval/>In, according to the following formula:
Wherein, Is weight,/>And/>N is the number of monitoring nodes,/>Is the average value of the memory occupancy rate,/>Is the average value of the occupancy rate of the video memory;
Step four, observing the use frequency of each item of data in the data set And generates the frequency of use/>, of each data setScreening and caching the data group to be cached, identifying the data in a cached state, and constructing a cache coefficient/>' by the data caching set obtained by identificationIf the obtained cache coefficient/>The method comprises the steps of exceeding a cache threshold value and sending out a resource release instruction; wherein,
After the resource to be loaded is acquired, classifying the resource data according to the type of the data, acquiring a plurality of resource data types, and acquiring the call data volume of the resource data of each type at each monitoring node in the monitoring periodData reading speed/>After summarizing, constructing a resource reading data set;
after receiving the resource evaluation instruction, setting an observation period comprising a plurality of subcycles, and obtaining the use frequency of the data group in each subcycle Summarizing the acquired frequency data, and constructing a use frequency set of the data set; generating frequency of use/>, of individual data sets from a set of frequency of useFrequency of use/>The acquisition mode of (a) is as follows: for the frequency of use/>After the linear normalization process, mapping the corresponding data value to interval/>In, according to the following formula:
wherein n is the number of data, Is/>Frequency of use difference between data and jth data,/>To use the mean value of the frequency,/>For/>The frequency of use of the data; if the frequency of use/>If the frequency threshold value is exceeded, the corresponding data set is determined to be the data set to be cached, and the data set is cached in the memory;
identifying the cached data and acquiring the frequency of use of the data in the caching period Buffer time length during buffer storageAfter summarizing, constructing a cache data set of the data group; construction of cache coefficient/>, from data cache setThe method is as follows: for the frequency of use/>Cache duration/>, when cachedAfter linear normalization processing, mapping the corresponding data value to the interval/>In, according to the following formula:
The frequency factor is used to determine the frequency, Duration factor,/>，/>Is a constant correction coefficient;
fifthly, predicting execution of the resource release strategy according to the resource release strategy corresponding to the data matching in the cache state with the cache characteristics, and constructing a performance coefficient according to the obtained predicted data Judging whether the current resource release strategy is feasible or not by using the method, and optimizing the resource release strategy if not, wherein,
After receiving the resource release instruction, identifying the resource data to be released and the cache state thereof, and obtaining corresponding cache characteristics; the resource release and related words thereof are used as target words, and a resource release knowledge graph is constructed in advance; matching the resource release strategy by using the trained matching model according to the correspondence between the cache characteristics and the resource release strategy by using the resource release knowledge graph;
testing the resource release strategy by using the trained resource release model, obtaining corresponding test data and summarizing to construct a test data set; construction of coefficient of performance from test data sets Coefficient of performance/>The acquisition mode of (a) is as follows: acquiring the data release speed/>, on each test nodeAfter linear normalization processing is carried out on the data values, the corresponding data values are mapped into intervalsIn, according to the following formula:
Wherein, The number of the test nodes; weight coefficient: /(I)And is also provided with；/>Is the mean value of the data release speed,/>Is a qualified standard value of the data release speed;
If coefficient of performance And (3) when the release threshold value is lower than the release threshold value, optimizing the current resource release strategy by using the trained optimization model, acquiring the optimized resource release strategy, and executing the optimized resource release strategy to release the cached resource data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410221439.0A CN117785332B (en) | 2024-02-28 | 2024-02-28 | Virtual three-dimensional space dynamic resource loading and releasing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410221439.0A CN117785332B (en) | 2024-02-28 | 2024-02-28 | Virtual three-dimensional space dynamic resource loading and releasing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117785332A CN117785332A (en) | 2024-03-29 |
| CN117785332B true CN117785332B (en) | 2024-05-28 |
Family
ID=90385781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410221439.0A Active CN117785332B (en) | 2024-02-28 | 2024-02-28 | Virtual three-dimensional space dynamic resource loading and releasing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117785332B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102436373A (en) * | 2011-09-13 | 2012-05-02 | 上海普元信息技术股份有限公司 | Method for realizing resource loading and resource hot-updating in distributed enterprise application system |
| CN103092696A (en) * | 2012-12-29 | 2013-05-08 | 深圳先进技术研究院 | Three-dimensional space data pre-dispatching method and system thereof |
| CN107789836A (en) * | 2016-09-06 | 2018-03-13 | 盛趣信息技术（上海）有限公司 | Implementation method and client of a kind of people of game on line thousand with screen |
| CN107846454A (en) * | 2017-10-25 | 2018-03-27 | 暴风集团股份有限公司 | A kind of resource regulating method, device and CDN system |
| CN109542603A (en) * | 2018-11-22 | 2019-03-29 | 北京航空航天大学 | It is a kind of improve different priorities task between isolation multi dimensional resource shielding system |
| CN110245021A (en) * | 2019-06-21 | 2019-09-17 | 上海创功通讯技术有限公司 | EMS memory management process, system, electronic equipment and the storage medium of mobile terminal |
| CN111104454A (en) * | 2019-12-19 | 2020-05-05 | 国网湖南省电力有限公司 | Multi-dimensional data processing method, system and medium for monitoring requirements of power grid on cloud |
| CN112860350A (en) * | 2021-03-15 | 2021-05-28 | 广西师范大学 | Task cache-based computation unloading method in edge computation |
| CN113521753A (en) * | 2021-07-21 | 2021-10-22 | 咪咕互动娱乐有限公司 | System resource adjusting method, device, server and storage medium |
| CN114972664A (en) * | 2022-04-21 | 2022-08-30 | 武汉市测绘研究院 | Integrated management method for urban full-space three-dimensional model data |
| CN115098448A (en) * | 2022-08-26 | 2022-09-23 | 深圳市必凡娱乐科技有限公司 | Software cleaning method and system |
| CN115641404A (en) * | 2022-05-07 | 2023-01-24 | 泰瑞数创科技(北京)股份有限公司 | Mobile rapid modeling system based on live-action three-dimensional modeling technology |
| CN115658171A (en) * | 2022-11-02 | 2023-01-31 | 厦门安胜网络科技有限公司 | Method and system for solving dynamic refreshing of java distributed application configuration in lightweight mode |
| CN116628068A (en) * | 2023-07-25 | 2023-08-22 | 杭州衡泰技术股份有限公司 | Data handling method, system and readable storage medium based on dynamic window |
| CN116916390A (en) * | 2023-09-11 | 2023-10-20 | 军事科学院系统工程研究院系统总体研究所 | Edge collaborative cache optimization method and device combining resource allocation |
| CN117253616A (en) * | 2023-11-17 | 2023-12-19 | 深圳市健怡康医疗器械科技有限公司 | Neurological rehabilitation degree evaluation method and system |
-
2024
- 2024-02-28 CN CN202410221439.0A patent/CN117785332B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102436373A (en) * | 2011-09-13 | 2012-05-02 | 上海普元信息技术股份有限公司 | Method for realizing resource loading and resource hot-updating in distributed enterprise application system |
| CN103092696A (en) * | 2012-12-29 | 2013-05-08 | 深圳先进技术研究院 | Three-dimensional space data pre-dispatching method and system thereof |
| CN107789836A (en) * | 2016-09-06 | 2018-03-13 | 盛趣信息技术（上海）有限公司 | Implementation method and client of a kind of people of game on line thousand with screen |
| CN107846454A (en) * | 2017-10-25 | 2018-03-27 | 暴风集团股份有限公司 | A kind of resource regulating method, device and CDN system |
| CN109542603A (en) * | 2018-11-22 | 2019-03-29 | 北京航空航天大学 | It is a kind of improve different priorities task between isolation multi dimensional resource shielding system |
| CN110245021A (en) * | 2019-06-21 | 2019-09-17 | 上海创功通讯技术有限公司 | EMS memory management process, system, electronic equipment and the storage medium of mobile terminal |
| CN111104454A (en) * | 2019-12-19 | 2020-05-05 | 国网湖南省电力有限公司 | Multi-dimensional data processing method, system and medium for monitoring requirements of power grid on cloud |
| CN112860350A (en) * | 2021-03-15 | 2021-05-28 | 广西师范大学 | Task cache-based computation unloading method in edge computation |
| CN113521753A (en) * | 2021-07-21 | 2021-10-22 | 咪咕互动娱乐有限公司 | System resource adjusting method, device, server and storage medium |
| CN114972664A (en) * | 2022-04-21 | 2022-08-30 | 武汉市测绘研究院 | Integrated management method for urban full-space three-dimensional model data |
| CN115641404A (en) * | 2022-05-07 | 2023-01-24 | 泰瑞数创科技(北京)股份有限公司 | Mobile rapid modeling system based on live-action three-dimensional modeling technology |
| CN115098448A (en) * | 2022-08-26 | 2022-09-23 | 深圳市必凡娱乐科技有限公司 | Software cleaning method and system |
| CN115658171A (en) * | 2022-11-02 | 2023-01-31 | 厦门安胜网络科技有限公司 | Method and system for solving dynamic refreshing of java distributed application configuration in lightweight mode |
| CN116628068A (en) * | 2023-07-25 | 2023-08-22 | 杭州衡泰技术股份有限公司 | Data handling method, system and readable storage medium based on dynamic window |
| CN116916390A (en) * | 2023-09-11 | 2023-10-20 | 军事科学院系统工程研究院系统总体研究所 | Edge collaborative cache optimization method and device combining resource allocation |
| CN117253616A (en) * | 2023-11-17 | 2023-12-19 | 深圳市健怡康医疗器械科技有限公司 | Neurological rehabilitation degree evaluation method and system |
Non-Patent Citations (1)
| Title |
|---|
| 虚拟现实技术在风景园林规划与设计中的应用研究――几种常见虚拟现实技术的应用评价分析;李国松;杨柳青;;中国园林;20080215(02);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117785332A (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951984B (en) | Dynamic analysis and prediction method and device for system health degree | |
| CN111614690B (en) | Abnormal behavior detection method and device | |
| EP2273431B1 (en) | Model determination system | |
| CN112702224B (en) | Method and device for analyzing quality difference of home broadband user | |
| CN113971527A (en) | Data risk assessment method and device based on machine learning | |
| Kwong et al. | Evaluation of the exact conditional spectrum and generalized conditional intensity measure methods for ground motion selection | |
| CN112101692B (en) | Identification method and device for mobile internet bad quality users | |
| CN113297393A (en) | Situation awareness and big data based information generation method and information security system | |
| CN112598443A (en) | Online channel business data processing method and system based on deep learning | |
| CN113297578A (en) | Information perception method and information security system based on big data and artificial intelligence | |
| CN117274259A (en) | Keyboard production equipment management method and system based on Internet of things | |
| CN115529315B (en) | Cloud edge cooperative system | |
| CN111178633A (en) | Method and device for predicting scenic spot passenger flow based on random forest algorithm | |
| CN115204648A (en) | Method and device for acquiring quality index of power engineering based on mathematical model | |
| CN113163353A (en) | Intelligent health service system of power supply vehicle and data transmission method thereof | |
| CN115145817A (en) | Software testing method, device, equipment and readable storage medium | |
| CN111612491B (en) | State analysis model construction method, analysis method and device | |
| CN116909534B (en) | Operator flow generating method, operator flow generating device and storage medium | |
| CN111340975A (en) | Abnormal data feature extraction method, device, equipment and storage medium | |
| CN117785332B (en) | Virtual three-dimensional space dynamic resource loading and releasing method | |
| CN116452747A (en) | BIM model rendering method and system based on multiple scenes | |
| WO2013026389A1 (en) | Method and device for simulation | |
| CN112100165B (en) | Traffic data processing method, system, equipment and medium based on quality assessment | |
| CN111680572A (en) | Power grid operation scene dynamic judgment method and system | |
| CN111898666A (en) | Random forest algorithm and module population combined data variable selection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |