Game resource loading optimization method oriented to distributed storage

Abstract

The invention relates to the technical field of game optimization, and discloses a distributed storage-oriented game resource loading optimization method, which comprises the following steps of: acquiring game modes of a user, performing game resource analysis on a target user, and executing corresponding loading optimization methods according to different game modes. According to the game resource loading optimization method oriented to distributed storage, through an intelligent resource loading strategy, timely loading of key resources is ensured, waiting time is shortened, game fluency and user satisfaction are improved, different optimization strategies are adopted for a single-person mode and a multi-person mode, respective characteristics are fully considered, resource utilization efficiency is improved, and user experience is improved. And automatic judgment and decision are performed according to the user behavior data and communication information, so that manual operation of the user is reduced, and the loading process is more intelligent and efficient.

Description

Technical Field

[0001] This invention relates to the field of game optimization technology, specifically to a method for optimizing game resource loading for distributed storage. Background Technology

[0002] With the evolution of modern games and the explosive growth of resources, the complexity and resource volume of modern games have exceeded the carrying capacity of traditional local storage and centralized distribution models. Meanwhile, players have increasingly higher demands for instant feedback and seamless gaming experiences. Distributed storage technology, with its elasticity, high performance, and high availability, provides a strong technical foundation for solving the pain points of game resource loading. Therefore, "A Game Resource Loading Optimization Method for Distributed Storage" has emerged. It aims to fully utilize the advantages of distributed storage, overcome the limitations of traditional loading methods, and achieve faster, more reliable, and more cost-effective game resource loading, thereby significantly improving the player's gaming experience. It aims to use the "decentralized" and "on-demand" characteristics of distributed storage to completely revolutionize the way game resources are accessed, thereby achieving a faster, smoother, and more immersive gaming experience. Existing optimization methods for game resource loading in distributed storage cannot determine the order and importance of resource loading based on the user's game mode, nor can they determine the user's preferences and habits regarding game instances based on historical data, nor can they determine the user's and teammates' choices and changes regarding game instances based on communication between the user and teammates, nor can they determine the likelihood of the user choosing a game instance based on their historical cooperation with teammates. Therefore, their practicality is somewhat limited. Summary of the Invention

[0003] This invention provides a method for optimizing game resource loading for distributed storage, which helps to solve the problems mentioned in the background art.

[0004] This invention provides the following technical solution: a method for optimizing game resource loading for distributed storage, comprising: Get the mode button for the target game: ; Collect the mode button selected by the target user, define it as the selection button, and denote it as . ; Define a key press analysis function to determine the game mode selected by the target user: ; like If so, the game mode is determined to be single-player mode; like If so, the game mode is determined to be multiplayer mode; Analyze game resources for the target user to determine whether all resources of the current game have been fully loaded. If the target user's resources are normal, no measures will be taken and the user's gaming experience will not be interfered with. If the game mode is single-player and the target user resources are abnormal, then perform single-player resource analysis. If the single-player instance resources have finished loading, then perform a background update; If the resources for a single-player instance have not been fully loaded, then perform optimized loading for single-player instances. If the game mode is multiplayer and the target user resources are abnormal, then collect the target user's communication data and perform multiplayer resource analysis.

[0005] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the following is included: Game resource analysis is performed on the target user, specifically: Obtain all resource data of the target game and generate a resource data list. : ; Define a resource loading analysis function For each resource in the resource data list, determine its loading status: ; like If so, it is determined that the resource data has been loaded completely; like If so, it is determined that the resource data has not been fully loaded; For each resource in the resource data list, retrieve all its dependent resource data, denoted as _____. : ; Define an associated load analysis function For each resource data in the resource data list, determine whether it and its dependent resource data have all been loaded: ; like If so, it is determined that the resource data and its dependent resource data have been loaded. like If so, it is determined that the resource data or its dependent resource data has not been fully loaded; Define a resource partitioning function that, for each resource data in the resource data list, determines whether it is a high-priority data: ; like If so, the resource data is determined to be high-priority data; like If so, the resource data is determined to be low-priority data; Define a resource analysis function Determine if all resources in the target game are loaded correctly for the target user: ; like If so, the target user's resources are deemed normal; like If so, the target user's resources are determined to be abnormal.

[0006] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the following is included: performing single-player resource analysis, specifically: Record the timestamp of the user's first click on the copy: ; Record the copy selected by the user for the first time: ; Check the replica resource loading status to determine if all replica resources and their dependent resource data have been loaded completely: ; like If so, it is determined that the single-player instance resources have been loaded. like If so, it is determined that the single-player instance resources have not been fully loaded.

[0007] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, wherein: single-player optimized loading is performed, specifically: Extract all instances of the target game and generate an instance set. : ; in, Indicates the target game's... One copy; Record the time of the target user's first click on each copy, denoted as . ; Record the click time of the target user switching between each copy, denoted as . ; Calculate the duration of the target user's stay on each copy. : ; Obtain the historical execution count of the target user for each replica, denoted as ; Prioritize loading replicas; Perform phased resource loading on the priority loading copy.

[0008] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the method involves: performing phased resource loading on the priority loading copy, specifically: Retrieve all resource data corresponding to the priority loading replica and form a replica resource list; Using the resource partitioning function, for each resource data in the replica resource list, determine whether it is high-priority data: ; like If so, the resource data is determined to be high-priority data; like If so, the resource data is determined to be low-priority data. Define all high-priority data from the replicas as necessary resources; All low-priority data in the priority replica will be designated as secondary resources. Updates prioritize loading the necessary resources for the replica; Once the necessary resources for the prioritized copy have been loaded and updated, the target user will be notified. While the target user is executing the priority loading of the replica, update and load the secondary resources of the priority loading replica; After the secondary resources of the priority replica have been loaded and updated, a background update is performed.

[0009] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the following is included: collecting communication data of the target user, specifically: Obtain all communication information between the target user and their teammates to form a communication data set. : ; Based on the communication data set, extract the copy keywords from the communication information, and denot them as follows: : ; The keywords for the copy are sorted from highest to lowest frequency: ; Obtain all instances of the target game; Obtain the historical execution count of each instance of the target game by the target user; Sort all instances of the target game according to their historical execution count: ; Identify the initial copy for the target user : ; Check the resource loading status of the replicas to determine if all resources and their dependent resources of the initially selected replicas have been loaded completely: ; like If so, it is determined that the multi-player instance resources have been loaded. Then execute a background update; like If so, it is determined that the multiplayer instance resources have not been fully loaded; Then perform multi-user resource analysis.

[0010] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the following is included: performing multi-user resource analysis, specifically: Determine whether the target user and their teammates choose to change instances; If the instance is changed, extract the instance selected by the target user and their teammates, define it as the changed instance, and denot it as . ; Check the resource loading status of the replicas to determine if all resources and their dependent resources of the initially selected replicas have been loaded completely: ; like If so, it is determined that the replacement of the copy resources has been completed; Then execute a background update; like If so, it is determined that the replacement of the copy resources has not been fully loaded; Then perform multi-user optimized loading; If the copy is not replaced, resources will be loaded in stages for the initially selected copy; If unsure about the replacement, perform multi-user optimization loading.

[0011] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the method includes: performing multi-user optimized loading, which includes new optimized loading, specifically: Obtain the target user's historical game records; Extract all team-up information from the target user's historical game records to form a historical team-up set. : ; Define a collaboration analysis function to determine if there is a historical collaboration record between the target user and their current teammate: ; like If so, it is determined that there is no historical cooperation record between the target user and the current teammate; Obtain all instances of the target game; Then obtain the historical execution count of each instance of the target game by the target user, denoted as . ; Obtain the historical execution count of each instance of the target game by the target user's current teammates, denoted as . ; Record the time of the target user's first click on each copy, denoted as . ; Record the click time of the target user switching between each copy, denoted as . ; Calculate the duration of the target user's stay on each copy. : ; Calculate the overall score for each instance of the target game. : ; Determine the priority copy : ; For priority copies Perform phased resource loading.

[0012] As an optional solution to the game resource loading optimization method for distributed storage described in this invention, the following is included: performing multi-user optimized loading, including the old optimized loading, specifically: Obtain the target user's historical game records; Extract all team-up information from the target user's historical game records to form a historical team-up set: ; Define a collaboration analysis function to determine if there is a historical collaboration record between the target user and their current teammate: ; like If so, it is determined that the target user and the current teammate have a historical cooperation record; Then extract the combination of the target user and the current teammate in the historical team set. All team information is used to form a teammate cooperation set; Extract the number of times each instance of the target game appears in the teammate co-op set, define it as the historical co-op count, and denot it as... ; Calculate the duration of the target user's stay on each copy. : ; Calculate the overall score for each instance of the target game. : ; Determine the priority copy : ; For priority copies Perform phased resource loading.

[0013] The present invention has the following beneficial effects: 1. This game resource loading optimization method for distributed storage determines whether it is single-player or multiplayer mode based on the user's selected mode button. At the same time, it checks the loading status of the game resource package. If it is normal, it does not interfere; if it is abnormal, it is further processed. It quickly identifies the game mode, provides a basis for subsequent resource loading and optimization strategy selection, ensures that the game experience meets user expectations, avoids unnecessary resource waste, detects resource loading problems early, avoids game lag or abnormalities caused by resource shortages, creates conditions for subsequent optimization of the loading process, and ensures the stability of game operation.

[0014] 2. This game resource loading optimization method for distributed storage collects the selected instance when the user selects single-player mode, and determines the resource status of the selected instance, i.e., whether the instance's resources are in an updated state. If the instance's resources are in an updated state, the instance's resource status is considered normal, and a background update is performed. That is, while the user is executing the selected instance, other unupdated resources in the game are updated and loaded. If the instance's resources are not in an updated state, the instance's resource status is considered abnormal, and single-player optimized loading is performed. That is, based on the user's historical execution count of each instance in the game and the duration of time spent in each instance after entering the game this time, the method determines the instance that the user is more likely to execute, i.e., determines the instance that the user will execute. The system identifies the instance with the highest probability of success and assesses its resource status. If the instance's resource status is normal, a background update is performed, updating and loading any un-updated resources while the user is playing the instance. If the instance's resource status is abnormal, a phased loading process is implemented. First, the core and necessary resources of the instance are loaded to allow the user to start the game. Then, during the user's gameplay, non-core and unnecessary resources are updated and loaded, followed by updates and loading of any other un-updated resources. This approach, tailored to the characteristics of single-player mode, accurately collects user behavior data to optimize instance loading strategies. Through phased loading and background updates, resource utilization is maximized while ensuring game smoothness.

[0015] 3. This game resource loading optimization method for distributed storage collects communication data when a user selects multiplayer mode to determine the instance selected by the user and teammates, and judges the resource status of the selected instance. If the resource status of the selected instance is normal, a background update is performed, that is, while the user is playing the instance, other un-updated resources of the game are updated and loaded. If the resource status of the instance is abnormal, the method determines whether to change the instance based on the communication content. If the user and teammates explicitly request to change the instance during communication, the resource status of the changed instance is judged. If the resource status of the changed instance is normal, a background update is performed. If the resource status of the changed instance is abnormal, the method is the same as when the user and teammates did not explicitly request to change the instance during communication. If the user and teammates explicitly request not to change the instance during communication, the instance is loaded in stages, that is, first the core and necessary resources of the instance are loaded so that the user can start the game, and then during the user's game, the non-core and unnecessary resources of the instance are updated and loaded, and then other un-updated resources of the game are updated and loaded. If the user and teammates did not explicitly request to change the instance during communication, the method is the same as when the user and teammates first cooperate. If this is the first time the user and teammates cooperate, the method determines whether to change the instance. The system intelligently selects priority dungeons based on the user's and teammates' historical execution counts for each dungeon, as well as the user's current playtime for each dungeon. Specifically, it selects the dungeon with the highest average execution count and longest playtime for all team members. If the user and teammates are not cooperating for the first time, the system further selects priority dungeons based on their historical cooperation records, including the highest execution counts and longest playtime for each dungeon in their past collaborations. Prioritized dungeons are loaded in stages: first, the core and necessary resources are loaded to allow the user to start playing; then, during gameplay, non-core and unnecessary resources are updated; finally, other un-updated resources are updated. By combining communication data and team collaboration characteristics, the system intelligently determines dungeon selection and loading strategies to enhance the team gaming experience, promote player collaboration, and improve the game's social and fun aspects. By comprehensively considering user behavior history and current needs, the system intelligently selects priority dungeons, ensuring a balance between game performance and resource utilization through staged loading and dynamic updates. Attached Figure Description

[0016] Figure 1 This is a flowchart of the game resource loading optimization method for distributed storage according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1: A method for optimizing game resource loading for distributed storage, see [link / reference] Figure 1 ,include: Get the mode button for the target game: ; in, This indicates a single-player button, which refers to single-player mode, where the user plays as a team by themselves. This indicates a multi-player button, which is a team mode, meaning that the user and their partner form a team to play the game. Collect the mode button selected by the target user, define it as the selection button, and denote it as . ; Define a key press analysis function to determine the game mode selected by the target user: ; like If the game mode is determined to be single-player mode, that is, the mode button selected by the target user is single-player mode. like If the game mode is determined to be multiplayer mode, that is, the mode button selected by the target user is team mode; Analyze game resources for the target user to determine whether all resources of the current game have been fully loaded. If the target user's resources are normal, no measures will be taken and the user's gaming experience will not be interfered with. If the game mode is single-player and the target user resources are abnormal, then perform single-player resource analysis. If the resources for a single-player instance are loaded, a background update will be performed. That is, when the user runs the instance, the resources that the target user has not updated and loaded in the current game will be updated and loaded. If the resources for a single-player instance have not been fully loaded, then perform optimized loading for single-player instances. If the game mode is multiplayer and the target user resources are abnormal, then collect the target user's communication data and perform multiplayer resource analysis.

[0019] This embodiment also provides game resource analysis for target users, specifically: Obtain all resource data of the target game and generate a resource data list. The resource data mentioned above refers to the game's resource package. ; Define a resource loading analysis function For each resource in the resource data list, determine its loading status: ; in, This indicates that the resource data is in a loaded state, meaning that the resource data has been loaded and updated. This indicates that the resource data is not in a loaded state, meaning the resource data has not been fully loaded and updated. This refers to a specific resource data in the resource data list; like If so, it is determined that the resource data has been loaded completely; like If so, it is determined that the resource data has not been fully loaded; For each resource in the resource data list, retrieve all its dependent resource data, denoted as _____. This refers to other resource packages that the resource package directly depends on or is associated with: ; in, This is a dependency analysis function used to describe the dependencies between resource packages. It records other resource packages that each resource package directly depends on in the form of a dictionary or table. Define an associated load analysis function For each resource data in the resource data list, determine whether it and its dependent resource data have all been loaded: ; in, This indicates that all dependent resource data for this resource has been loaded. This indicates that none of the resource data's dependent resource data has been fully loaded; like If so, it is determined that the resource data and its dependent resource data have been loaded. like If so, it is determined that the resource data or its dependent resource data has not been fully loaded; Define a resource partitioning function that, for each resource data in the resource data list, determines whether it is a high-priority data: ; in, This indicates that the resource data belongs to a high-priority resource package. This indicates that the resource data does not belong to a high-priority resource pack, meaning it belongs to a low-priority resource pack. The priority of resource packs is determined by conditions set by game designers. High-priority resources are usually those that directly affect the core functions of the game, such as player character models and main map textures. These resources must be loaded and their integrity confirmed when the game starts. Low-priority resources are those that have a smaller impact on the core functions of the game, such as background music, particle effects, and other sound and visual effects. These resources can be dynamically loaded in the background without affecting the initial operation of the game. In this way, the game's loading time can be optimized, improving the player's initial experience. like If so, the resource data is determined to be high-priority data; like If so, the resource data is determined to be low-priority data; Define a resource analysis function Determine if all resources in the target game are loaded correctly for the target user: ; in, Indicates that, This indicates that all high-priority resource data and all dependent resource data corresponding to each high-priority resource data have been loaded. like If so, the target user's resources are deemed normal; like If so, the target user's resources are determined to be abnormal.

[0020] By employing the above methods and intelligent resource loading strategies, we ensure that critical resources are loaded in a timely manner, reduce waiting time, improve game smoothness and user satisfaction, and adopt different optimization strategies for single-player and multiplayer modes, fully considering their respective characteristics to improve resource utilization efficiency. Based on user behavior data and communication information, we automatically judge and make decisions, reduce manual user operations, and make the loading process more intelligent and efficient.

[0021] Example 2 is an improvement upon Example 1. This method for optimizing game resource loading in distributed storage performs single-player resource analysis, specifically as follows: Record the timestamp of the user's first click on the copy: ; in, This represents the timestamp of the user's first click on a copy, used to determine when the user first selected a copy. This represents the current system timestamp, used to record the time of user operations. This indicates that the user is clicking on the copy for the first time. This means the timestamp will remain unchanged to avoid repeated updates; Record the copy selected by the user for the first time: ; in, This represents the copy identifier initially selected by the user, used for analyzing user behavior. This represents the replica identifier the user is currently in, and it changes as the user switches replicas. This indicates that the user is clicking on the copy for the first time. This indicates that the replica identifier remains unchanged to avoid duplicate updates; Check the replica resource loading status to determine if all replica resources and their dependent resource data have been loaded completely: ; in, This function is used to determine the loading status of replica resources and whether all resources and their dependent resources of the replica have been loaded. The replica identifier initially selected by the user. This indicates that all resources and their dependent resources for the replica have been loaded. like If so, it is determined that the single-player instance resources have been loaded. like If so, it is determined that the single-player instance resources have not been fully loaded.

[0022] This embodiment also provides a method for performing single-user optimized loading, specifically: Extract all instances of the target game and generate an instance set. : ; in, Indicates the target game's... One copy; Record the time of the target user's first click on each copy, denoted as . ; Record the click time of the target user switching between each copy, denoted as . The switching click time is the click time when the target user switches from this copy to another copy, that is, the initial click time of the other copy; Calculate the duration of the target user's stay on each copy. : ; Obtain the historical execution count of the target user for each replica, denoted as ; Prioritize loading replicas; Phased resource loading is performed on the priority copy; Specifically, the process of prioritizing loading replicas is as follows: All instances of the target game will be ranked based on their historical execution count. The values ​​are sorted from largest to smallest to form a sorting list, denoted as . ; All instances of the target game are ranked based on the duration of play. The values ​​are sorted from largest to smallest to form a sorted list, denoted as . ; Calculate the sequential composite score for each replica. : ; in, It is a weighting coefficient between 0 and 1, used to balance the importance of "number of executions" and "duration of stay"; Sort all instances of the target game by their combined scores in descending order to form a combined ranking list; Extract the copy corresponding to the highest overall score from the overall ranking list and designate it as the priority copy to load.

[0023] Specifically, for the priority loading replica, resources are loaded in stages, as follows: Obtain all resource data corresponding to the priority loading replica, that is, all resource packages that constitute the complete functionality of the replica, and form a replica resource list; Using the resource partitioning function, for each resource data in the replica resource list, determine whether it is high-priority data: ; in, This refers to loading resource data from a replica, meaning that a specific resource from the replica is loaded first. This indicates that the resource data does not belong to a high-priority resource package; that is, the resource data belongs to a low-priority resource package. This indicates that the resource data belongs to a high-priority resource package. The priority of resource packages is divided into high and low priorities based on the conditions set by the game designers. High-priority resources are usually those that directly affect the core functions of the game, such as player character models and main map textures. These resources must be loaded and their integrity confirmed when the game starts. Low-priority resources are those that have less impact on the core functions of the game, such as background music, particle effects and other sound and visual effects. These resources can be loaded dynamically in the background without affecting the initial operation of the game. In this way, the game loading time can be optimized and the player's initial experience can be improved. like If so, the resource data is determined to be high-priority data; like If so, the resource data is determined to be low-priority data. Define all high-priority data from the replicas as necessary resources; All low-priority data in the priority replica will be designated as secondary resources. Updates prioritize loading the necessary resources for the replica; After the necessary resources for the priority copy have been loaded and updated, the target user is prompted to execute the copy. While the target user is executing the priority loading of the replica, update and load the secondary resources of the priority loading replica; After the secondary resources of the priority replica have been loaded and updated, a background update is performed.

[0024] Example 3 is an improvement on Example 2. In this example, communication data of the target user is collected, specifically as follows: Obtain all communication information between the target user and their teammates to form a communication data set. The communication information includes in-game text chat logs, voice chat collections, and user behavior logs between users and teammates. This data contains information about player-teammate interactions, such as timestamps, player IDs, teammate IDs, and communication content, and is used to analyze player behavior and game interactions. ; in, Indicates the target user and their teammates' first... One communication message; Based on the communication data set, extract the copy keywords from the communication information, and denot them as follows: : ; in, For communication data sets, Indicates from communication information Keywords extracted from; The keywords for the copy are sorted from highest to lowest frequency: ; in, This is a list of keywords sorted by frequency of occurrence from highest to lowest, used to obtain the copy names most frequently mentioned in user communications. This is a sorting function used to sort data. This indicates that the keyword list is sorted from highest to lowest frequency. This indicates that the sorting is based on the number of times an element appears. This indicates the basis for sorting, specifying the criteria used to sort the data. This represents the count, specifically the number of times a keyword appears in the communication data set. This indicates that the sorting order is descending, i.e., from highest to lowest. Indicates the direction of sorting. This indicates descending order, meaning sorted from highest to lowest. Obtain all instances of the target game; Obtain the historical execution count of each instance of the target game by the target user; Sort all instances of the target game according to their historical execution count: ; in, This is a list of copy history records sorted from highest to lowest by the number of times they were executed. This is the historical data for the replicas, including information such as the number of times each replica has been executed. This indicates that the sorting is based on the number of times the replicas have been executed. This indicates the basis for sorting, specifying the criteria used to sort the data. This indicates the number of times the instance has been executed by players in the past. This indicates that the sorting order is in descending order. Indicates the direction of sorting. This indicates descending order, meaning sorted from highest to lowest. This is a sorting function used to sort data. This indicates that the copy history is sorted from highest to lowest by the number of times it was executed; Identify the initial copy for the target user : ; in, This represents the first element in the sorted keyword list, i.e., the keyword with the highest frequency. This indicates the length of the keyword list, i.e., the number of keywords. This is a function to calculate the length of a list, specifically the length of a keyword list, i.e., the number of keywords. This represents the first element in the sorted list of replica history records, i.e., the replica with the most executions in history. This indicates a check to see if the keyword list is empty. If the keyword list is not empty, then... If true, it means that keywords have been extracted from the communication data and can be used for subsequent analysis; otherwise, if empty, other auxiliary decision-making methods need to be considered. Check the resource loading status of the replicas to determine if all resources and their dependent resources of the initially selected replicas have been loaded completely: ; in, This function is used to determine the loading status of the replica resources and whether all resources and their dependent resource data of the initially selected replica have been loaded. The replica identifier for the initial replica selected by the target user. This indicates that all resources and their dependent resource data for the initial copy have been loaded. like If so, it is determined that the multi-player instance resources have been loaded. Then a background update is performed, that is, when the user executes the copy, the resources that the target user has not updated and loaded in the current game are updated and loaded. like If so, it is determined that the multiplayer instance resources have not been fully loaded; Then perform multi-user resource analysis.

[0025] This embodiment also provides the following: Performing multi-user resource analysis, specifically: Determine whether the target user and their teammates choose to change instances; If the instance is changed, extract the instance selected by the target user and their teammates, define it as the changed instance, and denot it as . ; Check the resource loading status of the replicas to determine if all resources and their dependent resources of the initially selected replicas have been loaded completely: ; in, This function is used to determine the loading status of the replica resources and whether all resources and their dependent resource data of the initially selected replica have been loaded. To change the replica identifier, This indicates that all resources and their dependent resource data for the initial copy have been loaded. like If so, it is determined that the replacement of the copy resources has been completed; Then a background update is performed, that is, when the user executes the copy, the resources that the target user has not updated and loaded in the current game are updated and loaded. like If so, it is determined that the replacement of the copy resources has not been fully loaded; Then perform multi-user optimized loading; If the copy is not replaced, the initial copy will be loaded with resources in stages. The overall steps and process of loading resources in stages for the initial copy are the same as those for loading resources in stages for the copy to be loaded first. If unsure about the replacement, perform multi-user optimized loading; Specifically, determining whether the target user and their teammates choose to change instances involves: Obtain the communication data set between the target user and their teammates. ; Collection of communication data The text communication data is segmented, keywords are extracted, and tagged as a set of text keywords, denoted as . ; Collection of communication data The voice communication data is converted into text, then segmented into words, keywords are extracted, and labeled as a set of voice keywords, denoted as . ; Define a keyword function This is used to determine whether there are explicit keywords related to "replace copy" or "do not replace copy" in the keyword set: ; in, Represents a set of keywords. To integrate a collection of all keywords related to "swapping instances", To integrate a collection of all keywords related to "no dungeon swapping", This represents the intersection, that is, collecting the common elements between two sets. The empty set is defined as two sets that have no common elements. If two sets have common elements, then the empty set is not the same as the empty set. like If so, it is determined that there are explicit "copy swapping" related keywords in the keyword set; like If so, it is determined that there are explicit keywords related to "no copy replacement" in the keyword set; like If no specific keywords related to "replace copy" or "do not replace copy" are found in the keyword set, then it is determined that there are no explicit keywords related to "replace copy" or "do not replace copy". Define an instruction expression function Used to determine whether an instructional expression exists in the keyword set: ; in, Represents a set of keywords. To integrate all sets of directive expressions, which are statements in communication that contain clear suggestions or requirements for action, such as "Let's exchange copies" or "This is it." They are expressed through specific verbs (such as "exchange" and "go") and modal particles (such as "let" and "then"), and are intended to guide or require the other party to perform a certain action. When analyzing communication data, these expressions can be identified through word segmentation and semantic understanding techniques to accurately grasp the communication intent. like If so, it is determined that there is an instructional expression in the keyword set; like If so, it is determined that there is no directive expression in the keyword set; Determine if the target user and their teammates have changed instances: ; in, This indicates that multiple conditions are met simultaneously. It means "or", that is, any condition that is met. like If so, then it is determined that a copy should be replaced; like Then determine whether the target user and their teammates will not change instances: ; like If so, it is determined that the copy will not be replaced; like Then determine whether the target user and their teammates are unsure whether to switch instances: ; like If the result is uncertain, then it is determined that a replacement is needed.

[0026] This embodiment also provides performing multi-user optimized loading, including new optimized loading, specifically: Obtain the target user's historical game records; Extract all team-up information from the target user's historical game records to form a historical team-up set. : ; in, This represents the nth team information of the target user. The team information includes the game win / loss record, game dungeon record, and the game IDs of the target user and the teammates in this team. Define a collaboration analysis function to determine if there is a historical collaboration record between the target user and their current teammate: ; in, The result of determining whether the target user and their current teammates have a historical cooperation record. Indicates historical team assembly This does not include the combination of the target user and the current teammate. For the target user's game ID, The current teammate's game ID. This indicates the combination of the target user and their current teammates; like If no previous cooperation record exists between the target user and the current teammate, it is determined that this is the first time the target user and the current teammate have cooperated. Obtain all instances of the target game; Then obtain the historical execution count of each instance of the target game by the target user, denoted as . ; Obtain the historical execution count of each instance of the target game by the target user's current teammates, denoted as . ; Record the time of the target user's first click on each copy, denoted as . ; Record the click time of the target user switching between each copy, denoted as . The switching click time is the click time when the target user switches from this copy to another copy, that is, the initial click time of the other copy; Calculate the duration of the target user's stay on each copy. : ; Calculate the overall score for each instance of the target game. : ; in, It is a weighting coefficient between 0 and 1, used to balance the impact of replica execution count and dwell time on the overall score. Indicates a specific instance of the target game; Determine the priority copy : ; in, This is a maximum value function used to extract the total score of all instances in the target game. The highest level of dungeon; For priority copies Perform phased resource loading, the priority copy Performing phased resource loading follows the same overall steps and process as performing phased resource loading on the priority loaded copy.

[0027] This embodiment also provides a method for performing multi-user optimized loading, including older optimized loading, specifically: Obtain the target user's historical game records; Extract all team-up information from the target user's historical game records to form a historical team-up set: ; Define a collaboration analysis function to determine if there is a historical collaboration record between the target user and their current teammate: ; in, The result of determining whether the target user and their current teammates have a historical cooperation record. Indicates historical team assembly It includes the combination of the target user and the current teammate. For the target user's game ID, The current teammate's game ID. This indicates the combination of the target user and their current teammates; like If so, it is determined that the target user and the current teammate have a historical cooperation record, that is, the target user and the current teammate are not cooperating for the first time; Then extract the combination of the target user and the current teammate in the historical team set. All team information is used to form a teammate cooperation set; Extract the number of times each instance of the target game appears in the teammate co-op set, define it as the historical co-op count, and denot it as... ; Calculate the duration of the target user's stay on each copy. : ; Calculate the overall score for each instance of the target game. : ; in, It is a weighting coefficient between 0 and 1, used to balance the impact of replica execution count and dwell time on the overall score. Indicates a specific instance of the target game; Determine the priority copy : ; in, This is a maximum value function used to extract the total score of all instances in the target game. The highest level of dungeon; For priority copies Perform phased resource loading, the priority copy Performing phased resource loading follows the same overall steps and process as performing phased resource loading on the priority loaded copy.

[0028] In this embodiment, a smart resource loading strategy ensures that critical resources are loaded in a timely manner, reducing waiting time, improving game smoothness and user satisfaction. Different optimization strategies are adopted for single-player and multiplayer modes, taking into full account their respective characteristics to improve resource utilization efficiency. Automatic judgment and decision-making are based on user behavior data and communication information, reducing manual operation by users and making the loading process more intelligent and efficient.

Claims

1. A method for optimizing game resource loading for distributed storage, characterized in that: include: Get the mode button for the target game: ； Collect the mode button selected by the target user, define it as the selection button, and denote it as . ; Define a key press analysis function to determine the game mode selected by the target user: ； like If so, the game mode is determined to be single-player mode; like If so, the game mode is determined to be multiplayer mode; Analyze game resources for the target user to determine whether all resources of the current game have been fully loaded. If the target user's resources are normal, no measures will be taken and the user's gaming experience will not be interfered with. If the game mode is single-player and the target user resources are abnormal, then perform single-player resource analysis. If the single-player instance resources have finished loading, then perform a background update; If the resources for a single-player instance have not been fully loaded, then perform optimized loading for single-player instances. If the game mode is multiplayer and the target user resources are abnormal, then collect the target user's communication data and perform multiplayer resource analysis.

2. The game resource loading optimization method for distributed storage according to claim 1, characterized in that: Game resource analysis for target users, specifically: Obtain all resource data of the target game and generate a resource data list. : ； Define a resource loading analysis function For each resource in the resource data list, determine its loading status: ； like If so, it is determined that the resource data has been loaded completely; like If so, it is determined that the resource data has not been fully loaded; For each resource in the resource data list, retrieve all its dependent resource data, denoted as _____. : ； Define an associated load analysis function For each resource data in the resource data list, determine whether it and its dependent resource data have all been loaded: ； like If so, it is determined that the resource data and its dependent resource data have been loaded. like If so, it is determined that the resource data or its dependent resource data has not been fully loaded; Define a resource partitioning function that, for each resource data in the resource data list, determines whether it is a high-priority data: ； like If so, the resource data is determined to be high-priority data; like If so, the resource data is determined to be low-priority data; Define a resource analysis function Determine if all resources in the target game are loaded correctly for the target user: ； like If so, the target user's resources are deemed normal; like If so, the target user's resources are determined to be abnormal.

3. The game resource loading optimization method for distributed storage according to claim 1, characterized in that: Perform individual resource analysis, specifically: Record the timestamp of the user's first click on the copy: ； Record the copy selected by the user for the first time: ； Check the replica resource loading status to determine if all replica resources and their dependent resource data have been loaded completely: ； like If so, it is determined that the single-player instance resources have been loaded. like If so, it is determined that the single-player instance resources have not been fully loaded.

4. The game resource loading optimization method for distributed storage according to claim 1, characterized in that: Perform single-player optimized loading, specifically: Extract all instances of the target game and generate an instance set. : ； in, Indicates the target game's... One copy; Record the time of the target user's first click on each copy, denoted as . ; Record the click time of the target user switching between each copy, denoted as . ; Calculate the duration of the target user's stay on each copy. : ； Obtain the historical execution count of the target user for each replica, denoted as ; Prioritize loading replicas; Perform phased resource loading on the priority loading copy.

5. The game resource loading optimization method for distributed storage according to claim 4, characterized in that: For the priority-loaded replica, resources are loaded in stages, specifically as follows: Retrieve all resource data corresponding to the priority loading replica and form a replica resource list; Using the resource partitioning function, for each resource data in the replica resource list, determine whether it is high-priority data: ； like If so, the resource data is determined to be high-priority data; like If so, the resource data is determined to be low-priority data. Define all high-priority data from the replicas as necessary resources; All low-priority data in the priority replica will be designated as secondary resources. Updates prioritize loading the necessary resources for the replica; Once the necessary resources for the prioritized copy have been loaded and updated, the target user will be notified. While the target user is executing the priority loading of the replica, update and load the secondary resources of the priority loading replica; After the secondary resources of the priority replica have been loaded and updated, a background update is performed.

6. The game resource loading optimization method for distributed storage according to claim 1, characterized in that: Collect communication data from the target users, specifically: Obtain all communication information between the target user and their teammates to form a communication data set. : ； Based on the communication data set, extract the copy keywords from the communication information, and denot them as follows: : ； The keywords for the copy are sorted from highest to lowest frequency: ； Obtain all instances of the target game; Obtain the historical execution count of each instance of the target game by the target user; Sort all instances of the target game according to their historical execution count: ； Identify the initial copy for the target user : ； Check the resource loading status of the replicas to determine if all resources and their dependent resources of the initially selected replicas have been loaded completely: ； like If so, it is determined that the multi-player instance resources have been loaded. Then execute a background update; like If so, it is determined that the multiplayer instance resources have not been fully loaded; Then perform multi-user resource analysis.

7. The game resource loading optimization method for distributed storage according to claim 6, characterized in that: Perform multi-person resource analysis, specifically: Determine whether the target user and their teammates choose to change instances; If the instance is changed, extract the instance selected by the target user and their teammates, define it as the changed instance, and denot it as . ; Check the resource loading status of the replicas to determine if all resources and their dependent resources of the initially selected replicas have been loaded completely: ； like If so, it is determined that the replacement of the copy resources has been completed; Then execute a background update; like If so, it is determined that the replacement of the copy resources has not been fully loaded; Then perform multi-user optimized loading; If the copy is not replaced, resources will be loaded in stages for the initially selected copy; If unsure about the replacement, perform multi-user optimization loading.

8. The game resource loading optimization method for distributed storage according to claim 7, characterized in that: Perform multi-user optimized loading, including new optimized loading, specifically: Obtain the target user's historical game records; Extract all team-up information from the target user's historical game records to form a historical team-up set. : ； Define a collaboration analysis function to determine if there is a historical collaboration record between the target user and their current teammate: ； like If so, it is determined that the target user and the current teammate have no historical cooperation record; Obtain all instances of the target game; Then obtain the historical execution count of each instance of the target game by the target user, denoted as . ; Obtain the historical execution count of each instance of the target game by the target user's current teammates, denoted as . ; Record the time of the target user's first click on each copy, denoted as . ; Record the click time of the target user switching between each copy, denoted as . ; Calculate the duration of the target user's stay on each copy. : ； Calculate the overall score for each instance of the target game. : ； Determine priority copy : ； For priority copies Perform phased resource loading.

9. The game resource loading optimization method for distributed storage according to claim 8, characterized in that: Perform multi-user optimized loading, including the old optimized loading, specifically: Obtain the target user's historical game records; Extract all team-up information from the target user's historical game records to form a historical team-up set: ； Define a collaboration analysis function to determine if there is a historical collaboration record between the target user and their current teammate: ； like If so, it is determined that the target user and the current teammate have a historical cooperation record; Then extract the combination of the target user and the current teammate in the historical team set. All team information is used to form a teammate cooperation set; Extract the number of times each instance of the target game appears in the teammate co-op set, define it as the historical co-op count, and denot it as... ; Calculate the duration of the target user's stay on each copy. : ； Calculate the overall score for each instance of the target game. : ； Determine priority copy : ； For priority copies Perform phased resource loading.

Images