A method for dynamically determining acceleration of data flow

Abstract

The application provides a dynamic acceleration method for data flow, comprising the following steps: obtaining a plurality of data flows corresponding to game data generated after a plurality of players enter a game and inputting the data flows into a flow filter; determining a combined classification attribute of each data flow based on a preset dynamic data flow determination standard, classifying a plurality of data flows according to the combined classification attribute, and obtaining a plurality of data flow sets; determining a corresponding server based on a preset data flow set-server corresponding table, and connecting the plurality of data flow sets to the corresponding server. The method can dynamically accelerate data flow and enable players in multiple regions to enjoy the optimal route without manually switching servers.

Description

Technical Field

[0001] This invention relates to the field of digital information transmission technology, and in particular to a method for accelerating dynamic judgment of data streams. Background Technology

[0002] Currently, for games of a certain scale, game companies often invest a large amount of money in equipping high-performance servers in the early stages of operation, in order to improve the stability and compatibility of the game by running servers deployed in multiple locations, thereby enhancing the player's gaming experience.

[0003] However, there are still some drawbacks in the application server approach. First, due to the geographically dispersed nature of the servers, it is impossible to accelerate data streams from all servers simultaneously. Second, when accelerating a single data stream, users need to perform multiple operations to achieve the desired effect, which often leads to player dissatisfaction. This not only reduces the player's gaming experience but also increases the game company's operation, maintenance, and network costs. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the aforementioned technologies. Therefore, the purpose of this invention is to propose a method for dynamically judging and accelerating data streams, aiming to achieve dynamic acceleration of data streams so that players in multiple regions can enjoy the optimal network connection without manually switching servers.

[0005] To achieve the above objectives, embodiments of the present invention propose an acceleration method for dynamically determining data streams, comprising:

[0006] Acquire several data streams corresponding to the game data generated after several players enter the game and input them into the traffic filter;

[0007] The flow filter determines the combined classification attributes of each data stream based on a preset dynamic judgment standard for data streams, and classifies several data streams according to the combined classification attributes to obtain several data stream sets;

[0008] Based on a preset data stream set-server mapping table, the corresponding server is determined, and the several data stream sets are connected to the corresponding server.

[0009] According to some embodiments of the present invention, the game data includes: player game data, state data of game objects, scene and environment data, social data, historical data, and future data.

[0010] According to some embodiments of the present invention, the data flow dynamic judgment criteria include: the destination address location, the destination address port, the destination domain name, and the protocol.

[0011] According to some embodiments of the present invention, the flow filter determines the combined classification attributes of each data flow based on a preset data flow dynamic judgment standard, including:

[0012] Based on the location of the target address, the first classification attribute of the data stream is obtained;

[0013] Based on the target address port, the second classification attribute of the data stream is obtained;

[0014] Based on the target domain name, the third classification attribute of the data stream is obtained;

[0015] Based on the protocol, the fourth classification attribute of the data stream;

[0016] The first classification attribute, the second classification attribute, the third classification attribute, and the fourth classification attribute are combined to obtain a combined classification attribute.

[0017] According to some embodiments of the present invention, it further includes:

[0018] Real-time monitoring and acquisition of status data from each server;

[0019] Based on the state data, the probability value that the current data stream set can be connected to the corresponding default server is calculated;

[0020] When the probability value is lower than a preset probability threshold, the current data stream set is switched to another server corresponding to the current data stream set.

[0021] According to some embodiments of the present invention, before connecting the plurality of data stream sets to the corresponding server, the method further includes:

[0022] Based on preset server synchronization rules, the servers are filtered to determine the servers that can be synchronized, which are then used as the target server set, and the corresponding data stream sets are synchronized and connected.

[0023] According to some embodiments of the present invention, the method further includes: detecting and displaying status information of the data stream set being connected to the corresponding server.

[0024] According to some embodiments of the present invention, the status information includes connection success, connection failure, and poor connection status.

[0025] According to some embodiments of the present invention, acquiring the data stream corresponding to the game data generated after a player enters the game includes:

[0026] Acquire game data generated after players enter the game, and use it as game data to be processed;

[0027] The game data to be processed is de-identified to obtain the first processed game data;

[0028] Obtain the working status of each data processing node and determine the set of available data processing nodes;

[0029] Based on the data type of the first processed game data and the preset processing method corresponding to the data type, a target data processing node set and the working order of the target data processing nodes in the target data processing node set are determined from the available data processing node set.

[0030] A data processing channel is constructed based on the target data processing node set and the working order of the target data processing nodes;

[0031] The first processed game data is input into the data processing channel for processing to obtain the second processed game data and acquire the corresponding data stream.

[0032] According to some embodiments of the present invention, obtaining the status data of each server includes:

[0033] Obtain historical status data for each server and obtain the corresponding historical performance data sequence set;

[0034] Based on the aforementioned historical performance data sequence set, construct the corresponding historical performance data detail table;

[0035] Based on historical performance data details, a server performance prediction model was built and trained.

[0036] Obtain and parse the server status detection command, and obtain the server performance data from the current time point to the preset time interval, as a performance data detection set;

[0037] Based on the server performance prediction model and the performance data detection set, several initial performance prediction values ​​are obtained; these initial performance prediction values ​​are then corrected to obtain several final performance prediction values.

[0038] Based on the aforementioned final performance prediction values, the server's performance prediction data for the current time node is obtained, and the performance prediction data and the performance data for the current time node are used as the server's status data.

[0039] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0040] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0041] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0042] Figure 1 This is a flowchart of a method for accelerating dynamic data flow determination according to the present invention;

[0043] Figure 2 This is a flowchart illustrating the determination of combined classification attributes for each of the data streams according to an embodiment of the present invention;

[0044] Figure 3 This is a flowchart illustrating, according to an embodiment of the present invention, a method for determining whether the correspondence between the current data stream set and the server needs to be changed. Detailed Implementation

[0045] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0046] like Figure 1 As shown in the figure, this embodiment of the invention proposes a method for accelerating dynamic data stream determination, including:

[0047] Acquire several data streams corresponding to the game data generated after several players enter the game and input them into the traffic filter;

[0048] The flow filter determines the combined classification attributes of each data stream based on a preset dynamic judgment standard for data streams, and classifies several data streams according to the combined classification attributes to obtain several data stream sets;

[0049] Based on a preset data stream set-server mapping table, the corresponding server is determined, and the several data stream sets are connected to the corresponding server.

[0050] The working principle of the above technical solution:

[0051] In this embodiment, the flow filter filters the data stream according to the dynamic judgment criteria of the data stream, thereby classifying the data stream;

[0052] In this embodiment, the preset dynamic judgment criteria for data flow include four aspects: the location of the target address, the port of the target address, the target domain name and protocol;

[0053] In this embodiment, the combined classification attribute is a combination of four aspects, including the target address location, target address port, target domain name, and protocol, obtained by the traffic filter based on a preset dynamic judgment standard for data flow; for example, target address location a + target address port b + target domain name c + protocol;

[0054] In this embodiment, the data stream set-server mapping table is established in advance based on the relationship between the data stream set and the server. For example, data stream set Q corresponds to server F1, and data stream set R corresponds to server F2.

[0055] The system acquires several data streams corresponding to game data generated after several players enter the game and inputs them into a traffic filter; this facilitates the classification of the data streams through the traffic filter.

[0056] The flow filter determines the combined classification attributes of each data stream based on a preset dynamic judgment standard for data streams, classifies several data streams according to the combined classification attributes, and obtains several data stream sets; and performs comprehensive classification of the data streams according to the combined classification attributes.

[0057] Based on a preset data stream set-server mapping table, the corresponding server is determined, and the data stream sets are connected to the corresponding server; this facilitates connecting the data stream sets to the most suitable server and improves the game's response speed.

[0058] The beneficial effects of the above technical solution are as follows: By acquiring several data streams corresponding to game data, it is convenient to transmit them quickly in the form of data streams. Classifying the data streams according to their combined classification attributes enables comprehensive and dynamic classification of the data streams, which helps to improve the overall acceleration of the data stream. By connecting the several data stream sets to the corresponding servers according to the data stream set-server mapping table, the compatibility between the data stream sets and the servers is improved, thereby increasing the success rate of data stream access to the server. Furthermore, without manual operation by the game player, the system can automatically switch servers based on the classification of the data streams, always maintaining the optimal connection line, thus improving the player's gaming experience.

[0059] According to some embodiments of the present invention, the game data includes: player game data, state data of game objects, scene and environment data, social data, historical data, and future data.

[0060] The working principle of the above technical solution:

[0061] In this embodiment, player game data includes player behavior in the game, such as the number of times a player buys items and the length of time a player watches the game.

[0062] In this embodiment, the state data of the game object includes the physical state and dynamic state of the game object, as well as its animation state, etc.

[0063] In this embodiment, scene and environment data include environmental parameters during game runtime, such as time, date, map coordinates, etc.

[0064] In this embodiment, social data includes player interaction behaviors, such as friend relationships, chat communication, game movement trends, etc.

[0065] In this embodiment, historical data includes game running history data, which is constantly changing, such as game score, level completion, recording time, etc.

[0066] In this embodiment, future data, including future data during game operation, can be preset or calculated in real time based on player behavior.

[0067] The beneficial effects of the above technical solution are as follows: by clarifying the specific aspects included in the game data, it is convenient to process the game data comprehensively, and at the same time, it is possible to process different game data in a targeted manner, thereby improving the processing efficiency of game data, obtaining the corresponding data stream, improving the game running speed, and enhancing the player's gaming experience.

[0068] According to some embodiments of the present invention, the data flow dynamic judgment criteria include: the destination address location, the destination address port, the destination domain name, and the protocol.

[0069] The working principle and beneficial effects of the above technical solution are as follows: The dynamic judgment criteria for data flow include: the location of the target address, the port of the target address, the domain name of the target, and the protocol.

[0070] In this embodiment, the destination address location refers to the location of the player's IP address;

[0071] In this embodiment, the target address port represents the player's port type;

[0072] In this embodiment, the target domain name is the identifier of the player's computer, and there is a mapping relationship between the target domain name and the IP address;

[0073] In this embodiment, the protocol identifies a set of conventions that the player's computer follows when communicating, including syntax, semantics, and timing.

[0074] Classifying data streams based on the aforementioned dynamic judgment criteria is more applicable to classifying data streams for games with multiple regions and servers. It can accelerate all types of data streams generated in different regions and optimize the player's gaming experience.

[0075] like Figure 2 As shown, according to some embodiments of the present invention, the traffic filter determines the combined classification attributes of each data stream based on a preset data stream dynamic judgment standard, including:

[0076] S1: Based on the location of the target address, obtain the first classification attribute of the data stream;

[0077] S2: Based on the target address port, obtain the second classification attribute of the data stream;

[0078] S3: Based on the target domain name, obtain the third classification attribute of the data stream;

[0079] S4: Based on the protocol, the fourth classification attribute of the data stream;

[0080] S5: Combine the first classification attribute, the second classification attribute, the third classification attribute, and the fourth classification attribute to obtain a combined classification attribute.

[0081] The working principle of the above technical solution:

[0082] In this embodiment, the first classification attribute is the attribute of the player's target address location. For example, if the player's target address location is r, then the first classification attribute is R1.

[0083] In this embodiment, the second classification attribute is the attribute of the player's target address port. For example, if the player's target address port is d, then the second classification attribute is D1.

[0084] In this embodiment, the third category attribute is the attribute of the player's target domain name. For example, if the player's target domain name is y, then the third category attribute is Y1.

[0085] In this embodiment, the fourth category attribute is the attribute of the protocol followed by the player. For example, if the player's protocol is p, then the third category attribute is P1.

[0086] In this embodiment, based on the above four classification attributes, the combined classification attribute is R1+D1+Y1+P1.

[0087] The beneficial effects of the above technical solution are as follows: Based on the destination address location, destination address port, destination domain name, and protocol of the data stream, it is easier to clearly define the four basic attributes of the data stream, comprehensively classifying the data stream type, resulting in more stable classification results. Furthermore, since the type of each data stream is not defined before its generation, but is dynamically determined after generation, it is easier to improve the accuracy of identifying the type of each data stream, thereby matching appropriate servers, improving data stream acceleration, and enhancing the player's gaming experience.

[0088] like Figure 3 As shown, according to some embodiments of the present invention, it further includes:

[0089] C1: Real-time monitoring and acquisition of status data from each server;

[0090] C2: Based on the state data, calculate the probability value that the current data stream set can access the corresponding default server;

[0091] C3: When the probability value is lower than the preset probability threshold, the current data stream set is switched to another server corresponding to the current data stream set.

[0092] The working principle of the above technical solution is as follows: C1: Real-time monitoring and acquisition of status data of each server; facilitating timely understanding of the server status and prompt response to abnormal situations;

[0093] C2: Based on the state data, calculate the probability value that the target data stream set can access the corresponding default server; this facilitates making a judgment in advance on whether the access can be successful and avoids wasting resources.

[0094] C3: When the probability value is lower than a preset probability threshold, the current data stream set is switched to another server corresponding to the current data stream set; this can promptly and effectively take measures to switch other servers for data streams with a very low success rate of access.

[0095] The beneficial effects of the above technical solution are as follows: By monitoring and acquiring server status data in real time, the server's status can be understood promptly, allowing for timely responses to abnormal server conditions, and providing data-driven support for these responses. Calculating the probability of successfully connecting to the corresponding default server allows for comparison with a probability threshold before connection. Based on the comparison result, a judgment can be made regarding the success rate of connection. If the success rate is low, other servers can be switched promptly to ensure a higher success rate, thereby accelerating data flow and improving the gaming experience.

[0096] According to some embodiments of the present invention, before connecting the plurality of data stream sets to the corresponding server, the method further includes:

[0097] Based on preset server synchronization rules, the servers are filtered to determine the servers that can be synchronized, which are then used as the target server set, and the corresponding data stream sets are synchronized and connected.

[0098] The working principle of the above technical solution:

[0099] In this embodiment, the server synchronization rules include state synchronization rules, stateless synchronization rules, and stateless synchronization rules; the stateless synchronization rules include frame synchronization rules and state frame synchronization rules.

[0100] The beneficial effects of the above technical solution are as follows: Based on the preset server synchronization rules, the servers are screened to determine the servers that can be synchronized, which are then used as the target server set, and the corresponding data stream sets are synchronized and connected; this facilitates the determination of servers that are compatible with the game's needs before connecting them, ensuring the safe and sustainable operation of the game.

[0101] According to some embodiments of the present invention, the method further includes: detecting and displaying status information of the data stream set being connected to the corresponding server.

[0102] The working principle and beneficial effects of the above technical solution:

[0103] In this embodiment, the status information includes connection success, connection failure, and poor connection status;

[0104] The system detects and displays the status information of the data stream collection connected to the corresponding server. This makes it easy to clearly understand the status of the data stream collection connected to the corresponding server. When the status information is a connection failure or a poor connection status, it is easy to adjust the subsequent data stream collection connection operations in a timely manner. Displaying the status information also helps players and technical personnel to understand the game's operation status in a timely manner.

[0105] According to some embodiments of the present invention, the status information includes connection success, connection failure, and poor connection status.

[0106] The working principle and beneficial effects of the above technical solution are: it facilitates the determination of whether the correspondence between data streams and server access needs to be adjusted based on different status information, and provides a basis for this determination.

[0107] According to some embodiments of the present invention, when the status information is determined to be successful access, the access duration of the data stream generated by the player through game operations after entering the game is determined to be accessed to the server.

[0108] The system determines to connect the data stream to several nodes of the server and obtains the execution data of the nodes, including the number of processes of the nodes and the power consumption per unit time corresponding to each process.

[0109] Based on the number of processes in the node and the power consumption per unit time for each process, determine the total power consumption of the node per unit time. Based on the total power consumption per unit time, query the preset total power consumption-response time data table to determine the response time of the node.

[0110] Calculate the server's latency parameters based on the response times and access times of several nodes:

[0111]

[0112] Where YC is the latency parameter of the server, N is the number of nodes included in the server; T n S represents the response time of the nth node in the server; S represents the access time.

[0113] The server's latency parameter is compared with a preset parameter threshold. When the server's latency parameter exceeds the preset parameter threshold, an alarm is issued and the server's performance is adjusted.

[0114] The working principle of the above technical solution is as follows: when the status information indicates successful access, the access time from when the player enters the game and generates a data stream through game operations to when it accesses the server is determined; this makes it easier to clarify the time taken for the data stream to successfully access the server.

[0115] The system determines which nodes will connect the data stream to the server and obtains the execution data of the nodes. The execution data includes the number of processes on the node and the power consumption per unit time for each process. This allows the system to understand the load of the nodes based on the power consumption.

[0116] Based on the number of processes in a node and the power consumption per unit time for each process, the total power consumption of the node per unit time is determined. Based on the total power consumption per unit time, a preset total power consumption-response time data table is queried to determine the response time of the node; this facilitates the quick retrieval of the node's response time through the total power consumption-response time data table.

[0117] The server's latency parameters are calculated based on the response time and access time of several nodes; this allows for assessment of the server's latency status.

[0118] The server's latency parameter is compared with a preset threshold. When the server's latency parameter exceeds the preset threshold, an alarm is issued and the server's performance is adjusted to ensure that the server always maintains a good state and guarantees the game's running speed.

[0119] The beneficial effects of the above technical solution are as follows: By acquiring the access duration of a data stream successfully connecting to the server, the server latency parameter is obtained by subtracting the access duration from the response duration of all parameter nodes. This latency parameter allows for a judgment of the server's performance. When the latency parameter exceeds a preset threshold, an alarm can be issued promptly, and server performance can be adjusted to ensure game speed and improve the player's gaming experience.

[0120] According to some embodiments of the present invention, acquiring the data stream corresponding to the game data generated after a player enters the game includes:

[0121] Acquire game data generated after players enter the game, and use it as game data to be processed;

[0122] The game data to be processed is de-identified to obtain the first processed game data;

[0123] Obtain the working status of each data processing node and determine the set of available data processing nodes;

[0124] Based on the data type of the first processed game data and the preset processing method corresponding to the data type, a target data processing node set and the working order of the target data processing nodes in the target data processing node set are determined from the available data processing node set.

[0125] A data processing channel is constructed based on the target data processing node set and the working order of the target data processing nodes;

[0126] The first processed game data is input into the data processing channel for processing to obtain the second processed game data and acquire the corresponding data stream.

[0127] The working principle of the above technical solution:

[0128] In this embodiment, the preset processing method includes data processing types and the order of data processing; the data processing types include data deduplication, data conversion, and image processing, etc.; the order of data processing is the sequence of the different data processing types. For example, when the data type of the first processed game data is A, the corresponding preset processing method is C1. At this time, the preset processing method C1 includes data processing types L1, L2, and L3, and the order of data processing is L1-L3-L2.

[0129] In this embodiment, the data processing channel is constructed by establishing an association between target data processing nodes according to their working order;

[0130] Acquire game data generated after players enter the game, and use it as game data to be processed;

[0131] The game data to be processed is input into a preset sensitive identification model to obtain sensitive identification parameters; the sensitive identification model is a learning model built based on preset sensitive identification rules; the sensitive identification rules include privacy information identification rules and uncivilized information identification rules.

[0132] The identification parameters are compared with a sensitive identification threshold. When the identification parameters are greater than the sensitive identification threshold, the game data to be processed is desensitized to obtain the first processed game data. This helps protect the privacy of players and ensures civilized gameplay.

[0133] The working status of each data processing node is obtained, and the set of available data processing nodes is determined. The set of available data processing nodes includes at least two data processing nodes. This ensures that the game data to be processed can be processed in a timely manner.

[0134] Based on the data type of the first game data to be processed and the preset processing method corresponding to the data type, a target data processing node set and the working order of the target data processing nodes in the target data processing node set are determined from the set of available data processing nodes. This facilitates the rapid determination of the target data processing node set, and by determining the order in which the game data to be processed is processed, it is easier to ensure that the processing work can be carried out in an orderly manner.

[0135] A data processing channel is constructed based on the target data processing node set and the working order of the target data processing nodes; this facilitates improved efficiency in processing game data. The target data processing node set is determined from the available data processing node set according to a preset processing method, identifying the nodes to process the data processing type. For example, given an existing available data processing node set (k1, k2, k3), the preset processing method corresponding to the first processed game data is C2 (where data processing types include L1 and L2). In this case, the target data processing node set (k1, k2) is determined for corresponding data processing. The working order of the target data processing nodes is determined based on the data processing types L1 and L2 and the data processing order relationship L2-L1 included in the preset processing method. The working order of the target data processing nodes is then determined as follows: k1 performs L2 processing, and then k2 performs L1 processing.

[0136] The first processed game data is input into the data processing channel for processing to obtain the second processed game data and acquire the corresponding data stream.

[0137] The beneficial effects of the above technical solution are as follows: By anonymizing the acquired player game data, it is easier to protect player privacy and provide a more civilized environment in terms of game social interaction, thus optimizing the user experience. By constructing a data processing channel, data processing speed can be accelerated, time consumption reduced, and data processing efficiency improved, thereby quickly obtaining the expected data stream.

[0138] According to some embodiments of the present invention, obtaining the status data of each server includes:

[0139] Obtain historical status data for each server and obtain the corresponding historical performance data sequence set;

[0140] Based on the aforementioned historical performance data sequence set, construct the corresponding historical performance data detail table;

[0141] Based on historical performance data details, a server performance prediction model was built and trained.

[0142] Obtain and parse the server status detection command, and obtain the server performance data from the current time point to the preset time interval, as a performance data detection set;

[0143] Based on the server performance prediction model and the performance data detection set, several initial performance prediction values ​​are obtained; these initial performance prediction values ​​are then corrected to obtain several final performance prediction values.

[0144] Based on the aforementioned final performance prediction values, the server's performance prediction data for the current time node is obtained; and the performance prediction data and the performance data for the current time node are used as the server's status data.

[0145] The working principle of the above technical solution:

[0146] In this embodiment, historical performance data includes CPU utilization, disk utilization, memory utilization, and I / O utilization.

[0147] Acquire historical status data for each server, including historical performance data and time data; this facilitates prediction of future performance based on the server's historical performance data.

[0148] The historical performance data of each server is extracted and sorted based on the time data to obtain the corresponding historical performance data sequence set; this facilitates the obtaining of a historical performance data sequence set with a timeline, providing reliable basic data for predicting future performance data;

[0149] Based on the aforementioned historical performance data sequence set, a corresponding historical performance data detail table is constructed; this facilitates viewing the data in a table format.

[0150] The historical performance data in the historical performance data details table corresponding to each server is used as sample data, and the feature data and label data corresponding to each sample data are obtained; this facilitates more effective training of the server performance prediction model based on the feature data and label data of the sample data; the feature data includes the correlation features between time-series sample data and the correlation features between each server.

[0151] The feature data and tag data are added to the corresponding historical performance data details table to obtain the historical performance data feature table for each server; this facilitates viewing and comparison in a table format.

[0152] Based on the data in the historical performance data feature table of each server, a server performance prediction model is constructed and trained; this facilitates performance prediction using the server performance prediction model.

[0153] Obtain and parse the server status detection command, and obtain the server performance data from the current time point to the preset time interval, as a performance data detection set;

[0154] The performance data in the performance data detection set is input into the server performance prediction model in chronological order to obtain several corresponding initial performance prediction values; this facilitates the initial determination of the corresponding performance prediction values.

[0155] Calculate and obtain the residual dataset corresponding to the historical performance data of the server, and input the residual dataset into a preset correction model to obtain correction parameters; this facilitates the improvement of the accuracy of server performance prediction.

[0156] Based on the correction parameters, several initial performance prediction values ​​are corrected to obtain several final performance prediction values.

[0157] Based on a pre-trained GM-HMM model, state labels are added to the several performance prediction final values ​​to obtain a final value label sequence; the GM-HMM model is a Gaussian Mixture Hidden Markov Model, used to find the patterns of server performance changes; the state labels include normal, idle, abnormal, etc.

[0158] Based on a preset prediction time interval, the final value identifier sequence is segmented into several prediction segments, which helps to improve the accuracy of short-term change prediction.

[0159] Several predicted sequences are used as target predicted sequences, and a preset DP algorithm and a basic sequence database are used to obtain the basic sequence with the highest similarity to the target predicted sequence; the DP algorithm is a dynamic programming algorithm; the basic sequence is the basic sequence in the basic sequence database that has the longest identical sequence with the target predicted sequence.

[0160] Obtain the performance trend of the basic sequence segment, and use the performance trend as the server's performance prediction data at the current time node;

[0161] The performance prediction data and the performance data at the current time point are used as the server's status data.

[0162] The beneficial effects of the above technical solution are as follows: By acquiring historical performance data from each server, future performance can be predicted based on the patterns in these historical performance data. Specifically, constructing detailed historical performance data tables and feature tables facilitates a clearer and more concise view of trends. While sample data allows for rapid and effective training of the server performance prediction model, the limited sample data means the results still contain significant errors. Therefore, correcting the initial performance prediction values ​​reduces these errors, resulting in a final performance prediction value with a smaller error, ultimately leading to more reliable performance prediction data. Predicting server performance significantly improves the success rate of data flow access to the server, ensuring stable game operation.

[0163] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for accelerating dynamic data stream determination, characterized in that, include: Acquire game data generated after players enter the game, and use it as game data to be processed; The game data to be processed is de-identified to obtain the first processed game data; Obtain the working status of each data processing node and determine the set of available data processing nodes; Based on the data type of the first processed game data and the preset processing method corresponding to the data type, a target data processing node set and the working order of the target data processing nodes in the target data processing node set are determined from the available data processing node set. A data processing channel is constructed based on the target data processing node set and the working order of the target data processing nodes; The first processed game data is input into the data processing channel for processing to obtain the second processed game data and acquire the corresponding data stream. Acquire several data streams corresponding to the game data generated after several players enter the game and input them into the traffic filter; The flow filter determines the combined classification attributes of each data stream based on a preset dynamic judgment standard for data streams, and classifies several data streams according to the combined classification attributes to obtain several data stream sets; Real-time monitoring and acquisition of status data from each server; The acquisition of status data from each server includes: Obtain historical status data for each server and obtain the corresponding historical performance data sequence set; Based on the aforementioned historical performance data sequence set, construct the corresponding historical performance data detail table; Based on historical performance data details, a server performance prediction model was built and trained. Obtain and parse the server status detection command, and obtain the server performance data from the current time point to the preset time interval, as a performance data detection set; Based on the server performance prediction model and the performance data detection set, several initial performance prediction values ​​are obtained; these initial performance prediction values ​​are then corrected to obtain several final performance prediction values. Based on the pre-trained GM-HMM model, state labels are added to the several performance prediction final values ​​to obtain a final value label sequence; the GM-HMM model is a Gaussian Mixture Hidden Markov Model. Based on a preset prediction time interval, the final value identifier sequence is segmented into several prediction segments. The predicted sequences are used as target predicted sequences, and a preset DP algorithm and a basic sequence database are used to obtain the basic sequence with the highest similarity to the target predicted sequence; the DP algorithm is a dynamic programming algorithm. Obtain the performance trend of the basic sequence segment, and use the performance trend as the server's performance prediction data at the current time node; The performance prediction data and the performance data at the current time point are used as the server's status data. Based on the state data, the probability value that the current data stream set can be connected to the corresponding default server is calculated; When the probability value is lower than a preset probability threshold, the current data stream set will be switched to another server corresponding to the current data stream set. Based on a preset data stream set-server mapping table, the corresponding server is determined, and the several data stream sets are connected to the corresponding server; The dynamic judgment criteria for the data stream include: the location of the target address, the port of the target address, the target domain name, and the protocol.

2. The method for accelerating dynamic data stream determination as described in claim 1, characterized in that, The game data includes: player game data, game object status data, scene and environment data, social data, historical data, and future data.

3. The method for accelerating dynamic data stream determination as described in claim 1, characterized in that, The flow filter, based on a preset dynamic judgment standard for data flows, determines the combined classification attributes of each data flow, including: Based on the location of the target address, the first classification attribute of the data stream is obtained; Based on the target address port, the second classification attribute of the data stream is obtained; Based on the target domain name, the third classification attribute of the data stream is obtained; Based on the protocol, the fourth classification attribute of the data stream; The first classification attribute, the second classification attribute, the third classification attribute, and the fourth classification attribute are combined to obtain a combined classification attribute.

4. The method for accelerating dynamic data stream determination as described in claim 1, characterized in that, Before connecting the aforementioned data stream sets to the corresponding servers, the process also includes: Based on preset server synchronization rules, the servers are filtered to determine the servers that can be synchronized, which are then used as the target server set, and the corresponding data stream sets are synchronized and connected.

5. The method for accelerating dynamic data stream determination as described in claim 1, characterized in that, Also includes: The system detects and displays the status information of the data stream collection connected to the corresponding server.

6. The method for accelerating dynamic data stream determination as described in claim 5, characterized in that, The status information includes connection success, connection failure, and poor connection status.

Images