CN116764235A

CN116764235A - Data processing method and related device

Info

Publication number: CN116764235A
Application number: CN202210228826.8A
Authority: CN
Inventors: 邵梦超; 许敏华
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2022-03-08
Filing date: 2022-03-08
Publication date: 2023-09-19

Abstract

The application provides a data processing method and a related device, which can be applied to a cloud game scheduling scene. The method comprises the following steps: when a device number prediction request of a target game is received, acquiring the number of reference objects of the target game in a target cluster, wherein the number of reference objects corresponds to a first time point; acquiring a first object number of a target game in a target cluster at a first time point in a historical time period; acquiring a second object number of the target game in the target cluster at a second time point in the history time period; predicting a second predicted number of objects of the target game in the target cluster at a second point in time according to the first number of objects, the second number of objects and the reference number of objects; and adjusting the number of devices for pre-starting the target game in the target cluster at a second time point according to the second predicted object number. The application can improve the accuracy of the number of the devices for pre-starting the target game, thereby improving the multiplexing rate of the devices.

Description

Data processing method and related device

Technical Field

The present application relates to the field of computer technology, and in particular, to a data processing method, a data processing apparatus, a computer device, a computer readable storage medium, and a computer program product.

Background

In a cloud game scenario, a real-time pull-up scheme and a pre-installation scheme are typically employed to run the cloud game. Practice shows that the real-time pull-up scheme has longer starting time and needs longer waiting time to see the game picture. The preassembly scheme mainly installs the game in the virtual machine in advance and starts the game according to the use condition of the game, and the preassembly starting mode causes resource waste if the number of the started devices is too large, and objects need to wait in a queue if the number of the started devices is too small. Therefore, how to accurately predict the number of devices to be started in a cloud game is a technical problem to be solved currently.

Disclosure of Invention

The embodiment of the application provides a data processing method, a data processing device, computer equipment, a computer readable storage medium and a computer program product, which can improve the accuracy of the number of devices for pre-starting a target game, thereby improving the multiplexing rate of the devices.

In one aspect, an embodiment of the present application provides a data processing method, where the data processing method includes:

when a device number prediction request of a target game is received, acquiring the number of reference objects of the target game in a target cluster, wherein the number of reference objects corresponds to a first time point;

Acquiring a first object number of a target game in a target cluster at a first time point in a historical time period;

acquiring a second object number of the target game in the target cluster at a second time point in the history time period;

predicting a second predicted number of objects of the target game in the target cluster at a second point in time according to the first number of objects, the second number of objects and the reference number of objects;

and adjusting the number of devices for pre-starting the target game in the target cluster at a second time point according to the second predicted object number.

In one aspect, an embodiment of the present application provides a data processing apparatus, including:

the acquisition unit is used for acquiring the number of reference objects of the target game in the target cluster when receiving the equipment number prediction request of the target game, wherein the number of the reference objects corresponds to a first time point;

an acquisition unit, configured to acquire a first number of objects of a target game in a target cluster at a first time point in a history period;

an acquisition unit, configured to acquire a second number of objects of the target game in the target cluster at a second time point in the history period;

a processing unit for predicting a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects and the reference number of objects;

The processing unit is further used for adjusting the number of devices for pre-starting the target game in the target cluster at the second time point according to the second predicted object number.

In one possible implementation manner, the obtaining unit obtains the number of reference objects of the target game in the target cluster, and is used for performing the following operations:

acquiring N contribution degrees of N associated objects associated with the target game to N target object numbers of the target cluster at a first time point, wherein the associated objects are objects which enter the target game or are objects which wait to enter the target game in a queuing manner at the first time point, and N is a positive integer;

and superposing the contribution degrees of the N target object quantities as the reference object quantities.

In one possible implementation, the target cluster is any one of K clusters, K is a positive integer, and the target associated object is any one of N associated objects;

the acquisition unit acquires a process of determining the contribution degree of the target associated object to the number of target objects of the target cluster at a first time point, and the process is used for executing the following operations:

obtaining K round trip delays between a client side where a target associated object is located and K clusters at a first time point;

and filtering the K round trip delays, and generating the contribution degree of the target associated object to the number of target objects of the target cluster according to the filtered round trip delays.

In one possible implementation, the historical time period includes P time periods, each time period including a first point in time, P being a positive integer;

the acquisition unit acquires a first object number of the target game in the target cluster at a first time point in the history period, for performing the following operations:

acquiring the number of first unit objects of a target game in a target cluster at a first time point in any time period;

and determining the first object number of the target game in the target cluster according to the obtained P first unit object numbers.

In one possible implementation, the target cluster is any one of the K clusters;

the acquisition unit acquires a first unit object number of the target game in the target cluster at a first time point in any time period, and is used for executing the following operations:

acquiring contribution degrees of M history-related objects associated with a target game to M history object numbers of a target cluster at a first time point in any time period, wherein the history-related objects are objects which enter the target game at the first time point in any time period or are objects which are queued to enter the target game, and M is an integer larger than 1;

And superposing the contribution degrees of the number M of the historical objects as the number of the first unit objects.

In one possible implementation, the processing unit adjusts, according to the second number of predicted objects, the number of devices in the target cluster that pre-start the target game at the second point in time, to perform the following operations:

acquiring the number of reference devices for pre-starting a target game in a target cluster at a first time point;

determining the number of target devices according to the number of second predicted objects and the number of reference devices;

if the number of the target devices is positive, adding devices for pre-starting the target game in the target cluster at a second time point;

if the number of target devices is negative, reducing the devices for pre-starting the target game in the target cluster at a second time point, wherein the number of newly added devices or the number of reduced devices is equal to the absolute value of the number of target devices.

In one possible implementation manner, the processing unit determines the number of target devices according to the second number of predicted objects and the number of reference devices, and is configured to perform the following operations:

acquiring a first predicted object number of a target game in a target cluster at a first time point;

determining a corresponding error rate at a first point in time according to the first number of predicted objects and the reference number of objects;

And determining the number of target devices according to the error rate, the number of second predicted objects, the number of reference devices and the number of reference objects.

In one possible implementation, the processing unit determines the number of target devices according to the error rate, the number of second predicted objects, the number of reference devices, and the number of reference objects, and is configured to perform the following operations:

if the error rate is greater than or equal to the preset error threshold, determining the number of target devices according to the number of second predicted objects, the number of reference objects and the number of reference devices;

and if the error rate is smaller than the preset error threshold value, determining the number of target devices according to the error rate, the number of second predicted objects, the number of reference objects and the number of reference devices.

In one possible implementation, the processing unit determines the target device number according to the second predicted object number, the reference object number, and the reference device number, for performing the following operations:

acquiring a first difference value between the second predicted object number and the reference device number, and acquiring a second difference value between the reference object number and the reference device number;

the maximum value between the first difference and the second difference is determined as the target device number.

In one possible implementation, the processing unit determines the target device number according to the error rate, the second predicted object number, the reference object number, and the reference device number, for performing the following operations:

obtaining a product between the error rate and the number of reference objects, and obtaining a third difference value between the second predicted object number, the number of reference devices and the product;

acquiring a second difference between the number of reference objects and the number of reference devices;

and determining the maximum value between the third difference value and the second difference value as the device adjustment quantity corresponding to the target cluster.

In one possible implementation, the processing unit predicts a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects, and the reference number of objects, for performing the following operations:

acquiring a ratio between the second object number and the first object number, and determining a product between the ratio and the reference object number as a second predicted object number; or alternatively, the process may be performed,

and determining object characteristics according to the first object quantity, the second object quantity and the reference object quantity, and calling a deep learning model to identify the object characteristics so as to obtain the second predicted object quantity.

In one possible implementation, the device in the target cluster that pre-starts the target game includes one or more of: virtual machine, development board, container.

In one possible implementation, the processing unit is further configured to perform the following operations:

when a starting request of a target object for a target game is received, distributing target equipment for the target object according to the equipment performance of equipment for pre-starting the target game in a target cluster; or alternatively, the process may be performed,

when a starting request of a target object for a target game is received, distributing target equipment for the target object according to the historical starting times of equipment for pre-starting the target game in a target cluster;

the initiation request is forwarded to the target device to cause the target device to respond to the initiation request.

In one aspect, an embodiment of the present application provides a computer apparatus, where the computer apparatus includes a memory and a processor, and the memory stores a computer program, and when the computer program is executed by the processor, causes the processor to execute the data processing method described above.

In one aspect, embodiments of the present application provide a computer-readable storage medium storing a computer program that, when read and executed by a processor of a computer device, causes the computer device to perform the above-described data processing method.

In one aspect, embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the data processing method described above.

In the embodiment of the application, when the equipment number prediction request of the target game is received, the reference object number of the target game in the target cluster can be obtained, and the reference object number corresponds to a first time point; acquiring the first object number of the target game in the target cluster at a first time point in the historical time period; and obtaining a second number of objects of the target game in the target cluster at a second point in time within the historical time period. Then, a second predicted number of objects of the target game in the target cluster at a second point in time may be predicted based on the first number of objects, the second number of objects, and the reference number of objects. Finally, the number of devices for pre-starting the target game in the target cluster at the second time point is adjusted according to the second predicted object number. It follows that the corresponding second predicted object number at the second time point may be determined according to the first object number at the first time point in the history period, the second object number at the second time point in the history period, and the reference object number at the first time point where the current is located, and the predicted second predicted object number is more accurate since both the history data and the current data are referenced. Further, the number of devices for pre-starting the target game in the target cluster is adjusted based on the second predicted object number with higher accuracy, so that the accuracy of the number of devices for pre-starting the target game can be improved, and the device multiplexing rate in the target cluster is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1a is a schematic diagram of a login scenario for a cloud game according to an embodiment of the present application;

FIG. 1b is a graph of the number of objects in a game scene provided by an embodiment of the present application;

FIG. 1c is a graph of the number of objects in another game scenario provided by an embodiment of the present application;

FIG. 1d is a schematic flow chart of predicting the number of objects according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a data processing system according to an embodiment of the present application;

FIG. 3 is a schematic flow chart of a data processing method according to an embodiment of the present application;

FIG. 4 is a flowchart of another data processing method according to an embodiment of the present application;

FIG. 5a is a schematic diagram of a prediction curve of a target game in a target cluster according to an embodiment of the present application;

FIG. 5b is a graph illustrating an error rate provided by an embodiment of the present application;

FIG. 6 is a schematic diagram of a data processing apparatus according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application.

The embodiment of the application provides a data processing scheme which can be applied to a cloud game scheduling scene and can improve the accuracy of the number of devices for pre-starting a target game, thereby improving the device multiplexing rate in a target cluster. Next, real data collected from the cloud game scene and the cloud game scene are described in detail, respectively.

1. Cloud game scene:

first, related introduction is performed on an object login scene in a cloud game. Referring to fig. 1a, fig. 1a is a schematic diagram of a login scenario of a cloud game according to an embodiment of the present application. As shown in fig. 1a, at least one game cover of a cloud game is displayed in the page S10, for example, the cloud game may include a game 1 and a game 2, and an opening control is set in the game cover of each cloud game (for example, a second play control 101 is set in the game cover of the game 2); the target object may select any one of the cloud game play experiences, for example, game 2 may be selected as the target game. In one possible implementation manner, after the target object clicks the second play control 101 in the page S10, if the machine room equipment near the client where the target object is located is sufficient, the login page S20 of the target game may be displayed, and one or more login controls corresponding to the login modes, for example, the login mode 1 and the login mode 2, are displayed in the login page S20. After the target object triggers the login control corresponding to any login mode, a game page S30 of the target game is displayed. In another possible implementation manner, after the target object clicks the second play control 101 in the page S10, if the machine room device near the client where the target object is located is insufficient, a popup page S40 related to the target game may be displayed, and a prompt message is displayed in the popup page S40, where for example, the prompt message may be: the current game number is more, please wait for the patience, currently row 1, and expects to wait for 1 minute. Therefore, when the starting requests of a plurality of objects for the target game are detected at the same time point, the cloud game server needs to ensure that each object can be allocated to corresponding equipment to respond to the starting requests as far as possible, and the utilization rate of the current equipment can be ensured to be higher (namely, more equipment cannot exist in an idle state). Therefore, in the object login scene of the cloud game, the device multiplexing rate needs to be improved while the nearby scheduling is guaranteed.

2. Data analysis in a cloud game scene:

second, to get a clearer view of the cloud game scenario. The real data collected in the cloud game scene (e.g. the number of objects experiencing the target game) can be specifically analyzed as follows with reference to fig. 1b-1 d:

1) Referring to fig. 1b, fig. 1b is a graph illustrating the number of objects in a game scene according to an embodiment of the present application. As shown in fig. 1b, the number of subjects corresponding to each day of 11 months 8-11 months 11 is collected at four hour intervals. As can be seen from the graph shown in fig. 1a, the number of objects experiencing the target game for any day has a rule that the number of objects corresponding to the morning increases all the time, then increases to 12 pm to a small peak, starts to decrease from 12 pm, until the number of objects starts to increase again around 5 pm, until 8 pm reaches a peak, and starts to decrease from 8 pm.

2) Referring to fig. 1c, fig. 1c is a graph illustrating the number of objects in another game scene according to an embodiment of the present application. As shown in fig. 1c, the number of subjects corresponding to each day of 11 months 10-11 months 13 was also collected at four hour intervals. It will be appreciated that the days include weekdays and holidays, so it can be seen from the graph shown in fig. 1c that the number of objects of the holiday experience target game is much greater than the number of objects of the weekday experience target game, but that the 2 peak regularity as analyzed in fig. 1b is maintained for either of the holidays or the weekdays.

3) In combination with the above analysis of the number of objects experiencing the target game in the game scene, the following features can be derived:

(1) the trend of the object number in a certain time period in the future can be deduced by using the object number in the history time period, that is, a small peak at 2 pm is predicted according to the object number in the history, and a small peak exists at 2 pm in the future with a high probability.

(2) The number of objects on the weekend is greater than the number of objects on the weekday, but the curve characteristic corresponding to the weekend is approximately the same as the curve characteristic corresponding to the weekday.

(3) Referring to fig. 1d in conjunction with the above analysis, fig. 1d is a schematic flow chart of predicting the number of objects according to an embodiment of the present application. As shown in fig. 1d, the characteristics of the curve are linear. Assuming that the number of objects at time T (first time point) in the history period is M (T), the number of objects at time t+1 (second time point) in the history period is M (t+1). If the number of objects at time T is N (T), the predicted amount of object data at time t+1 may be represented by formula (1):

N(T+1)＝N(T)*M(T+1)/M(T) (1)

3. in view of the foregoing, the present application provides a data processing scheme, the general principle of which is as follows:

When a device number prediction request of a target game is received, the number of reference objects of the target game in a target cluster can be acquired, wherein the number of reference objects corresponds to a first time point; acquiring the first object number of the target game in the target cluster at a first time point in the historical time period; and obtaining a second number of objects of the target game in the target cluster at a second point in time within the historical time period. Then, a second predicted number of objects of the target game in the target cluster at a second point in time may be predicted based on the first number of objects, the second number of objects, and the reference number of objects. Finally, the number of devices for pre-starting the target game in the target cluster at the second time point is adjusted according to the second predicted object number. It follows that the corresponding second predicted object number at the second time point may be determined according to the first object number at the first time point in the history period, the second object number at the second time point in the history period, and the reference object number at the first time point where the current is located, and the predicted second predicted object number is more accurate since both the history data and the current data are referenced. Further, the number of devices for pre-starting the target game in the target cluster is adjusted based on the second predicted object number with higher accuracy, so that the accuracy of the number of devices for pre-starting the target game can be improved, and the device multiplexing rate in the target cluster is improved.

4. The above mentioned data processing schemes are described in connection with the technical terms to which the present application relates:

1) Cloud technology:

cloud gaming (Cloud gaming), which may also be referred to as game on demand, is an online gaming technology based on Cloud computing technology. Cloud gaming technology enables lightweight devices (thin clients) with relatively limited graphics processing and data computing capabilities to run high quality games. In a cloud game scene, the game is not run in a player game terminal, but is run in a cloud server, the cloud server renders the game scene into a video and audio stream, and the video and audio stream is transmitted to the player game terminal through a network. The player game terminal does not need to have strong graphic operation and data processing capability, and only needs to have basic streaming media playing capability and the capability of acquiring player input instructions and sending the player input instructions to the cloud server.

In one possible implementation, the data processing scheme provided by the application can be applied to a cloud game scene. Specifically, the method can be applied to an object login scene in a cloud game, and a page schematic diagram of the object login scene in the cloud game can be shown in fig. 1 a. Therefore, when an object logs in a game in a certain cloud game client, the cloud game server can predict the number of objects (second predicted number of objects) corresponding to the cloud game at the next moment (second time point) based on the scheme provided by the application, so that the number of devices for pre-starting the target game in the target cluster can be adjusted (for example, increased or decreased based on the current devices) based on the second predicted number of objects. By the scheme, the accuracy of the number of the devices for pre-starting the target game can be improved, so that the device multiplexing rate in the target cluster is improved.

In addition, when executing the data processing scheme of the present application, the second predicted object number of the target game in the target cluster at the second time point is predicted according to the first object number, the second object number and the reference object number, which involves larger-scale calculation and requires larger calculation power and storage space, so that in one possible implementation of the present application, the computer device can acquire enough calculation power and storage space through the cloud computing technology, and further execute the determination of the second predicted object number involved in the present application.

2) Artificial intelligence:

artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

In one possible implementation, the data processing scheme of the present application may be combined with machine learning techniques in the field of artificial intelligence. For example, a second predicted number of objects in the target cluster for the target game at a second point in time may be predicted using machine learning techniques. Among them, machine Learning (ML) is a multi-domain interdisciplinary, and involves multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, confidence networks, reinforcement learning, transfer learning, induction learning, teaching learning, and the like.

3) Blockchain:

blockchain (Blockchain) networks are networks of point-to-point networks (P2P networks) and blockchains, which are a new model of application of computer technology such as distributed data storage, point-to-point transmission, consensus mechanisms, encryption algorithms, etc., and are essentially a de-centralized database, which is a string of data blocks (or called blocks) that are generated in association using cryptographic methods.

In one possible implementation, the data processing scheme of the present application may be combined with blockchain technology. For example, the number of reference objects, the number of first objects, the number of second predicted objects, and the like can be uploaded to a blockchain of the blockchain network for storage, so that internal data of the computer device is prevented from being tampered, and safety and privacy of log data are improved.

It should be noted that, in the following embodiments of the present application, data related to object information (such as nicknames, IDs of objects) and the like is required to obtain permission or consent of objects when the above embodiments of the present application are applied to specific products or technologies, and collection, use and processing of related data are required to comply with related laws and regulations and standards of related countries and regions.

With reference to fig. 2, fig. 2 is a schematic diagram of a data processing system according to an embodiment of the present application. As shown in fig. 2, the schematic structural diagram of the data processing system may include: scheduling server 240, a server cluster, and a terminal device cluster, wherein the terminal device cluster may include: terminal device 210, terminal device 220, terminal device 230, etc.; the server cluster may include: server 250, server 260, server 270, etc. Any terminal device in the terminal device cluster may be directly or indirectly connected to the scheduling server 240 through a wired or wireless communication manner, which is not limited herein; any terminal device in the terminal device cluster may be directly or indirectly connected to any server in the server cluster through a wired or wireless communication manner, which is not limited herein. In addition, the types of any two terminal devices in the terminal device cluster may be the same or different, and the application is not limited herein.

The dispatch server 240 shown in fig. 2, and any server in the server cluster may be an independent physical server, or may be a server cluster or a distributed system formed by a plurality of physical servers, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs (Content Delivery Network, content delivery networks), and basic cloud computing services such as big data and artificial intelligence platforms, and so on.

The terminal devices 210, 220, 230 shown in fig. 2 may include, but are not limited to: a mobile phone, a tablet computer, a notebook computer, a palm computer, a mobile internet device (MID, mobile internet device), an intelligent voice interaction device, an on-board terminal, a roadside device, an aircraft, a wearable device, an intelligent home appliance, or a wearable device with a data processing function such as a smart watch, a smart bracelet, a pedometer, or the like.

In one possible implementation, taking the terminal device 210 as an example, the terminal device 210 and the scheduling server 240 together execute the data processing scheme in the present application. When the terminal device 210 receives a device number prediction request of the target game, the number of reference objects of the target game in the target cluster may be acquired, where the number of reference objects corresponds to a first time point; then, the terminal device 210 may acquire a first number of objects of the target game in the target cluster at a first point in time within the history period; and the terminal device 210 may also obtain a second number of objects of the target game in the target cluster at a second point in time within the historical period of time. Next, the terminal device 210 may transmit the first object number, the second object number, and the reference object number to the scheduling server 240. The scheduling server 240 may predict a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects, and the reference number of objects.

Subsequently, the scheduling server 240 may send the second predicted object number to the terminal device 210, and the terminal device 210 may adjust the number of devices in the target cluster (e.g., the server 250 closest to the terminal device 210) that pre-start the target game at the second point in time according to the second predicted object number.

It should be understood that the foregoing is merely illustrative of various steps that the terminal device 210 and the scheduling server 240 are responsible for performing, and embodiments of the present application are not limited thereto. For example, in another possible implementation manner, when the terminal device 210 receives a game request submitted by an object, the terminal device 210 sends a device number prediction request of the target game to the scheduling server 240, and when the scheduling server 240 receives the device number prediction request of the target game, the reference object number of the target game in the target cluster may be acquired, where the reference object number corresponds to a first time point; then, the scheduling server 240 may obtain a first number of objects of the target game in the target cluster at a first point in time within the historical period of time; and the dispatch server 240 may also obtain a second number of objects of the target game in the target cluster at a second point in time over the historical period of time. Next, the scheduling server 240 may predict a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects, and the reference number of objects. Subsequently, the scheduling server 240 may send the second predicted object number to the terminal device 210, and the terminal device 210 may adjust the number of devices in the target cluster (e.g., the server 250 closest to the terminal device 210) that pre-start the target game at the second point in time according to the second predicted object number.

Further, the data processing system provided in fig. 2 may be deployed at a node of a blockchain, for example, the terminal device 210, the terminal device 220, and the scheduling server 240 may all be regarded as node devices of the blockchain, to jointly form a blockchain network, specifically, the terminal device 210, the terminal device 220, etc. may be regarded as working nodes in the blockchain, and the scheduling server 240 may be regarded as a management node of the blockchain. Therefore, the data processing flow related in the application can be executed on the blockchain, so that the fairness and fairness of the data processing flow can be ensured, and the data processing flow can be provided with traceability, thereby improving the safety of the data processing flow.

It may be understood that the schematic diagram of the system architecture described in the embodiment of the present application is for more clearly describing the technical solution of the embodiment of the present application, and does not constitute a limitation on the technical solution provided by the embodiment of the present application, and those skilled in the art can know that, with the evolution of the system architecture and the appearance of a new service scenario, the technical solution provided by the embodiment of the present application is equally applicable to similar technical problems.

Based on the above description of the data processing scheme and the data processing system, the embodiment of the application provides a data processing method. Referring to fig. 3, fig. 3 is a schematic flow chart of a data processing method according to an embodiment of the present application, where the data processing method may be executed by the above-mentioned terminal device or the scheduling server, and for convenience of explanation, the data processing method is hereinafter described by taking a computer device as an example. The data processing method may include the following steps S301 to S305:

S301: when a device number prediction request of the target game is received, the number of reference objects of the target game in the target cluster is acquired, and the number of reference objects corresponds to a first time point.

In the embodiment of the application, when an object (such as an operation and maintenance personnel) requests to start a target game, the generation of a device quantity prediction request about the target game can be triggered; alternatively, the computer device may generate a device number prediction request for the target game (e.g., 10:00 am a day) at regular time, which may be a cloud game. In one possible implementation manner, the device number prediction request carries an identifier of a client that requests to start the target game, when the computer device receives the device number prediction request, the client that requests to start the target game may be checked according to the identifier of the client, for example, security check, validity check, and the like, and if the check is successful, the step of acquiring the number of reference objects of the target game in the target cluster is triggered to be executed. In this way, the security of the data processing process can be improved.

By reference object number is meant the sum of the number of objects that start or run the target game at the first point in time. The first time point may include a time point corresponding to the current time, for example, the current time may be 8 a.m. 1 month and 1 day: 00. it should be noted that, a cluster may be considered as a group formed by a plurality of computer devices (e.g., servers), and thus, in the subsequent embodiments of the present application, the cluster may also be referred to as a machine room.

In the embodiment of the application, the following cases are considered:

1) In general, when an object starts a target game, the object is usually dispatched to a nearest machine room according to the location of the object. Due to errors caused by predictions or insufficient equipment resources, an object may be dispatched to a room that is not closest to the object. For example, at a certain time point, the object A located in the Shenzhen machine room should be dispatched to the Shenzhen machine room when the target game is started, but the Shenzhen machine room has enough equipment at the moment due to insufficient equipment of the Shenzhen machine room, so as to avoid waiting of the object, and the object can be dispatched to the Shenzhen machine room at the moment. From the above, if the number of objects in the long sand machine room is simply counted, it is considered that the objects should be dispatched to the long sand machine room instead of the Shenzhen machine room in the future, and such prediction and assumption are not reasonable.

2) When the equipment of the machine room where the object is nearby is insufficient, the object may be unwilling to wait and leave directly. Therefore, if the number of the historical objects is simply counted, the historical objects may be missed, so that the prediction result is inaccurate.

In summary, in the embodiment of the present application, the number of objects of each game in each cluster at each time point in the historical time period needs to be counted. Next, the embodiment of the present application may take "obtaining the number of reference objects of the target game in the target cluster" as an example, and perform a related description.

In one possible implementation, the computer device obtaining the number of reference objects of the target game in the target cluster may include: firstly, acquiring N target object quantity contribution degrees of N associated objects associated with a target game on a target cluster at a first time point, wherein the associated objects are objects which enter the target game or are objects which wait to enter the target game in a queuing way at the first time point, and N is an integer larger than 1; then, the N target object quantity contribution degrees are superimposed as the reference object quantity.

In one possible implementation, the target cluster is any one of K clusters, K is a positive integer, and the target associated object is any one of N associated objects. The process of the computer device obtaining the contribution degree of the number of the target objects of the target associated objects on the target cluster at the first time point can comprise: firstly, at a first time point, the computer equipment acquires K round trip delays between a client where a target associated object is located and K clusters, where the round trip delays are used to represent a time difference between a sending time corresponding to sending data (i.e., a test packet of a target game) from a sending end (i.e., the client where the target associated object is located) and a receiving time when the sending end (i.e., the client where the target associated object is located) receives an acknowledgement message from a receiving end (a machine room), where the receiving end can be considered to send the acknowledgement message immediately after receiving the data. And then, the computer equipment filters the K round trip delays, and generates the contribution degree of the target associated object to the number of the target objects of each cluster according to the filtered round trip delays. Specifically, when the computer device filters K round trip delays, the method specifically may include: the computer device compares each of the K round trip delays with a round trip delay threshold, and filters the round trip delay greater than or equal to the round trip delay threshold, where filtering means that deleting is performed on the round trip delay greater than or equal to the round trip delay threshold, and the part of the round trip delay that has been deleted is not included in the process of calculating the contribution of the number of target objects.

It will be appreciated that a velocimetry procedure is required before the apparatus is allocated to the subject. The speed measurement means that a client side where an object is located sends a test packet of a target game to each machine room, and each machine room responds after receiving the test packet. The client side where the object is located can judge whether the current client side corresponds to the network situation or not by calculating round trip time RTT between the sending time (i.e. the sending time and the receiving time) of the test packet of the target game (i.e. the time of sending the test packet of the target game from the client side where the target associated object is located) and the receiving time (i.e. the time of receiving the acknowledgement message of the machine room by the client side where the target associated object is located) of the response. Generally, the lower the RTT, the better, generally the RTT will be, and generally the experience will be poor beyond 50ms, and allocation to the room is not recommended. Next, a detailed description will be given of a process in which the computer device determines the target object number contribution degree of the target associated object on the target cluster according to the Round Trip Time (RTT):

assuming that the speed from the client where the object a (target associated object) is located to the cluster n is v (n), the transmission distance between the object a and the cluster n is shown in formula (2):

Distance(n)＝v(n)*RTT(n) (2)

it will be appreciated that, in terms of the near scheduling principle, the closer an object a is to a cluster n, the higher the probability/likelihood (i.e., the number of objects contributing) that the object a is scheduled to the cluster n; if object a is farther from cluster n, the probability/likelihood that object a is scheduled to that cluster n (i.e., the target object number contribution) is lower. That is, in the embodiment of the present application, the distance and the contribution degree of the number of target objects may be considered as an inversely proportional relationship.

Thus, from equation (2), equation (3) can be derived:

Weight(n)/Weight(m)＝Distance(m)/Distance(n)＝(v(m)*RTT(m))/(v(n)*RTT(n)) (3)

in equation (3), weight (n) refers to the target object number contribution of object a to be scheduled to cluster n, and Weight (m) refers to the target object number contribution of object a to be scheduled to cluster m.

Next, assuming that the speeds of the objects to the respective clusters are the same, that is, v (m) =v (n), equation (4) can be derived:

Weight(n)/Weight(m)＝RTT(m)/RTT(n) (4)

because each object can only contribute one object quantity, the sum of the object quantity contribution degrees of the object on each cluster is 1, and the formula (5) is obtained:

Weight(1)+Weight(2)+…+Weight(N)＝1 (5)

thus, for K clusters, the target object number contribution of the target associated object on the target cluster (e.g., cluster n) may be expressed as equation (6):

Weight(n)＝1/RTT(n)/(1/RTT(1)+1/RTT(2)+…+1/RTT(K)) (6)

for example, the object a of Shenzhen sends test packets of the target game to the machine rooms of Shenzhen, guangzhou, changsha, shanghai, beijing and the like through the speed measurement flow, and RTTs obtained through the flow are 10ms, 15ms, 30ms, 40ms and 70ms respectively. Since Beijing cluster RTT >50ms, this RTT can be filtered out. The contribution degree of the computing object A to the target object number of the Shenzhen cluster, guangzhou cluster, changsha cluster and Shanghai cluster is shown in the following table 1.

TABLE 1 degree of contribution of target associated objects to the number of target objects for each Cluster

Cluster	Contribution degree of number of target objects
		Weight (Shenzhen)	1/10/(1/10+1/15+1/30+1/70)＝0.46
Weight (Guangzhou)	1/15/(1/10+1/15+1/30+1/70)＝0.31
		Weight (Changsha)	1/30/(1/10+1/15+1/30+1/70)＝0.15
Weight (Shanghai)	1/40/(1/10+1/15+1/30+1/70)＝0.12
		Weight (Beijing)	0

According to the degree of contribution of the target associated objects to the number of target objects contributed by each cluster shown in table 1, it can be understood that there are the following cases:

(1) assuming that the target associated object was scheduled to a long sand cluster for 10 minutes in the period of 8 points 10 minutes to 8 points 20 minutes, it can be recorded that the target associated object contributed 0.46 objects to the Shenzhen cluster, 0.31 objects to the Guangzhou cluster, 0.15 objects to the long sand cluster, and 0.12 objects to the Shanghai cluster in the period of 8 points 10 minutes to 8 points 20 minutes.

(2) Assume that the target association object is in the period of 9 to 9 minutes, 5 minutes, because the device is under-allocated to wait 5 minutes before leaving unused. Then it can be recorded in 9-5 that the target associated object contributed 0.46 objects to the Shenzhen cluster, 0.31 objects to the Guangzhou cluster, 0.15 objects to the Changsha cluster, and 0.12 objects to the Shanghai cluster.

(3) Assume that the target associated object is in the period of 9 at 10 minutes and 9 at 25 minutes because the device is under-allocated to wait 5 minutes before using the Shanghai cluster for 10 minutes. Then it can be recorded that 9 points 10 are divided into 9 points 25, the target associated object contributes 0.46 objects to the Shenzhen cluster, 0.31 objects to the Guangzhou cluster, 0.15 objects to the Changsha cluster, and 0.12 objects to the Shanghai cluster.

In combination with the above analysis, for any cluster, the corresponding number of objects can be sampled in the dimension per minute, and thus the reference number of objects F (D, T) of the target game in the target cluster can be obtained. Where D represents day D (e.g., day 1 month 1), and T represents the first time point (e.g., 8:00) within day D. For example, the number of objects at 10 points 21 on 11/25/1/2021/certain game cluster may be specifically denoted as F (2021-11-25, 10:21).

For example, the contribution degree of each associated object in the N associated objects to the target object number of each cluster in the K clusters can be determined in the above manner. It may be understood that determining the contribution degree of any associated object to the number of target objects of each of the K clusters may refer to determining the contribution degree of the target associated object to the number of target objects of each cluster, which is not described herein. Thus, the contribution degree of the N target object numbers of the N associated objects on each cluster can be obtained, as shown in table 2.

TABLE 2 degree of contribution of N associated objects to the target object quantity of K clusters

The contribution degree of each associated object to the number of target objects of each cluster can be obtained according to the table 2. For example, S11 may represent the contribution of the associated object 1 to the target object number of the cluster 1, and S12 may represent the contribution of the associated object 2 to the target object number of the cluster 1, and so on, SKN may represent the contribution of the associated object N to the target object number of the cluster K. Then, as can be seen from table 2, the number of reference objects in the target cluster of the target game may be the sum of the contribution degrees of the respective associated objects to the number of target objects of the target cluster, for example, the target cluster is cluster 1, and then the number of reference objects in the target cluster 1 may be: s11+s12+ & gt, s1n.

It can be understood that, since the contribution degree of each target object number refers to a number between 0 and 1, after the N target object number contribution degrees are superimposed, the reference object number can be obtained after the rounding process. The rounding process may include rounding up or rounding down, where rounding up refers to adding 1 to a fraction of the preceding integer, for example, the number of reference objects is "100.3", and rounding up "100.3" to obtain 101, regardless of the rounding rule. By rounding down is meant that, regardless of the rounding rule, if a fraction is followed by an ignoring fraction, for example the number of reference objects is "100.3", then "100.3" is rounded down to get 100. In this way, compared with the traditional method that the number of the objects is 0 or 1, the method and the device can calculate the contribution degree of the number of the target objects through round trip delay, and then determine the number of the reference objects of the target cluster according to the contribution degree of the number of the target objects, so that the determined number of the reference objects is more accurate.

It should be noted that, in the embodiment of the present application, the contribution degree of the number of the target objects is related to time, that is, the contribution degree of the number of the target objects may or may not change with the change of time. For example, at a first time point, the contribution degree of the number of target objects of the target associated objects on the target cluster, which is acquired by the computer equipment, is x1; if the target associated object is still experiencing the target game and is not exited at the second time point, the contribution degree of the number of target objects, on the target cluster, of the target associated object acquired by the computer equipment is still x1 at the second time point. For another example, at a first time point, the contribution degree of the number of target objects of the target associated objects on the target cluster, which is acquired by the computer equipment, is x1; if the target associated object exits the target game at the second time point, the contribution degree of the number of target objects on the target cluster, obtained by the computer device, of the target associated object needs to be calculated again according to the above-mentioned speed measurement flow, and at this time, the determined round trip time RTT may be different due to different geographic positions of the target object, so that the contribution degree of the number of target objects calculated at the second time point may also be different.

It will be appreciated that if the target associated object does not exit the target game, then the geographic location where the target associated object is highly probable is unchanged, and the contribution of the number of target associated objects remains unchanged. In this case, the round trip delay is not recalculated, so that network resources can be saved.

If the target associated object exits from and reenters the target game in the time interval of the first time point and the second time point, the geographic position of the target associated object is considered to be possibly changed, so that the contribution degree of the number of the objects at the second time point needs to be recalculated, and the influence caused by the address position of the target associated object can be considered, so that the calculated contribution degree of the number of the target objects is more accurate.

S302: a first number of objects of the target game in the target cluster at a first point in time within the historical time period is obtained.

In one possible implementation, the historical time period may include P time periods, each time period including a first point in time, P being a positive integer. The computer device obtaining a first number of objects of the target game in the target cluster at a first point in time over a historical period of time may include: first, the computer device obtains a first number of unit objects of the target game in the target cluster at a first point in time within any period of time. Then, the computer device determines the first object number of the target game in the target cluster according to the acquired P first unit object numbers.

Specifically, as can be seen from the analysis in step S301 described above, the history period can be set to the past one week. The reason why the number of subjects of the nearest 7 days was adopted is as follows:

1) The more recent the day the more accurate the historical data is, generally the more regular the number of objects increases or decreases. The number of objects in the target cluster for yesterday target games increases, and today the number of objects in the target cluster for target games is not more probable than yesterday.

2) The habit of the object to experience the target game takes 7 days of the week as a time period, so that the number of objects in 7 days of the sampling history can be ensured to cover a complete time period.

Then, if the historical time period is one week, the historical time period may include 7 days, one day is a time period, and each day may include the first time point. The first time point here refers to a specific time of day, for example 8 am: 00, 5 pm: 00, etc.

In one possible implementation, the target cluster is any one of the K clusters. The computer device obtaining a first number of unit objects of the target game in the target cluster at a first point in time within any time period may include: first, the contribution degree of the number of M historical objects on the target cluster of M historical associated objects associated with the target game at a first time point in any time period is acquired, wherein the historical associated objects are objects which enter the target game at the first time point in any time period or are objects which are queued to enter the target game, and M is an integer greater than 1. And then superposing the contribution degrees of the number of M historical objects into the number of first unit objects.

In one possible implementation, the target correlation object is any one of the M history correlation objects. A process for a computer device to obtain a number of contribution of M historical objects on a target cluster of M historical associated objects associated with a target game at a first point in time during any time period may include: firstly, K round trip delays between a client where a target history associated object is located and K clusters at a first time point in any time period are obtained. And then, filtering the K round trip delays, and generating the contribution degree of the target history associated object to the number of the history objects of each cluster according to the filtered round trip delays. The filtering here refers to a process of performing a deletion process, that is, performing a deletion process on a round trip delay greater than or equal to a round trip delay threshold, without incorporating the round trip delay that has been subjected to the deletion process into the calculation of the contribution of the number of history objects.

It should be noted that, the first time point in the history period corresponding to the number of reference objects mentioned in step S301 means that a specific time in a day is aligned, for example, the first time point corresponding to the number of reference objects may specifically refer to a first time point (for example, 8:00 a.m.) in a reference time period (for example, in the day, specifically, 1 month and 10 days). The number of objects of each first unit refers to the number of objects corresponding to a first time point (e.g. 8:00 am) in any one of 7 days (e.g. the last 1 day, and may be 1 month and 9 days) based on the current day.

It can be understood that, for details of the process of the computer device "obtaining the first unit object number of the target game in the target cluster" can be specifically referred to the process of the computer device "obtaining the reference object number of the target game in the target cluster" in the step S301, and the embodiments of the present application are not described herein.

In one possible implementation manner, the determining, by the computer device, the first object number of the target game in the target cluster according to the acquired P first unit object numbers may include: the computer equipment carries out weighted average operation on the number of the P first unit objects to obtain the number of the first objects. The computer device determining, according to the obtained P first unit object numbers, the first object numbers of the target game in the target cluster, and may further include: the computer equipment takes the largest first unit number in the P first unit object numbers as the first object number; alternatively, the computer device takes a smallest first unit number among the P first unit object numbers as the first object number.

For example, the number of reference objects for game N at D day T on M clusters may be expressed as: f (D, T); the 7 first unit object numbers included in the first object number may be expressed as: f (D-7, T), F (D-6, T), F (D-5, T), F (D-4, T), F (D-3, T), F (D-2, T), F (D-1, T). Then, the first number of objects of the target game in the target cluster may be denoted as f_history (T), where the f_history (T) may be calculated as shown in equation (7):

F_history(T)＝(F(D-7,T)+F(D-6,T)+…+F(D-1,T))/7 (7)

S303: a second number of objects of the target game in the target cluster at a second point in time within the historical time period is obtained.

In one possible implementation, the historical time period may include P time periods, each time period including the second point in time, P being a positive integer. The computer device obtaining a second number of objects of the target game in the target cluster at a second point in time within the historical period of time may include: first, the computer device obtains a second number of unit objects of the target game in the target cluster at a second point in time within any one time period. Then, the computer device determines the second object number of the target game in the target cluster according to the acquired P second unit object numbers.

In one possible implementation, the target cluster is any one of the K clusters. The computer device obtaining a second number of unit objects of the target game in the target cluster at a second point in time within any time period may include: first, the contribution degree of the number of the Q historical object numbers of the Q historical association objects which are associated with the target game on the target cluster at a second time point in any time period is acquired, wherein the historical association objects are objects which enter the target game at the second time point in any time period or are objects which are queued to enter the target game, and Q is an integer larger than 1. And then the Q historical object quantity contribution degrees are superimposed into a second unit object quantity.

It should be noted that the second time point refers to a specific time point located after the first time point in the day. For example, the first time point is 8 am: 00, the second point in time may be 8 a.m.: 01. then, the number of objects per second unit refers to the number of objects corresponding to the second time point (e.g. 8:01 am) in any one of the 7 days (e.g. the last 1 day, and may be 1 month and 9 days) based on the current day.

It can be understood that, the detailed process of the computer device "obtaining the second unit object number of the target game in the target cluster" can be specifically referred to the process of the computer device "obtaining the reference object number of the target game in the target cluster" in the step S301, and the embodiments of the present application are not described herein.

In one possible implementation, the determining, by the computer device, the second number of objects of the target game in the target cluster according to the acquired Q second number of unit objects may include: the computer equipment carries out weighted average operation on the Q second unit object numbers to obtain the second object numbers.

For example, the number of reference objects for game N at D day T on M clusters may be expressed as: f (D, T); the 7 second unit object numbers included in the second object number may be expressed as: f (D-7, T+1), F (D-6, T+1), F (D-5, T+1), F (D-4, T+1), F (D-3, T+1), F (D-2, T+1), F (D-1, T+1). Then, the second number of objects of the target game in the target cluster may be expressed as f_history (t+1), where f_history (t+1) may be calculated as shown in formula (8):

F_history(T+1)＝(F(D-7,T+1)+F(D-6,T+1)+…+F(D-1,T+1))/7 (8)

S304: and predicting a second predicted object number of the target game in the target cluster at a second time point according to the first object number, the second object number and the reference object number.

It can be understood that, as shown in fig. 1 b-1 c, the corresponding curve features of the number of objects experiencing the target game at different time points are approximately the same, and conform to the linear rule, so that the linear prediction model can be used to predict the second predicted number of objects at the second time point. In one possible implementation, the computer device may invoke the linear model to predict a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects, and the reference number of objects. Specifically, the computer device may first obtain a ratio between the second number of objects and the first number of objects, and then may determine a product between the ratio and the reference number of objects as the second number of predicted objects. By the method, the number of the predicted objects based on the linear prediction model can ensure the operation rule of the target game, the prediction process is more convenient and quick, and the processing efficiency is improved while the data accuracy is ensured.

For example, assume that the first object number is denoted as f_history (T), the second object number is denoted as f_history (t+1), and the reference object number is denoted as F (D, T). The second prediction object number f_expect (D, t+1) may be calculated as shown in equation (9):

F_expect(D,T+1)＝F(D,T)*(F_history(T+1)/F_history(T)) (9)

in another possible implementation, the computer device may invoke the deep learning model to predict a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects, and the reference number of objects. Specifically, the computer device may determine the object feature according to the first object number, the second object number and the reference object number, and call the deep learning model to identify the object feature, so as to obtain the two predicted object numbers.

The deep learning model may include a neural network model, among other things. The neural network model may include, for example, but is not limited to: DNN (Deep Neural Networks, deep neural network) model, LSTM (Long Short-Term Memory network) model, GRU (Gated Recurrent Neural network ) model, and the like. It should be noted that, in the embodiment of the present application, the model structure of the deep learning model is not particularly limited.

S305: and adjusting the number of devices for pre-starting the target game in the target cluster at a second time point according to the second predicted object number.

In one possible implementation, the computer device adjusts the number of devices in the target cluster that pre-start the target game at the second point in time according to the second predicted object number, and may include: firstly, the computer equipment acquires the number of reference equipment for pre-starting a target game in a target cluster at a first time point; then, the computer device determines the number of target devices according to the second number of predicted objects and the number of reference devices; finally, the computer and the device adjust the number of devices in the target cluster that pre-start the target game at the second point in time based on the number of target devices. If the number of the target devices is positive, the number of the predicted objects at the next moment is smaller than the number of the devices running the target game at the current moment, and devices for pre-starting the target game need to be added in the target cluster at a second time point so as to reduce the occurrence of queuing of the objects caused by insufficient devices; if the number of the target devices is negative, the number of the predicted objects at the next time is larger than the number of the devices running the target game at the current time, and the devices pre-starting the target game need to be reduced in the target cluster at the second time point, so that resource waste caused by excessive device starting is avoided. Wherein the number of newly added devices or the number of reduced devices is equal to the absolute value of the number of target devices.

For example, assuming that the second predicted object number is denoted as f_expect (D, t+1), and the reference device number is denoted as G (T), the target device number O (T) may be: f_expect (D, t+1) -G (T). Then, the number of devices of the pre-start O (T) target games may be increased or decreased in the target cluster at the second point in time on the basis of the number of devices of the target cluster being G (T).

In another possible implementation manner, the computer device adjusts the number of devices that pre-start the target game in the target cluster at the second time point according to the second predicted object number, and may further include: first, the computer device acquires an error rate at a first point in time; the error rate is the basis for evaluating the accuracy of the prediction algorithm, wherein the error rate is positive value, which indicates that the number of the predicted objects is higher than the number of the objects actually used, and the error rate is negative value, which indicates that the number of the predicted objects is lower than the number of the objects actually used. The smaller the absolute value of the error rate and the more toward 0 indicates the more accurate the prediction result. The error rate is generally continuous from the viewpoint of observation, and if the number of objects predicted before is high, the probability of the number of objects predicted after is also high. The number of objects predicted later is also less probable if the number of objects predicted earlier is lower. If the number of previously predicted objects is relatively accurate, then the number of later predicted objects is also relatively accurate with a high probability. Thus, determining the number of target devices by introducing an error rate may improve accuracy of the number of target devices. Then, the computer device determines the number of target devices according to the second predicted object number and the error rate; finally, the computer and the device adjust the number of devices in the target cluster that pre-start the target game at the second point in time based on the number of target devices. If the number of the target devices is positive, adding devices for pre-starting the target game in the target cluster at a second time point; if the number of target devices is negative, reducing the devices for pre-starting the target game in the target cluster at a second time point, wherein the number of newly added devices or the number of reduced devices is equal to the absolute value of the number of target devices.

In the embodiment of the application, the number of the reference objects, the number of the first objects and the number of the second objects of the target game in the target cluster can be calculated according to the round trip delay obtained by the velocity measurement flow, and compared with the traditional method that the number of the reference objects, the number of the first objects and the number of the second objects is 0 or 1, the accuracy of the determined number of the reference objects, the number of the first objects and the number of the second objects can be improved. In addition, the number of the second predicted objects corresponding to the second time point may be determined according to the number of the first objects at the first time point in the history period, the number of the second objects at the second time point in the history period, and the number of the reference objects at the first time point where the current is located, and the predicted number of the second predicted objects is more accurate since both the history data and the current data are referred to. Further, the number of devices for pre-starting the target game in the target cluster is adjusted based on the second predicted object number with higher accuracy, so that the accuracy of the number of devices for pre-starting the target game can be improved, and the device multiplexing rate in the target cluster is improved.

Referring to fig. 4, fig. 4 is a flowchart illustrating another data processing method according to an embodiment of the application. The data processing method may be performed by the above-mentioned terminal device or the scheduling server, and for convenience of explanation, the data processing method will be described below by taking a computer device as an example. The data processing method may include the following steps S401 to S404:

S401: the method comprises the steps of obtaining the number of reference devices for pre-starting a target game in a target cluster at a first time point.

In the embodiment of the present application, the first time point herein specifically refers to a first time point (for example, 8:00 am) within a reference time period (for example, within the day, specifically, may be 1 month and 10 days).

S402: and determining the number of target devices according to the second predicted object number and the reference device number.

In one possible implementation, the determining, by the computer device, the number of target devices according to the second predicted object and the number of reference devices may include: the difference between the second prediction object and the reference device number is taken as the target device number.

In another possible implementation, the determining, by the computer device, the number of target devices according to the second predicted object and the number of reference devices may further include: firstly, acquiring a first predicted object number of a target game in a target cluster at a first time point; then, determining a corresponding error rate at a first point in time according to the first predicted object number and the reference object number; and finally, determining the number of target devices according to the error rate, the number of second predicted objects, the number of reference devices and the number of reference objects. The detailed process of determining the number of the first predicted objects may refer to the detailed process of determining the number of the second predicted objects in the embodiment of fig. 3, which is not described herein.

Specifically, assuming that the first prediction object number is denoted as f_expected (D, T), the calculation method of the error rate corresponding at the first time point (time T) is as shown in formula (10):

E(D,T)＝(F_expect(D,T)-F(D,T))/F(D,T) (10)

the error rate is an accurate basis for evaluating the prediction algorithm, and positive values of the error rate indicate that the number of the predicted objects is higher than the number of the objects actually used, and negative values of the error rate indicate that the number of the predicted objects is lower than the number of the objects actually used. The smaller the absolute value of the error rate and the more toward 0 indicates the more accurate the prediction result. The error rate is generally continuous from the viewpoint of observation, and if the number of objects predicted before is high, the probability of the number of objects predicted after is also high. The number of objects predicted later is also less probable if the number of objects predicted earlier is lower. If the number of previously predicted objects is relatively accurate, then the number of later predicted objects is also relatively accurate with a high probability. Thus, the accuracy of the number of target devices can be improved by introducing an error rate to determine the number of target devices.

S403: if the number of target devices is positive, adding devices for pre-starting the target game in the target cluster at a second time point.

In the embodiment of the application, the number of the newly added devices is equal to the absolute value of the number of the target devices.

In one possible implementation, if the error rate is greater than or equal to the preset error threshold, the process of determining, by the computer device, the target device number according to the second predicted object number, the reference object number, and the reference device number may include: firstly, acquiring a first difference value between a second predicted object number and a reference device number, and acquiring a second difference value between the reference object number and the reference device number; then, the maximum value between the first difference value and the second difference value is determined as the target device number. In this way, the number of target devices to be adjusted is determined according to the predicted number of objects and the actual number of objects (the number of reference objects), so that not only the prediction result obtained by the prediction algorithm (i.e., the first difference between the second predicted number of objects and the number of reference devices) can be compatible, but also the real-time number of objects at the current time T (i.e., the second difference between the number of reference objects and the number of reference devices) can be considered, and the probability of queuing the objects (i.e., the number of target devices is equal to the maximum value between the first difference and the second difference) can be reduced, thereby improving the accuracy of the number of devices that pre-start the target game at the second time point. It should be noted that the error rate is not incorporated into the calculation of the number of subsequent device adjustments here, since this part of the device redundancy can be tolerated.

For example, for a target cluster and target game, the following data may be obtained: a second prediction object number f_expect (D, t+1) at time t+1; reference object number F (D, T) at time T; the number G (T) of reference devices included in the target cluster at the moment T; error rate E (D, T) corresponding to time T. From the above data, the target device number O (T) can be determined.

If the error rate E (D, T) > =0 at time T, the current prediction result (the second number of predicted objects) is considered to be generally larger than the number of actually used objects, and this part of error can be tolerated (it can be understood that the larger the error rate, the lower the device utilization). At this time, the target device number O (T) is calculated as shown in the following formula (11):

O(T)＝max(F_expect(D,T+1)-G(T)，F(D,T)-G(T)) (11)

wherein, if O (T) is positive, the scheduling system will add O (T) devices, and if O (T) is negative, the scheduling system will reduce O (T) devices. As can be seen from the formula (11), taking the maximum value between f_expect (D, t+1) -G (T) and F (D, T) -G (T) on the basis of the devices included in the target cluster at the current time (T time) has the following advantages: the method can be compatible with the prediction result obtained by the prediction algorithm, and can also consider the real-time object number at the current T moment, so that the accuracy of the number of devices for pre-starting the target game at the second time point can be improved.

S404: if the number of target devices is negative, devices in the target cluster that pre-start the target game are reduced at a second point in time.

In the embodiment of the application, the number of the reduced devices is equal to the absolute value of the number of the target devices.

In one possible implementation, if the error rate is less than the preset error threshold, the process of determining, by the computer device, the target device number according to the error rate, the second predicted object number, the reference object number, and the reference device number may include: firstly, obtaining a product between an error rate and the number of reference objects, and obtaining a third difference value among the second predicted object number, the number of reference devices and the product; then, a second difference value between the number of reference objects and the number of reference devices is obtained; and finally, determining the maximum value between the third difference value and the second difference value as the device adjustment quantity corresponding to the target cluster. In this way, taking the maximum between the third difference and the second difference, based on the devices comprised by the target cluster at the current time T, has the advantage that: the prediction result (the third difference value between the second predicted object number, the reference device number and the product) obtained by the prediction algorithm can be compatible, the real-time object number (the second difference value between the reference object number and the reference device number) at the current time T can be considered, and the probability of queuing the objects (namely, the object device number is equal to the maximum value between the third difference value and the second difference value) can be reduced, so that the accuracy of the device number of the pre-starting object game at the second time point can be improved. Here, considering the influence of the error rate, since the number of objects at the next time is predicted to be smaller than the number of devices running the target game at the current time, it is necessary to compensate for the error due to the error rate (that is, E (D, T) F (D, T) in the following formula (12), where the error rate E (D, T) is negative, and subtracting E (D, T) F (D, T) is positive, that is, to compensate for/add the error), and the error rate needs to be included in the calculation of the device adjustment number.

If the error rate E (D, T) <0 at time T, the current prediction result (second prediction object number) is considered to be generally smaller than the actual object number, and this partial error may be compensated for in order to reduce the probability of queuing objects (similarly, the greater the error rate, the lower the device utilization). At this time, the target device number O (T) is calculated as shown in the following formula (12):

O(T)＝max(F_expect(D,T+1)-G(T)-E(D,T)*F(D,T)，F(D,T)-G(T)) (12)

wherein, if O (T) is positive, the scheduling system will add O (T) devices, and if O (T) is negative, the scheduling system will reduce O (T) devices. Similarly, as can be seen from the formula (12), taking the maximum value between f_expect (D, t+1) -G (T) -E (D, T) F (D, T) and F (D, T) -G (T) on the basis of the device included in the target cluster at the current time (T time) has the following advantages: the method can be compatible with the prediction result obtained by the prediction algorithm, and can also consider the real-time object number at the current T moment, so that the accuracy of the number of devices for pre-starting the target game at the second time point can be improved.

In the embodiment of the application, the device for pre-starting the target game in the target cluster comprises one or more of the following: virtual machine, development board, container. One development board (for example, an Android development board) may include an Android system, and one development board may specifically include a complete graphics processor (Graphics Processing Unit, GPU), a memory, a central processing unit (Central Processing Unit, CPU), a storage system, and so on. In addition, one or more development boards (for example, android development boards) may be integrated in the same server, and in this case, the server integrated with one or more Android development boards may be referred to as an ARM array server. The container may be running with a mirror image, which means a special file system, and may contain data related to the system program and data related to the target game. It should be noted that an Android development board in a virtual machine or an ARM array server may be referred to as an instance.

Subsequently, a virtual system of the target game may be installed in each device that pre-starts the target game, and the target game may be started and run based on the virtual system. By virtual system is meant a complete computer system with complete hardware system functionality that is software emulated and that operates in a completely isolated environment, the virtual system may be windows, android or other system. In addition, the virtual system may also comprise a system that runs inside the container.

In one possible implementation, when a start request of a target object for a target game is received, a target device may be allocated to the target object according to device capabilities of devices in the target cluster that pre-start the target game. The initiation request is then forwarded to the target device to cause the target device to respond to the initiation request.

Specifically, the starting request may carry an identifier (such as a nickname, an ID, etc.) of the target object, after the computer device receives the starting request of the target object for the target game, the identifier of the target object may be obtained, and then a historical game duration corresponding to the target object in the historical time period may be obtained based on the identifier of the target object. If the historical game time length corresponding to the target object is greater than or equal to the time length threshold, the device (such as a container) with better device performance of the device for pre-starting the target game in the target cluster can be used as the target device allocated to the target object. If the historical game duration corresponding to the target object is smaller than the duration threshold, the device with common device performance (such as a development board) of the device pre-starting the target game in the target cluster can be used as the target device allocated to the target object. It will be appreciated that so-called better device performance and device performance may generally be relative, e.g. devices in a target cluster include containers, virtual machines, development boards, then the device performance may be considered to be ordered from good to poor: container > virtual machine > development board. In this way, the target equipment is allocated for the target object according to the historical game duration of the target object experience target game, and the game duration of the target object experience target game in the future can be considered, so that the target equipment is allocated according to the requirement, and the equipment performance of the equipment can be considered.

In another possible implementation manner, when a starting request of the target object for the target game is received, the target device may be further allocated to the target object according to the historical starting times of devices for pre-starting the target game in the target cluster. The initiation request is then forwarded to the target device to cause the target device to respond to the initiation request.

Specifically, the starting request may carry an identifier (such as a nickname, an ID, etc.) of the target object, after the computer device receives the starting request of the target object for the target game, the identifier of the target object may be obtained, and then a historical game duration corresponding to the target object in the historical time period may be obtained based on the identifier of the target object. If the time length of the historical game corresponding to the target object is greater than or equal to the time length threshold, the device with the historical starting times smaller than or equal to the starting times threshold of the device for pre-starting the target game in the target cluster can be used as the target device allocated to the target object. If the historical game duration corresponding to the target object is smaller than the duration threshold, the device with the device performance of the device for pre-starting the target game in the target cluster being larger than the starting frequency threshold can be used as the target device allocated to the target object. In this way, the target devices allocated to the target objects are determined from the historical starting times of the devices, for example, devices with smaller starting times of the target objects with longer playing time periods can be allocated, and for example, devices with larger starting times of the target objects with smaller playing time periods can be allocated, so that load balancing of the devices in the target cluster can be ensured.

It will be appreciated that the response of the target device to the start-up request may specifically include: a virtual system that can run the target game (for example, may include related data for running the target game) is installed in the target device in advance, and then, after the target device is allocated to the target object, the virtual system of the target game may be executed in response to the start request, so that a game screen of the target game may be presented in the client of the target object (as in page S30 shown in fig. 1 a).

In summary, according to the calculation method provided by the embodiment of the application, the larger the error rate is, the lower the utilization rate of the devices in the target cluster is. Practice has shown that it is often necessary to ensure that the device utilization corresponding to the peak period of the experience target game is greater than 90%, i.e. the corresponding error rate at this time is required to be lower than 10%. For example, referring to fig. 5a, fig. 5a is a schematic diagram of a prediction curve of a target game in a target cluster according to an embodiment of the present application. As shown in fig. 5a, regarding the number of second predicted objects of the target game in the target cluster, the predicted value and the actual value at each time point are predicted by the prediction method provided by the embodiment of the present application, for example, at 20: about 00, the predicted value and the actual value reach the maximum value, wherein the predicted maximum value is 3.258k, and the actual maximum value is 3.509k, so that the error between the predicted maximum value and the actual maximum value is smaller, that is, the two curves shown in fig. 5a are almost identical, and therefore, the accuracy of the data processing scheme provided by the embodiment of the application is higher. In addition, referring to fig. 5b, fig. 5b is a schematic diagram of an error rate according to an embodiment of the application. As shown in fig. 5b, the error rate is relatively large when the number of objects is relatively small, decreases as the number of objects increases, and approaches 0. And, the error rate corresponding to the peak period (for example, about 20:00) of the experience target game is lower than 10%. Therefore, the data processing scheme provided by the embodiment of the application not only can accurately predict the number of objects at each time point, but also can ensure that the error rate at the peak period is lower than the actual scene requirement, and meanwhile, compared with the traditional prediction scheme, the prediction scheme provided by the application has the advantage that the near scheduling rate of the target game is increased from 26% to 78%, so that the multiplexing rate of equipment is increased.

In the embodiment of the application, the number of target devices can be determined by introducing the error rate, the error rate can be used as an accurate basis for evaluating the prediction algorithm, the error rate is positive value to indicate that the number of the predicted objects is higher than the number of the objects actually used, and the error rate is negative value to indicate that the number of the predicted objects is lower than the number of the objects actually used. The smaller the absolute value of the error rate and the more toward 0 indicates the more accurate the prediction result. The number of target devices is determined by the positive and negative of the error rate, so that the accuracy of the number of target devices can be improved.

Referring to fig. 6, fig. 6 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present application. The data processing apparatus 600 may be applied to the computer device in the foregoing embodiment. The data processing apparatus 600 may be a computer program (including program code) running in a computer device, for example the data processing apparatus 600 is an application software; the data processing device may be used to execute corresponding steps in the data processing method provided by the embodiment of the present application. The data processing apparatus 600 may include:

an obtaining unit 601, configured to obtain, when a device number prediction request of a target game is received, a reference object number of the target game in a target cluster, where the reference object number corresponds to a first time point;

An obtaining unit 601, configured to obtain, at a first time point in a historical time period, a first number of objects of a target game in a target cluster;

an obtaining unit 601, configured to obtain a second number of objects of the target game in the target cluster at a second time point in the history period;

a processing unit 602, configured to predict a second predicted object number of the target game in the target cluster at a second time point according to the first object number, the second object number and the reference object number;

the processing unit 602 is further configured to adjust, according to the second number of predicted objects, the number of devices in the target cluster that pre-start the target game at the second point in time.

In one possible implementation manner, the obtaining unit 601 obtains the number of reference objects of the target game in the target cluster, for performing the following operations:

the acquiring unit 601 acquires a process of determining a contribution degree of a target associated object to the number of target objects of the target cluster at a first time point, for performing the following operations:

the acquisition unit 601 acquires a first object number of a target game in a target cluster at a first time point in a history period, for performing the following operations:

the acquisition unit 601 acquires a first unit object number of the target game in the target cluster at a first time point in any time period, for performing the following operations:

In a possible implementation manner, the processing unit 602 adjusts the number of devices for pre-starting the target game in the target cluster at the second time point according to the second predicted object number, so as to perform the following operations:

In a possible implementation manner, the processing unit 602 determines the target device number according to the second predicted object number and the reference device number, and is configured to perform the following operations:

In a possible implementation manner, the processing unit 602 determines the number of target devices according to the error rate, the number of second predicted objects, the number of reference devices, and the number of reference objects, and is configured to perform the following operations:

In one possible implementation, the processing unit 602 determines the target device number according to the second predicted object number, the reference object number, and the reference device number, for performing the following operations:

In one possible implementation, the processing unit 602 determines the target device number according to the error rate, the second predicted object number, the reference object number, and the reference device number, and is configured to perform the following operations:

In one possible implementation, the processing unit 602 predicts a second predicted number of objects of the target game in the target cluster at the second point in time based on the first number of objects, the second number of objects, and the reference number of objects, for performing the following operations:

In one possible implementation, the processing unit 602 is further configured to perform the following operations:

Referring to fig. 7, fig. 7 is a schematic structural diagram of a computer device according to an embodiment of the application. The computer device 700 is configured to perform the steps performed by the computer device in the foregoing method embodiment, the computer device 700 comprising: one or more processors 710; one or more input devices 720, one or more output devices 730, and a memory 740. The processor 710, the input device 720, the output device 730, and the memory 740 are connected by a bus 750. The memory 740 is used for storing a computer program comprising program instructions, and the processor 710 is used for calling the program instructions stored in the memory 740 to perform the following operations:

In one possible implementation, the processor 710 obtains a reference object number of the target game in the target cluster for performing the following operations:

the processor 710 obtains a process of determining a contribution degree of the target associated object to the number of target objects of the target cluster at a first point in time, for performing the following operations:

the processor 710 obtains a first number of objects of the target game in the target cluster at a first point in time over the historical period of time for performing the following operations:

the processor 710 obtains a first number of unit objects of the target game in the target cluster at a first point in time within any one time period for performing the following operations:

In one possible implementation, the processor 710 adjusts the number of devices in the target cluster that pre-start the target game at the second point in time according to the second predicted object number, for performing the following operations:

In one possible implementation, the processor 710 determines a target device number according to the second predicted object number and the reference device number, for performing the following operations:

In one possible implementation, the processor 710 determines the target device number according to the error rate, the second predicted object number, the reference device number, and the reference object number, and is configured to:

In one possible implementation, the processor 710 determines the target device number according to the second predicted object number, the reference object number, and the reference device number, for performing the following operations:

In one possible implementation, the processor 710 determines the target device number based on the error rate, the second predicted object number, the reference object number, and the reference device number, for performing the following operations:

In one possible implementation, the processor 710 predicts a second predicted number of objects of the target game in the target cluster at a second point in time based on the first number of objects, the second number of objects, and the reference number of objects, for performing the operations of:

In one possible implementation, the processor 710 is further configured to:

Furthermore, it should be noted here that: the embodiment of the present application further provides a computer storage medium, in which a computer program is stored, and the computer program includes program instructions, when executed by a processor, can perform the method in the corresponding embodiment, so that a detailed description will not be given here. For technical details not disclosed in the embodiments of the computer storage medium according to the present application, please refer to the description of the method embodiments of the present application. As an example, the program instructions may be deployed on one computer device or executed on multiple computer devices at one site or distributed across multiple sites and interconnected by a communication network.

According to one aspect of the present application, there is provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device can perform the method in the foregoing corresponding embodiment, and therefore, a detailed description will not be given here.

Those skilled in the art will appreciate that implementing all or part of the above-described embodiment methods may be accomplished by way of a computer program for instructing relevant hardware, where the program may be stored on a computer readable storage medium, and where the program, when executed, may comprise the embodiment flow of the above-described methods. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random-access Memory (Random Access Memory, RAM), or the like.

The foregoing disclosure is illustrative of the present application and is not to be construed as limiting the scope of the application, which is defined by the appended claims.

Claims

1. A method of data processing, comprising:

acquiring a first object number of the target game in the target cluster at a first time point in a historical time period;

Predicting a second predicted number of objects of the target game in the target cluster at the second point in time according to the first number of objects, the second number of objects and the reference number of objects;

and adjusting the number of devices for pre-starting the target game in the target cluster at the second time point according to the second predicted object number.

2. The method of claim 1, wherein the obtaining the number of reference objects of the target game in a target cluster comprises:

acquiring contribution degrees of N associated objects associated with the target game to N target object numbers of the target cluster at the first time point, wherein the associated objects are objects which enter the target game or are objects which are queued to enter the target game at the first time point, and N is a positive integer;

and superposing the contribution degrees of the N target object quantities into the reference object quantity.

3. The method of claim 2, wherein the target cluster is any one of K clusters, K is a positive integer, and the target associated object is any one of the N associated objects;

The process of obtaining the contribution degree of the target associated object to the target object quantity of the target cluster at the first time point comprises the following steps:

obtaining K round trip delays between the client side where the target associated object is located and the K clusters at the first time point;

4. A method according to any one of claims 1-3, wherein the historical time period comprises P time periods, each time period comprising a first point in time, P being a positive integer;

the obtaining, at a first point in time within a historical time period, a first number of objects of the target game in the target cluster, including:

acquiring a first unit object number of the target game in the target cluster at a first time point in any time period;

and determining the first object number of the target game in the target cluster according to the acquired P first unit object numbers.

5. The method of claim 4, wherein the target cluster is any one of K clusters; the obtaining, at a first time point in any time period, a first unit object number of the target game in the target cluster, including:

Acquiring contribution degrees of M history associated objects associated with the target game to M history object numbers of the target cluster at a first time point in any time period, wherein the history associated objects are objects which enter the target game at the first time point in any time period or are objects which are queued to enter the target game, and M is an integer greater than 1;

and superposing the contribution degrees of the number of the M historical objects into the number of the first unit objects.

6. The method of claim 1, wherein the adjusting the number of devices in the target cluster that pre-start the target game at the second point in time based on the second predicted number of objects comprises:

acquiring the number of reference devices in the target cluster for pre-starting the target game at the first time point;

if the number of the target devices is positive, adding devices for pre-starting the target game in the target cluster at the second time point;

and if the number of the target devices is negative, reducing the devices for pre-starting the target game in the target cluster at the second time point, wherein the number of the newly added devices or the number of the reduced devices is equal to the absolute value of the number of the target devices.

7. The method of claim 6, wherein the determining the number of target devices from the second number of predicted objects, the number of reference devices, comprises:

acquiring a first predicted object number of the target game in a target cluster at the first time point;

determining a corresponding error rate at the first point in time according to the first predicted object number and the reference object number;

and determining the number of target devices according to the error rate, the second predicted object number, the reference device number and the reference object number.

8. The method of claim 7, wherein said determining a target device number based on said error rate, said second predicted object number, said reference device number, said reference object number, comprises:

if the error rate is greater than or equal to a preset error threshold, determining the number of target devices according to the number of second predicted objects, the number of reference objects and the number of reference devices;

and if the error rate is smaller than a preset error threshold, determining the number of target devices according to the error rate, the number of second predicted objects, the number of reference objects and the number of reference devices.

9. The method of claim 8, wherein the determining the number of target devices from the second number of predicted objects, the number of reference objects, and the number of reference devices comprises:

obtaining a first difference between the second predicted object number and the reference device number, and obtaining a second difference between the reference object number and the reference device number;

and determining the maximum value between the first difference value and the second difference value as the number of target devices.

10. The method of claim 8, wherein said determining a target device number based on said error rate, said second predicted object number, said reference object number, and said reference device number comprises:

obtaining a product between the error rate and the number of reference objects, and obtaining a third difference between the second number of predicted objects, the number of reference devices, and the product;

and determining the maximum value between the third difference value and the second difference value as the equipment adjustment quantity corresponding to the target cluster.

11. The method of claim 1, wherein predicting a second predicted number of objects for the target game in the target cluster at the second point in time based on the first number of objects, the second number of objects, and the reference number of objects comprises:

acquiring a ratio between the second object number and the first object number, and determining a product between the ratio and the reference object number as the second predicted object number; or alternatively, the process may be performed,

12. The method of claim 1, wherein the method further comprises:

when a starting request of a target object for the target game is received, distributing target equipment for the target object according to the equipment performance of equipment for pre-starting the target game in the target cluster; or alternatively, the process may be performed,

when a starting request of a target object for the target game is received, distributing target equipment for the target object according to the historical starting times of equipment for pre-starting the target game in the target cluster;

And forwarding the starting request to the target equipment so that the target equipment responds to the starting request.

13. A data processing apparatus, comprising:

the device comprises an acquisition unit, a judgment unit and a judgment unit, wherein the acquisition unit is used for acquiring the number of reference objects of a target game in a target cluster when receiving a device number prediction request of the target game, and the number of the reference objects corresponds to a first time point;

the acquisition unit is further used for acquiring the first object number of the target game in the target cluster at a first time point in a historical time period;

the acquiring unit is further configured to acquire a second number of objects of the target game in the target cluster at a second time point in the historical time period;

a processing unit, configured to predict a second predicted object number of the target game in the target cluster at the second time point according to the first object number, the second object number and the reference object number;

the processing unit is further configured to adjust, according to the second number of predicted objects, a number of devices that pre-start the target game in the target cluster at the second time point.

14. A computer device, comprising:

a processor adapted to execute a computer program;

a computer readable storage medium having stored therein a computer program which, when executed by the processor, implements the data processing method according to any of claims 1-12.

15. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program comprising program instructions which, when executed by a processor, cause the processor to perform the data processing method according to any of claims 1-12.

16. A computer program product, characterized in that the computer program product comprises a computer program adapted to be loaded by a processor and to perform the data processing method according to any of claims 1-12.