An intelligent management system for high-performance computing user quotas

Abstract

The present invention relates to the field of quota management system design, and in particular to an intelligent management system for high-performance computing user quotas. The present invention provides a data acquisition module for acquiring operation data input by a user terminal and recording resource usage data when game characters controlled by different user terminals interact with modeling targets pre-set in a game scene; a quota analysis module for calculating resource demand characterization parameters for modeling targets based on resource usage data corresponding to a number of modeling targets, so as to classify the resource demand degree categories of the modeling targets; and a resource allocation module for determining pre-interaction modeling targets based on the operation data of the user terminal, and adaptively allocating resources to the user terminal in advance according to the resource demand degree categories of the modeling targets, thereby ensuring the smoothness of game operation and the rationality of resource quotas, and improving the overall resource utilization efficiency.

Description

Technical Field

[0001] The present invention relates to the field of quota management system design, and in particular to an intelligent quota management system for high-performance computing users. Background Art

[0002] With the development of information technology and the explosive growth of data volume, the rational allocation of resources has become an urgent problem that needs to be solved. Therefore, user quota management can further understand and meet user needs by monitoring and analyzing user-side resource usage, flexibly manage and allocate resources, and allocate quotas to users by optimizing resource allocation, ensuring the rational use of resources and efficient operation of the system, thereby improving overall resource utilization.

[0003] Chinese Patent Publication No. CN111447577A discloses a quota management method and device, which includes: receiving a quota allocation request carrying a target user identifier sent by a network element, inputting data usage information of the target user within a second preset time period into a set prediction model corresponding to the target user to obtain a target quota, the network element providing mobile data services to the target user according to the target quota, and dynamically adjusting the target quota based on the actual situation of the user's recent data usage by using the target quota obtained by the set prediction model, so that the obtained target quota can accurately match the user's current data usage.

[0004] However, the existing technology still has the following problems: when the user side controls the game character to interact with the NPC character in the game, the user side will issue intensive, continuous and rapid control instructions. At the same time, the NPC character itself will also perform actions according to the settings, which will result in a large amount of resource consumption. Due to the failure to allocate resources in advance, the current resources may not be able to guarantee the smoothness of the game operation, thereby reducing the overall resource utilization efficiency. Summary of the Invention

[0005] To this end, the present invention provides a high-performance computing user quota intelligent management system to overcome the problem in the prior art that when the user side controls the game character to interact with the NPC character in the game, the user side will issue intensive, continuous and rapid control instructions. At the same time, the NPC character itself will also issue actions according to the settings, which will generate a large amount of resource consumption. Due to the failure to allocate resources in advance, the current resources may not be able to guarantee the smoothness of the game operation, thereby reducing the overall resource utilization efficiency.

[0006] To achieve the above objectives, the present invention provides a high-performance computing user quota intelligent management system, which includes:

[0007] A data acquisition module is used to acquire operation data input by the user terminal and record resource usage data when game characters controlled by different user terminals interact with modeling objects pre-set in the game scene. The operation data includes several instructions used by the user terminal to control the game characters to generate actions;

[0008] a quota analysis module connected to the data acquisition module and configured to calculate a resource demand characterizing parameter for a plurality of modeling targets based on resource usage data corresponding to the plurality of modeling targets, so as to classify the resource demand levels of the plurality of modeling targets, wherein the resource usage data corresponding to the plurality of modeling targets includes a throughput of a user terminal and a number of instructions issued by the user terminal;

[0009] The resource allocation module is connected to the data acquisition module and the quota analysis module respectively, and is used to determine the pre-interaction modeling target according to the operation data of the user terminal, and allocate resources to the user terminal according to the resource demand level category of the modeling target, including:

[0010] Adjusting the upper limit of the number of service nodes on the user side based on the resource demand characterization parameter, constructing a time domain curve of the number of service nodes, analyzing the slope of the time domain curve segment at each moment to determine whether a dense phase has been entered, calling the response time of each service node on the user side during the dense phase to determine whether congestion exists, and thereby revising the upper limit of the number of service nodes on the user side;

[0011] Alternatively, determining operational stability characterization parameters based on system load characteristics to determine whether to adjust the upper limit of the number of user-side service nodes;

[0012] Among them, the system load characteristics include the average number of active processes on the user side and the data transmission volume.

[0013] Furthermore, the quota analysis module calculates the resource demand characterization parameter according to formula (1):

[0014]

[0015] In formula (1), R represents the parameter representing resource demand, P represents the throughput of the user end, P0 represents the throughput threshold, U represents the number of instructions issued by the user end, U0 represents the instruction number threshold, α represents the throughput weight coefficient, and β represents the instruction number weight coefficient.

[0016] Furthermore, the quota analysis module is used to classify the resource demand degree of the modeling target into categories, including:

[0017] If the resource demand characterization parameter is greater than or equal to the resource demand characterization parameter threshold, the resource demand of the modeling target is determined to be a high degree category;

[0018] If the resource demand characterization parameter is less than the resource demand characterization parameter threshold, the resource demand of the modeling target is determined to be a low-level category.

[0019] Furthermore, the resource allocation module is used to allocate resources to the user terminal according to the resource demand level category of the modeling target, including:

[0020] If the resource demand of the modeling target is high, the upper limit of the number of service nodes on the user side is adjusted based on the resource demand characterization parameter, a time domain curve of the number of service nodes is constructed, and the slope of the time domain curve segment at each moment is analyzed to determine whether it has entered a dense phase. The response time of each service node on the user side during the dense phase is called to determine whether there is congestion, so as to adjust the upper limit of the number of service nodes on the user side;

[0021] If the resource demand of the modeling target is low, the operation stability characterization parameters are determined according to the system load characteristics to determine whether to adjust the upper limit of the number of user-side service nodes.

[0022] Furthermore, the resource allocation module is used to adjust the upper limit of the number of user-side service nodes, including:

[0023] The upper limit of the number of service nodes on the user side is increased, and the amount of increase in the upper limit of the number of service nodes is positively correlated with the resource demand characterization parameter.

[0024] Furthermore, the resource allocation module is used to construct the time domain curve of the number of service nodes and analyze the slope of the time domain curve at each moment, including:

[0025] Construct a rectangular coordinate system with time as the horizontal axis and the number of service nodes as the vertical axis;

[0026] Marking the coordinate points of the number of service nodes at each moment in the rectangular coordinate system;

[0027] Connecting the coordinate points with a smooth curve to obtain the time domain curve;

[0028] If there is a time domain curve segment whose slope is greater than or equal to the slope threshold, it is determined to enter the intensive stage.

[0029] Furthermore, the resource allocation module is used to determine whether congestion exists, including:

[0030] If the response time of each service node of the user end in the intensive phase is greater than or equal to the response time threshold, it is determined that congestion exists.

[0031] Furthermore, the resource allocation module is used to modify the upper limit of the number of user-side service nodes, including:

[0032] If congestion occurs, the upper limit of the number of user-side service nodes will be adjusted.

[0033] Furthermore, the resource allocation module is used to determine the operational stability characterization parameters including:

[0034] Calculating a ratio of an average number of active processes within a predetermined time period to an average number of active processes within a predetermined time period;

[0035] for calculating a data transmission volume ratio of a data transmission volume within a predetermined time period to a data transmission volume threshold within a preset predetermined time period;

[0036] The operation stability characterization parameter is obtained by respectively assigning corresponding weight values to the ratio of the average number of active processes and the ratio of the data transmission volume and summing them up.

[0037] Furthermore, the resource allocation module is used to determine whether to adjust the upper limit of the number of user-side service nodes, including:

[0038] If the operation stability characteristic parameter is greater than or equal to the operation stability characteristic parameter threshold, the upper limit of the number of client service nodes is adjusted.

[0039] Compared with the prior art, the present invention sets up a data acquisition module to obtain the operation data input by the user terminal and record the resource usage data when the game characters controlled by different user terminals interact with the modeling targets pre-set in the game scene; the quota analysis module is used to calculate the resource demand characterization parameters for the modeling targets based on the resource usage data corresponding to several modeling targets, so as to classify the resource demand degree categories of the modeling targets; the resource allocation module is used to determine the pre-interaction modeling targets based on the operation data of the user terminal, and adaptively allocate resources to the user terminal in advance according to the resource demand degree categories of the modeling targets, thereby ensuring the smoothness of the game operation and the rationality of the resource quota, and improving the overall resource utilization efficiency.

[0040] In particular, the present invention calculates resource demand characterization parameters for modeling targets through resource usage data corresponding to several modeling targets. During game operation, the user terminal continuously issues operation instructions, which will generate a certain amount of resource consumption and data transmission. In particular, when the user terminal controls the game character to interact with the modeling targets pre-set in the game scene, more resource consumption and data transmission will be generated. For example, when confronting an NPC character that needs to be defeated in a high-difficulty task, the operation instructions issued by the user terminal to control the game character to attack are intensive, continuous and fast. Therefore, sufficient resources are required to ensure the smoothness of the user terminal's control of the game character operation and avoid lag and other situations. Therefore, the present application calculates the resource demand characterization parameters for the modeling target through the user terminal's throughput and the number of instructions issued to characterize the degree of resources required when the user terminal controls the game character to interact with the modeling target, providing data support for the subsequent division of the resource demand degree of the modeling target, determining the pre-interaction modeling target based on the user terminal's operation data, and adaptively allocating resources to the user terminal in advance, thereby ensuring the smoothness of the game operation and the rationality of the resource quota, and improving the overall resource utilization efficiency.

[0041] In particular, the present invention pre-adjusts the upper limit of the number of user-side service nodes based on resource demand characterization parameters under the category of high resource demand of the modeling target. Based on the high resource demand of the pre-interacted modeling target, the upper limit of the number of user-side service nodes is pre-raised in advance to ensure the smoothness of game operation. During the continuous operation of the game, the number of service nodes used by the user side is detected, a time domain curve of the number of service nodes is constructed, and the slope of the time domain curve segment at each moment is analyzed. For example, if the slope of the time domain curve segment is greater than or equal to the slope of the time domain segment when the game is running stably, it is determined that the game has entered an intensive phase, and the response time data of each service node on the user side in the intensive phase is called. In actual situations, when data transmission is stable and resources can meet the needs of the user side, each service node on the user side will respond to the operation request instruction in a timely manner. If a service node fails to respond in time and a longer waiting time is required for the response, it is determined that congestion exists in this case, and the upper limit of the number of user-side service nodes is corrected to ensure the timely response of each service node and the smoothness of data transmission, thereby avoiding congestion.

[0042] In particular, the present invention determines the operation stability characterization parameters based on the system load characteristics under the low resource demand category of the modeling target, and predicts the resource demand based on the average number of active processes and the data transmission volume when the existing resources can meet the user-side needs. For example, a large number of active processes will generate high demand for resources, and a large amount of data transmission will increase the system load and increase the usage of memory resources, and at the same time will use more network bandwidth, CPU and other resources. Therefore, the present application uses the operation stability characterization parameters to characterize the user-side demand tendency for resources to determine whether to adjust the upper limit of the number of user-side service nodes, thereby ensuring the smoothness of game operation and the rationality of resource quotas, and improving the overall resource utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a functional module diagram of the high-performance computing user quota intelligent management system according to an embodiment of the invention;

[0044] Figure 2 A logical decision diagram for classifying resource demand levels for embodiments of the present invention;

[0045] Figure 3 A logic decision diagram for determining whether to enter the intensive phase for an embodiment of the invention;

[0046] Figure 4 A logical decision diagram for determining whether congestion exists according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0048] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0049] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the term "connection" should be understood in a broad sense. For example, it can refer to a fixed connection, a detachable connection, or an integral connection; it can refer to a mechanical connection or an electrical connection. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0050] See also Figures 1 to 4 As shown, Figure 1 This is a functional module diagram of the high-performance computing user quota intelligent management system according to an embodiment of the present invention. Figure 2 A logical decision diagram for classifying resource demand levels according to an embodiment of the invention, Figure 3 A logic decision diagram for determining whether to enter the intensive phase in an embodiment of the invention. Figure 4 The logical decision diagram for determining whether congestion exists in an embodiment of the present invention is provided. The intelligent management system for high-performance computing user quotas in an embodiment of the present invention includes:

[0051] A data acquisition module is used to acquire operation data input by the user terminal and record resource usage data when game characters controlled by different user terminals interact with modeling objects pre-set in the game scene. The operation data includes several instructions used by the user terminal to control the game characters to generate actions;

[0052] a quota analysis module connected to the data acquisition module and configured to calculate a resource demand characterizing parameter for a plurality of modeling targets based on resource usage data corresponding to the plurality of modeling targets, so as to classify the resource demand levels of the plurality of modeling targets, wherein the resource usage data corresponding to the plurality of modeling targets includes a throughput of a user terminal and a number of instructions issued by the user terminal;

[0053] The resource allocation module is connected to the data acquisition module and the quota analysis module respectively, and is used to determine the pre-interaction modeling target according to the operation data of the user terminal, and allocate resources to the user terminal according to the resource demand level category of the modeling target, including:

[0054] Adjusting the upper limit of the number of service nodes on the user side based on the resource demand characterization parameter, constructing a time domain curve of the number of service nodes, analyzing the slope of the time domain curve segment at each moment to determine whether a dense phase has been entered, calling the response time of each service node on the user side during the dense phase to determine whether congestion exists, and thereby revising the upper limit of the number of service nodes on the user side;

[0055] Alternatively, determining operational stability characterization parameters based on system load characteristics to determine whether to adjust the upper limit of the number of user-side service nodes;

[0056] Among them, the system load characteristics include the average number of active processes on the user side and the data transmission volume.

[0057] Specifically, the operation data input by the user terminal includes operation instructions such as instructions for controlling the movement of the game character and instructions for controlling the game character to use skills.

[0058] It can be understood that the modeling targets pre-set in the game scene include NPC characters in the game and character models such as bosses that need to be defeated in tasks of different difficulty levels. At the same time, controlling the game characters to interact with them can be a dialogue with the NPC characters, or a number of actions generated by fighting against the bosses that need to be defeated in the tasks. The relevant data can be obtained by calling the background running data, which will not be repeated here.

[0059] Specifically, there is no limitation on the specific structures of the data acquisition module, quota analysis module and resource allocation module, and they themselves or each unit therein can be composed of logic components or a combination of logic components, and the logic components include field programmable processors, computers or microprocessors in computers.

[0060] Specifically, the quota analysis module calculates the resource demand characterization parameter according to formula (1):

[0061]

[0062] In formula (1), R represents the parameter representing resource demand, P represents the throughput of the user end, P0 represents the throughput threshold, U represents the number of instructions issued by the user end, U0 represents the instruction number threshold, α represents the throughput weight coefficient, and β represents the instruction number weight coefficient.

[0063] In this example, the user-side throughput is the amount of data that the user-side device can receive and send per unit time in network communication.

[0064] The instruction number threshold U0 and throughput threshold P0 issued by the user end are pre-set. The instruction number and throughput within a predetermined time period recorded by the data acquisition module when different user ends interact with the corresponding modeling targets are obtained, and the mean instruction number ΔU and throughput mean ΔP corresponding to each user end are solved. It is set that U0 = r1×ΔU, P0 = r2×ΔP, r1 is the first deviation coefficient, r2 is the second deviation coefficient, 1.05<r1<1.10, 1.1<r2<1.15, α is 0.55, and β is 0.45.

[0065] The predetermined time period is selected within the interval [0.3ΔT, 0.5ΔT], where ΔT is the average interaction duration when different user terminals interact with the modeling target.

[0066] The present invention calculates resource demand characterization parameters for modeling targets through resource usage data corresponding to several modeling targets. During game operation, the user terminal continuously issues operation instructions, which will generate a certain amount of resource consumption and data transmission. In particular, when the user terminal controls the game character to interact with the modeling targets pre-set in the game scene, more resource consumption and data transmission will be generated. For example, when confronting an NPC character that needs to be defeated in a high-difficulty task, the operation instructions issued by the user terminal to control the game character to attack are intensive, continuous and fast. Therefore, sufficient resources are required to ensure the smoothness of the user terminal's control of the game character operation and avoid lag and other situations. Therefore, the present application calculates the resource demand characterization parameters for the modeling target through the user terminal's throughput and the number of instructions issued to characterize the degree of resources required when the user terminal controls the game character to interact with the modeling target, providing data support for the subsequent division of the resource demand degree of the modeling target, determining the pre-interaction modeling target based on the user terminal's operation data, and adaptively allocating resources to the user terminal in advance, thereby ensuring the smoothness of game operation and the rationality of resource quotas, and improving the overall resource utilization efficiency.

[0067] Specifically, the quota analysis module is used to classify the resource demand degree of the modeling target into categories, including:

[0068] If the resource demand characterization parameter is greater than or equal to the resource demand characterization parameter threshold, the resource demand of the modeling target is determined to be a high degree category;

[0069] If the resource demand characterization parameter is less than the resource demand characterization parameter threshold, the resource demand of the modeling target is determined to be a low-level category.

[0070] The resource demand characterization parameter threshold R0 is selected in the interval [1.35, 1.47].

[0071] Specifically, the resource allocation module is used to allocate resources to the user terminal according to the resource demand level category of the modeling target, including:

[0072] If the resource demand of the modeling target is high, the upper limit of the number of service nodes on the user side is adjusted based on the resource demand characterization parameter, a time domain curve of the number of service nodes is constructed, and the slope of the time domain curve segment at each moment is analyzed to determine whether it has entered a dense phase. The response time of each service node on the user side during the dense phase is called to determine whether there is congestion, so as to adjust the upper limit of the number of service nodes on the user side;

[0073] If the resource demand of the modeling target is low, the operation stability characterization parameters are determined according to the system load characteristics to determine whether to adjust the upper limit of the number of user-side service nodes.

[0074] Specifically, there is no limitation on the method of determining the pre-interaction modeling target. It can be to identify the modeling target corresponding to the jump position according to the jump position instruction issued by the user end, and determine the corresponding modeling target as the pre-interaction modeling target. This will not be repeated.

[0075] Specifically, the resource allocation module is used to adjust the upper limit of the number of user-side service nodes, including:

[0076] The upper limit of the number of service nodes on the user side is increased, and the amount of increase in the upper limit of the number of service nodes is positively correlated with the resource demand characterization parameter.

[0077] In this embodiment, optionally,

[0078] Compare the resource demand characterization parameter R with the first resource demand characterization parameter comparison threshold R1 and the second resource demand characterization parameter comparison threshold R2,

[0079] If R>R2, the service node number upper limit increase amount is determined to be the first service node number upper limit increase amount c1, and c1=[0.45c0];

[0080] If R1≤R≤R2, then the service node number upper limit increase amount is determined to be the second service node number upper limit increase amount c2, and c2 is set to [0.33c0];

[0081] If R<R1, the service node number upper limit increase amount is determined to be the third service node number upper limit increase amount c3, and c3 is set to [0.21c0];

[0082] Wherein, c0 represents the upper limit of the initial service node, R1=1.2R0, R2=1.3R0.

[0083] Regarding the upper limit of the initial service node, those skilled in the art can determine it based on the scale, design, number of players and complexity of the server architecture of the game, which will not be elaborated here.

[0084] Specifically, the resource allocation module is used to construct the time domain curve of the number of service nodes and analyze the slope of the time domain curve at each moment, including:

[0085] Construct a rectangular coordinate system with time as the horizontal axis and the number of service nodes as the vertical axis;

[0086] Marking the coordinate points of the number of service nodes at each moment in the rectangular coordinate system;

[0087] Connecting the coordinate points with a smooth curve to obtain the time domain curve;

[0088] If there is a time domain curve segment whose slope is greater than or equal to the slope threshold, it is determined to enter the intensive stage.

[0089] The slope threshold F0 is selected in the interval [0.24, 0.36].

[0090] Specifically, the resource allocation module is used to determine whether congestion exists, including:

[0091] If the response time of each service node of the user end in the intensive phase is greater than or equal to the response time threshold, it is determined that congestion exists.

[0092] In the dense phase, if the response time of each service node on the user side is less than the response time threshold, it is determined that there is no congestion.

[0093] The response time threshold T0 is pre-set. The response time data of each service node during the user-side game is obtained, and the average response time ΔT is calculated. It is set that T0 = k × ΔT, where k is the offset coefficient, 1.15 < r < 1.3.

[0094] Specifically, the resource allocation module is used to modify the upper limit of the number of user-side service nodes, including:

[0095] If congestion occurs, the upper limit of the number of user-side service nodes will be adjusted.

[0096] It is understandable that if there is congestion, the upper limit of the number of user-side service nodes will be increased to ensure smooth data transmission and timely response of service nodes, avoid freezes and other situations, and at the same time, meet the resource needs of the user side. The increase can be 0.1-0.3 times the current upper limit of the number of service nodes.

[0097] In the present invention, under the category of high resource demand of the modeling target, the upper limit of the number of user-side service nodes is adjusted in advance according to the resource demand characterization parameter. According to the high resource level required by the pre-interacted modeling target, the upper limit of the number of user-side service nodes is raised in advance to ensure the smoothness of game operation. During the continuous operation of the game, the number of service nodes used by the user side is detected, a time domain curve of the number of service nodes is constructed, and the slope of the time domain curve segment at each moment is analyzed. For example, if the slope of the time domain curve segment is greater than or equal to the slope of the time domain segment when the game is running stably, it is determined that the intensive phase has begun, and the response time data of each service node of the user side in the intensive phase is called. In actual situations, when data transmission is stable and resources can meet the needs of the user side, each service node of the user side will respond to the operation request instruction in a timely manner. If the service node fails to respond in time and a longer waiting time for the response is required, it is determined that congestion exists in this case, and the upper limit of the number of user-side service nodes is corrected to ensure the timely response of each service node and the smoothness of data transmission, thereby avoiding congestion.

[0098] Specifically, the resource allocation module is used to determine the operational stability characterization parameters including:

[0099] Calculating a ratio of an average number of active processes within a predetermined time period to an average number of active processes within a predetermined time period;

[0100] for calculating a data transmission volume ratio of a data transmission volume within a predetermined time period to a data transmission volume threshold within a preset predetermined time period;

[0101] The operation stability characterization parameter is obtained by respectively assigning corresponding weight values to the ratio of the average number of active processes and the ratio of the data transmission volume and summing them up.

[0102] In this embodiment, the operational stability characterization parameter D is calculated according to the following formula:

[0103]

[0104] In the formula, D represents the operation stability characterization parameter, H represents the mean number of active processes within a predetermined time period, H0 represents the threshold number of active processes, J represents the data transmission volume within a predetermined time period, J0 represents the data transmission volume threshold, γ represents the weight coefficient of the ratio of the mean number of active processes, and δ represents the weight coefficient of the data transmission volume ratio.

[0105] The threshold value H0 for the number of active processes and the threshold value J0 for the amount of data transmitted are preset. The number of active processes and the amount of data transmitted within a predetermined time period when different user terminals interact with the corresponding modeling targets as recorded by the data acquisition module are obtained. The mean value ΔH of the number of active processes and the mean value ΔJ of the amount of data transmitted corresponding to each user terminal are solved. It is set that H0 = q1 × ΔH, J0 = q2 × ΔJ, q1 is the first precision coefficient, q2 is the second precision coefficient, 1.07 < q1 < 1.14, 1.15 < q2 < 1.19, γ is 0.57, and δ is 0.43.

[0106] The predetermined time period is selected within the interval [0.3ΔT, 0.5ΔT], where ΔT is the average interaction duration when different user terminals interact with the modeling target.

[0107] Specifically, the resource allocation module is used to determine whether to adjust the upper limit of the number of user-side service nodes, including:

[0108] If the operation stability characteristic parameter is greater than or equal to the operation stability characteristic parameter threshold, the upper limit of the number of client service nodes is adjusted.

[0109] The upper limit of the number of service nodes will be increased by 0.1-0.3 times the initial upper limit of the number of service nodes.

[0110] The present invention determines the operation stability characterization parameters according to the system load characteristics under the low resource demand category of the modeling target. When the existing resources can meet the user-side demand, the resource demand is predicted based on the average number of active processes and the data transmission volume. For example, a large number of active processes will generate high demand for resources. A large amount of data transmission will increase the system load and the usage of memory resources, and will also use more network bandwidth, CPU and other resources. Therefore, the present application uses the operation stability characterization parameters to characterize the user-side demand tendency for resources to determine whether to adjust the upper limit of the number of user-side service nodes, thereby ensuring the smoothness of game operation and the rationality of resource quotas, and improving the overall resource utilization efficiency.

[0111] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

Claims

1. A high-performance computing user quota intelligent management system, characterized in that: include: A data acquisition module is used to acquire operation data input by the user terminal and record resource usage data when game characters controlled by different user terminals interact with modeling objects pre-set in the game scene. The operation data includes several instructions used by the user terminal to control the game characters to generate actions; a quota analysis module connected to the data acquisition module and configured to calculate a resource demand characterizing parameter for a plurality of modeling targets based on resource usage data corresponding to the plurality of modeling targets, so as to classify the resource demand levels of the plurality of modeling targets, wherein the resource usage data corresponding to the plurality of modeling targets includes a throughput of a user terminal and a number of instructions issued by the user terminal; The resource allocation module is connected to the data acquisition module and the quota analysis module respectively, and is used to determine the pre-interaction modeling target according to the operation data of the user terminal, and allocate resources to the user terminal according to the resource demand level category of the modeling target, including: Adjusting the upper limit of the number of service nodes on the user side based on the resource demand characterization parameter, constructing a time domain curve of the number of service nodes, analyzing the slope of the time domain curve segment at each moment to determine whether a dense phase has been entered, calling the response time of each service node on the user side during the dense phase to determine whether congestion exists, and thereby revising the upper limit of the number of service nodes on the user side; Alternatively, determining operational stability characterization parameters based on system load characteristics to determine whether to adjust the upper limit of the number of user-side service nodes; Among them, the system load characteristics include the average number of active processes and data transmission volume on the user side; The quota analysis module calculates the resource demand characterization parameter according to formula (1): In formula (1), R represents the parameter representing resource demand, P represents the throughput of the user end, P0 represents the throughput threshold, U represents the number of instructions issued by the user end, U0 represents the instruction number threshold, α represents the throughput weight coefficient, and β represents the instruction number weight coefficient; The quota analysis module is used to classify the resource demand degree of the modeling target into categories, including: If the resource demand characterization parameter is greater than or equal to the resource demand characterization parameter threshold, the resource demand of the modeling target is determined to be a high degree category; If the resource demand characterization parameter is less than the resource demand characterization parameter threshold, the resource demand of the modeling target is determined to be a low-level category.

2. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to allocate resources to the user end according to the resource demand level category of the modeling target. include, If the resource demand of the modeling target is high, the upper limit of the number of service nodes on the user side is adjusted based on the resource demand characterization parameter, a time domain curve of the number of service nodes is constructed, and the slope of the time domain curve segment at each moment is analyzed to determine whether it has entered a dense phase. The response time of each service node on the user side during the dense phase is called to determine whether there is congestion, so as to adjust the upper limit of the number of service nodes on the user side; If the resource demand of the modeling target is low, the operation stability characterization parameters are determined according to the system load characteristics to determine whether to adjust the upper limit of the number of user-side service nodes.

3. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to adjust the upper limit of the number of user-side service nodes, including: The upper limit of the number of service nodes on the user side is increased, and the amount of increase in the upper limit of the number of service nodes is positively correlated with the resource demand characterization parameter.

4. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to construct the time domain curve of the number of service nodes and analyze the slope of the time domain curve at each moment, including: Construct a rectangular coordinate system with time as the horizontal axis and the number of service nodes as the vertical axis; Marking the coordinate points of the number of service nodes at each moment in the rectangular coordinate system; Connecting the coordinate points with a smooth curve to obtain the time domain curve; If there is a time domain curve segment whose slope is greater than or equal to the slope threshold, it is determined to enter the intensive stage.

5. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to determine whether congestion exists, including: If the response time of each service node of the user end in the intensive phase is greater than or equal to the response time threshold, it is determined that congestion exists.

6. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to modify the upper limit of the number of user-side service nodes, including: If congestion occurs, the upper limit of the number of user-side service nodes will be adjusted.

7. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to determine the operational stability characterization parameters including: Calculating a ratio of an average number of active processes within a predetermined time period to an average number of active processes within a predetermined time period; for calculating a data transmission volume ratio of a data transmission volume within a predetermined time period to a data transmission volume threshold within a preset predetermined time period; The operation stability characterization parameter is obtained by respectively assigning corresponding weight values to the ratio of the average number of active processes and the ratio of the data transmission volume and summing them up.

8. The high performance computing user quota intelligent management system according to claim 1, characterized in that: The resource allocation module is used to determine whether to adjust the upper limit of the number of user-side service nodes, including: If the operation stability characteristic parameter is greater than or equal to the operation stability characteristic parameter threshold, the upper limit of the number of client service nodes is adjusted.

Images