A Big Data-Based Dynamic Scheduling Service System for Computer Resources

Abstract

This invention discloses a dynamic scheduling service system for computer resources based on big data, relating to the field of computer technology. It includes: a computing power classification module; a coefficient calculation module, which obtains the importance coefficient of local data blocks; a defense generation module, which forms a defense scheme; a data analysis module, which analyzes the computing power consumption value of the defense scheme and the amount of data loss caused by network attacks; a scheme formation module, which calculates the data loss coefficient of the computing power allocation scheme; and a computing power scheduling module, which selects the computing power allocation scheme with the smallest data loss coefficient as the target allocation scheme. By setting up the coefficient calculation module, defense generation module, scheme formation module, and computing power scheduling module, computing power can be rationally allocated even when the defense against attacks is insufficient, ensuring that the overall impact of data loss is minimized.

Description

Technical Field

[0001] This invention relates to the field of computer technology, specifically to a dynamic scheduling service system for computer resources based on big data. Background Technology

[0002] In computer science and related fields, resources refer to the various elements or components required for a system to perform tasks or provide services, including hardware, software, data, and human resources. Resources can be divided into hardware resources and software resources. Computer resources are diverse. Data protection is crucial in computer use. Allocating computing power and defending against network attacks are essential for data protection. However, current technology is insufficient in analyzing situations where computing power is inadequate, making it difficult to rationally allocate computer computing resources and thus limiting the impact of data loss. Summary of the Invention

[0003] To address the aforementioned technical problems, a dynamic scheduling service system for computer resources based on big data is provided. This technical solution resolves the issues raised in the background section.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A big data-based dynamic scheduling service system for computer resources includes: The computing power classification module updates the redundant computing power values ​​that are not called in the computer in real time at preset intervals, forming the computing power demand fluctuation value of computer data processing, and dividing the redundant computing power values ​​into a first computing power value and a second computing power value. The coefficient calculation module divides the data stored in the computer into at least one local data block, performs importance analysis on the data in the local data block, and obtains the importance coefficient of the local data block. A defense generation module identifies at least one network attack on a local data block and, based on historical data, formulates a defense scheme against the network attack. The data analysis module analyzes the computing power consumption of the defense scheme and the amount of data loss caused by the network attack. The scheme forming module forms at least one computing power allocation scheme for a local data block based on a second computing power value. Under the condition of implementing the computing power allocation scheme, at least one secondary allocation scheme is formed for the local computing power value allocated to the local data block. Based on the secondary allocation scheme, the data loss coefficient of the computing power allocation scheme is calculated. The computing power scheduling module selects the computing power allocation scheme with the smallest data loss coefficient as the target allocation scheme, and schedules the computer according to the local computing power value allocated to the local data block in the target allocation scheme and the second computing power value of the computer.

[0005] Preferably, the process of generating a fluctuation value for computing power demand in computer data processing and dividing redundant computing power values ​​into a first computing power value and a second computing power value includes the following steps: Obtain at least one historical data processing status, the duration of which is equal to a preset time. In the historical data processing status, obtain the historical maximum and minimum values ​​of computing power usage. The difference between the historical maximum and minimum values ​​is used to obtain the historical fluctuation value. The maximum value of the historical fluctuation value is used as the computing power demand fluctuation value. The value of computing power demand fluctuation is used as the first computing power value, and the result of subtracting the first computing power value from the redundant computing power value is used as the second computing power value.

[0006] Preferably, the step of performing importance analysis on the data in a local data block to obtain the importance coefficient of the local data block includes the following steps: Obtain at least one historical data call scenario, and use the data called in the historical data call scenario as a historical data packet; The historical data retrieval scenario corresponding to the historical data packet that overlaps with the local data block is used as the target data retrieval scenario; The overlapping portion of historical data packets and local data blocks in the target data retrieval scenario is taken as the target portion; The total frequency of all historical data call scenarios is taken as the first frequency, and the total frequency of the target data call scenarios is taken as the second frequency. Divide the second frequency by the first frequency to obtain the weight of the target data call scenario. Multiply the total cost of the project applied to the target data call scenario by the weight of the target data call scenario to obtain the application value of the target data call scenario. The proportion of the target portion to the historical data packets of the target data call scenario is taken as the first proportion. The application value of the target data call scenario is multiplied by the first proportion to obtain the data value of the target portion. The proportion of the target portion to the local data block is used as the second proportion. The data value of the target portion is multiplied by the second proportion to obtain the data contribution of the target portion. The data contributions of the target portion are accumulated to obtain the importance coefficient of the local data block.

[0007] Preferably, identifying at least one network attack on a local data block includes the following steps: In big data, at least one historical attack is obtained, and the attack on a local data block is compared with the historical attack. The historical attack that attacked the local data block is regarded as a network attack.

[0008] Preferably, the method for developing a network attack defense scheme based on historical data includes the following steps: Historical data will be used to summarize defense measures against cyberattacks and form a cyberattack defense plan.

[0009] Preferably, the analysis to obtain the computational cost of the defense scheme includes the following steps: The defense scheme is evenly divided into at least one defense block, and the data from the network attacks defended by the defense block are aggregated as the feature data set of the defense block. Defense blocks with the same feature data set are merged to obtain defense steps, and the computing power consumed by the defense steps is used as the computing power value of the defense steps. During the course of a cyberattack, at least one consecutive time interval of the defensive steps is combined to form the time range of the defensive steps. During the course of a cyberattack, at least one time point is taken evenly, and the time ranges containing these time points are summarized into a set of time ranges of action of the time points. The computing power values ​​of the defense steps corresponding to the effective time ranges in the set of effective time ranges are summed to obtain the total computing power value of the time points. The maximum value of the total computing power value of at least one time point is taken as the computing power consumption value of the defense scheme.

[0010] Preferably, the analysis to determine the amount of data loss caused by the network attack includes the following steps: In big data, we obtain historical attack data of at least one network attack under undefended conditions, and take the average of the amount of data destroyed in the historical attack data to obtain the amount of data loss caused by the network attack.

[0011] Preferably, the at least one computing power allocation scheme for forming a local data block includes the following steps: The second computing power value is evenly divided into at least one local computing power value, and at least one local computing power value is randomly matched to a local data block. Each matching method of the local computing power value is recorded as at least one computing power allocation scheme.

[0012] Preferably, the process of forming at least one secondary allocation scheme for the local computing power values ​​allocated to local data blocks includes the following steps: The defense steps in the defense scheme against network attacks are to evenly divide the local computing power value allocated to a local data block into at least one secondary computing power value and randomly allocate the secondary computing power value to the corresponding local data block. Each matching method of the secondary computing power value is recorded as at least one secondary allocation scheme.

[0013] Preferably, the calculation of the data loss coefficient of the computing power allocation scheme includes the following steps: The total computing power consumption of the local data block is obtained by summing the computing power consumption values ​​of defense schemes against at least one network attack acting on the local data block. If the local computing power value allocated to a local data block in the computing power allocation scheme is not lower than the total computing power consumption value of the local data block, then the loss value of the local data block with respect to the computing power allocation scheme is 0. Otherwise, in the secondary allocation scheme, the secondary computing power values ​​allocated to the defense step are summed to obtain the total computing power allocation value for the defense step; If the total computing power allocation value of the defense step is less than the computing power value of the defense step, then the defense step will be regarded as an abnormal defense step. Subtracting the total computing power allocation of the anomaly defense steps from the computing power value of the anomaly defense steps yields the computing power gap of the anomaly defense steps. The computing power gaps of the anomaly defense steps for local data blocks are superimposed to obtain the total computing power gap for the local data blocks; The ratio of the total computing power gap of a local data block to the total computing power consumption of the local data block is obtained by dividing the total computing power gap of the local data block. The total loss of a local data block is obtained by summing up the data loss caused by network attacks on local data blocks. The data loss coefficient of the secondary allocation scheme is obtained by multiplying the importance coefficient of the local data block, the total loss of the local data block, and the gap ratio of the local data block. The average data loss coefficients of at least one secondary allocation scheme for local data blocks generated under the condition of implementing the computing power allocation scheme are taken to obtain the loss value of the local data block with respect to the computing power allocation scheme. The data loss coefficient of the computing power allocation scheme is obtained by superimposing the loss values ​​of at least one local data block with respect to the computing power allocation scheme.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: By setting up a coefficient calculation module, a defense generation module, a scheme formation module, and a computing power scheduling module, the importance of the data is identified, and multiple allocation methods are formed based on different possibilities for the allocation of computing power. The data loss under each allocation method is estimated. Based on the estimated situation, the computing power allocation scheme can be screened, so that computing power can be reasonably allocated when the defense against attacks is insufficient, and the overall impact of data loss can be minimized. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the big data-based dynamic scheduling service system for computer resources of the present invention. Figure 2 This is a flowchart illustrating the process of generating the computing power demand fluctuation value for computer data processing and dividing the redundant computing power value into a first computing power value and a second computing power value according to the present invention. Figure 3 This is a flowchart illustrating the process of performing importance analysis on data in a local data block to obtain the importance coefficient of the local data block according to the present invention. Figure 4 This is a schematic diagram illustrating the process of obtaining the computational cost of the defense scheme based on the analysis of this invention. Figure 5 This is a flowchart illustrating the process of calculating the data loss coefficient of the computing power allocation scheme according to the present invention. Detailed Implementation

[0016] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0017] Reference Figure 1 As shown, a computer resource dynamic scheduling service system based on big data includes: The computing power classification module updates the redundant computing power values ​​that are not called in the computer in real time at preset intervals, forming the computing power demand fluctuation value of computer data processing, and dividing the redundant computing power values ​​into a first computing power value and a second computing power value. The coefficient calculation module divides the data stored in the computer into at least one local data block, performs importance analysis on the data in the local data block, and obtains the importance coefficient of the local data block. A defense generation module identifies at least one network attack on a local data block and, based on historical data, formulates a defense scheme against the network attack. The data analysis module analyzes the computing power consumption of the defense scheme and the amount of data loss caused by the network attack. The scheme forming module forms at least one computing power allocation scheme for a local data block based on a second computing power value. Under the condition of implementing the computing power allocation scheme, at least one secondary allocation scheme is formed for the local computing power value allocated to the local data block. Based on the secondary allocation scheme, the data loss coefficient of the computing power allocation scheme is calculated. The computing power scheduling module selects the computing power allocation scheme with the smallest data loss coefficient as the target allocation scheme, and schedules the computer according to the local computing power value allocated to the local data block in the target allocation scheme and the second computing power value of the computer.

[0018] This solution primarily focuses on allocating computing power for defending against network attacks. Since computing power serves many purposes beyond attack defense, only redundant computing power can be used for defense. This avoids impacting the use of other computing power. Furthermore, because the demand for computing power fluctuates in other areas, redundant computing power cannot be used entirely for attack defense; a portion must be reserved to handle fluctuations in other computing power needs. Secondly, when computing power is sufficient, there is little need to consider this, as no data loss will occur regardless of the defense. However, when there are many attacks, insufficient defense may occur, resulting in data loss from some attacks. Different allocations of computing power will lead to different defense effectiveness. Therefore, computing power needs to be allocated strategically based on the type of attack and the type of data to minimize the impact of data loss. A series of steps will be implemented to address this issue in subsequent steps.

[0019] Reference Figure 2 As shown, the process of generating fluctuation values ​​in computing power demand for computer data processing and dividing redundant computing power values ​​into a first computing power value and a second computing power value includes the following steps: Obtain at least one historical data processing status, the duration of which is equal to a preset time. In the historical data processing status, obtain the historical maximum and minimum values ​​of computing power usage. The difference between the historical maximum and minimum values ​​is used to obtain the historical fluctuation value. The maximum value of the historical fluctuation value is used as the computing power demand fluctuation value. The value of computing power demand fluctuation is used as the first computing power value, and the result of subtracting the first computing power value from the redundant computing power value is used as the second computing power value.

[0020] Here, the redundant computing power value is updated periodically. That is, the redundant computing power value is reacquired at preset intervals. The preset interval is set based on experience and usage needs. Because the tasks being processed differ greatly over time, the use of computing power will also be different, and the redundant computing power value will also vary significantly, requiring updates to the redundant computing power value. In addition, computing power usage fluctuates within the preset time. Therefore, the redundant computing power value cannot be used entirely for defense. A portion of it needs to be used to cope with fluctuations in the use of other computing power. This portion is the first computing power value. The first computing power value is set based on historical fluctuations and is sufficient to cope with fluctuations. The remaining computing power, the second computing power value, is used entirely for attack defense.

[0021] Reference Figure 3 As shown, the importance analysis of data in a local data block to obtain the importance coefficient of the local data block includes the following steps: Obtain at least one historical data call scenario, and use the data called in the historical data call scenario as a historical data packet; The historical data retrieval scenario corresponding to the historical data packet that overlaps with the local data block is used as the target data retrieval scenario; The overlapping portion of historical data packets and local data blocks in the target data retrieval scenario is taken as the target portion; The total frequency of all historical data call scenarios is taken as the first frequency, and the total frequency of the target data call scenarios is taken as the second frequency. Divide the second frequency by the first frequency to obtain the weight of the target data call scenario. Multiply the total cost of the project applied to the target data call scenario by the weight of the target data call scenario to obtain the application value of the target data call scenario. The proportion of the target portion to the historical data packets of the target data call scenario is taken as the first proportion. The application value of the target data call scenario is multiplied by the first proportion to obtain the data value of the target portion. The proportion of the target portion to the local data block is used as the second proportion. The data value of the target portion is multiplied by the second proportion to obtain the data contribution of the target portion. The data contributions of the target portion are accumulated to obtain the importance coefficient of the local data block.

[0022] The importance of data stored in a computer varies, and the impact of loss on data of different importance is different. Therefore, by dividing the data into blocks, the data in each local data block are adjacent and have high similarity, thus their importance is relatively consistent. Thus, the importance coefficient of the local data block can be identified and used to evaluate the loss situation later. The importance of data is primarily determined by its application. The target portion refers to the part of the local data block that appears in the target data call scenario. The higher the frequency of the target data call scenario, the more times the target portion is called, and therefore, its importance increases accordingly. Secondly, the higher the proportion of the target portion in the target data call scenario, the higher its importance. At the same time, the target portion is also related to the total cost of the project in which the target data call scenario is applied. The higher the cost, the more important the target data call scenario, and the more important the corresponding target portion. Therefore, these factors are combined to form the data value of the target portion. However, since a local data block consists of at least one target portion, the data value of the target portion is added proportionally to obtain the importance coefficient of the local data block.

[0023] Identifying at least one network attack affecting a local data block includes the following steps: In big data, at least one historical attack is obtained, and the attack on a local data block is compared with the historical attack. The historical attack that attacked the local data block is regarded as a network attack.

[0024] Based on historical data, a defense plan against cyberattacks includes the following steps: Historical data will be used to summarize defense measures against cyberattacks and form a cyberattack defense plan.

[0025] Reference Figure 4 As shown, the analysis of the computational cost of the defense scheme includes the following steps: The defense scheme is evenly divided into at least one defense block, and the data from the network attacks defended by the defense block are aggregated as the feature data set of the defense block. Defense blocks with the same feature data set are merged to obtain defense steps, and the computing power consumed by the defense steps is used as the computing power value of the defense steps. During the course of a cyberattack, at least one consecutive time interval of the defensive steps is combined to form the time range of the defensive steps. During the course of a cyberattack, at least one time point is taken evenly, and the time ranges containing these time points are summarized into a set of time ranges of action of the time points. The computing power values ​​of the defense steps corresponding to the effective time ranges in the set of effective time ranges are summed to obtain the total computing power value of the time points. The maximum value of the total computing power value of at least one time point is taken as the computing power consumption value of the defense scheme.

[0026] When defending against cyberattacks, a defense plan is implemented according to the plan. The plan may have multiple defense steps, but during the defense process, some steps may run in parallel rather than all steps running synchronously. Therefore, it is necessary to sum up the computing power consumed by each defense step at each point in time to calculate the peak computing power consumption and use it as the computing power consumption value of the defense plan. Instead of summing up the computing power consumed by all defense steps, this would result in a large discrepancy with the actual computing power required for defense, leading to greater analysis errors. It should be noted here that the effective time range may be the union of several intervals such as (1,2), (4,5) and (8,9), rather than a continuous interval range.

[0027] Analyzing the amount of data loss caused by a cyberattack includes the following steps: In big data, we obtain historical attack data of at least one network attack under undefended conditions, and take the average of the amount of data destroyed in the historical attack data to obtain the amount of data loss caused by the network attack.

[0028] At least one computing power allocation scheme for forming local data blocks includes the following steps: The second computing power value is evenly divided into at least one local computing power value, and at least one local computing power value is randomly matched to a local data block. Each matching method of the local computing power value is recorded as at least one computing power allocation scheme.

[0029] Local computing power can be allocated to any one of at least one local data block. The defensive effect produced by the local computing power will be different depending on the local data block it is allocated to. For example, there are local computing power 1, local computing power 2 and local computing power 3, and local data block 1, local data block 2 and local data block 3. Local computing power 1 is allocated to local data block 1, local computing power 2 is allocated to local data block 2, and local computing power 3 is allocated to local data block 3. This is one allocation method. Local computing power 1 and local computing power 2 are both allocated to local data block 1, and local computing power 3 is allocated to local data block 3. This is also an allocation method. All possible allocations are recorded as at least one computing power allocation scheme. The allocation method is similar for subsequent secondary allocation schemes.

[0030] Forming at least one secondary allocation scheme for the local computing power values ​​allocated to local data blocks includes the following steps: The defense steps in the defense scheme against network attacks are to evenly divide the local computing power value allocated to a local data block into at least one secondary computing power value and randomly allocate the secondary computing power value to the corresponding local data block. Each matching method of the secondary computing power value is recorded as at least one secondary allocation scheme.

[0031] Reference Figure 5 As shown, the calculation of the data loss coefficient of the computing power allocation scheme includes the following steps: The total computing power consumption of the local data block is obtained by summing the computing power consumption values ​​of defense schemes against at least one network attack acting on the local data block. If the local computing power value allocated to a local data block in the computing power allocation scheme is not lower than the total computing power consumption value of the local data block, then the loss value of the local data block with respect to the computing power allocation scheme is 0. Otherwise, in the secondary allocation scheme, the secondary computing power values ​​allocated to the defense step are summed to obtain the total computing power allocation value for the defense step; If the total computing power allocation value of the defense step is less than the computing power value of the defense step, then the defense step will be regarded as an abnormal defense step. Subtracting the total computing power allocation of the anomaly defense steps from the computing power value of the anomaly defense steps yields the computing power gap of the anomaly defense steps. The computing power gaps of the anomaly defense steps for local data blocks are superimposed to obtain the total computing power gap for the local data blocks; The ratio of the total computing power gap of a local data block to the total computing power consumption of the local data block is obtained by dividing the total computing power gap of the local data block. The total loss of a local data block is obtained by summing up the data loss caused by network attacks on local data blocks. The data loss coefficient of the secondary allocation scheme is obtained by multiplying the importance coefficient of the local data block, the total loss of the local data block, and the gap ratio of the local data block. The average data loss coefficients of at least one secondary allocation scheme for local data blocks generated under the condition of implementing the computing power allocation scheme are taken to obtain the loss value of the local data block with respect to the computing power allocation scheme. The data loss coefficient of the computing power allocation scheme is obtained by superimposing the loss values ​​of at least one local data block with respect to the computing power allocation scheme.

[0032] Here, if the local computing power allocated to a local data block in the computing power allocation scheme is not lower than the total computing power consumption of the local data block, it means that the local computing power allocated to the local data block is sufficient to defend against all network attacks acting on the local data block. Because the defense is sufficient, no data loss will occur. However, if the local computing power allocated to a local data block in the computing power allocation scheme is lower than the total computing power consumption of the local data block, then insufficient computing power will inevitably lead to a gap in the defense, that is, some defense steps cannot be implemented, and thus, some data will be lost. Since this is a case of insufficient computing power, it is necessary to estimate the defense effect of all allocation scenarios and combine their effects. Here, at least one secondary allocation is required. The scheme can approximately represent all allocation scenarios. Since the total loss of a local data block caused by at least one network attack targeting that local data block can be calculated, the proportion of the gap can be obtained by calculating the computing power gap. Since the allocation relationship between computing power and defense steps in the secondary allocation scheme is determined, the existing gap can be calculated. Thus, the data loss coefficient of the secondary allocation scheme is obtained by multiplying the importance coefficient of the local data block, the total loss of the local data block, and the proportion of the gap of the local data block. Here, the importance of the data is taken into account, that is, the more important the data, the greater the data loss coefficient. It is easy to see that the result obtained by this calculation method is consistent with common sense. Therefore, the calculation result is reliable.

[0033] Furthermore, this solution also proposes a storage medium on which a computer-readable program is stored, which, when invoked, executes the aforementioned big data-based dynamic scheduling service system for computer resources.

[0034] It is understandable that the storage medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid-state drive (SSD).

[0035] In summary, the advantages of this invention are as follows: by setting up a coefficient calculation module, a defense generation module, a scheme formation module, and a computing power scheduling module, the importance of data is identified, and multiple allocation methods are formed according to different possibilities of computing power allocation. The data loss under each allocation method is estimated. Thus, based on the estimated situation, the computing power allocation scheme can be screened, thereby enabling reasonable allocation of computing power even when the defense against attacks is insufficient, ensuring that the overall impact of data loss is minimized.

[0036] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A big data-based computer resource dynamic scheduling service system, characterized in that, include: The computing power classification module updates the redundant computing power values ​​that are not called in the computer in real time at preset intervals, forming the computing power demand fluctuation value of computer data processing, and dividing the redundant computing power values ​​into a first computing power value and a second computing power value. The coefficient calculation module divides the data stored in the computer into at least one local data block, performs importance analysis on the data in the local data block, and obtains the importance coefficient of the local data block. A defense generation module identifies at least one network attack on a local data block and, based on historical data, formulates a defense scheme against the network attack. The data analysis module analyzes the computing power consumption of the defense scheme and the amount of data loss caused by the network attack. The scheme forming module forms at least one computing power allocation scheme for a local data block based on a second computing power value. Under the condition of implementing the computing power allocation scheme, at least one secondary allocation scheme is formed for the local computing power value allocated to the local data block. Based on the secondary allocation scheme, the data loss coefficient of the computing power allocation scheme is calculated. The computing power scheduling module selects the computing power allocation scheme with the smallest data loss coefficient as the target allocation scheme, and schedules the computer according to the local computing power value allocated to the local data block in the target allocation scheme and the second computing power value of the computer. 2.The big data based computer resource dynamic scheduling service system according to claim 1, wherein, The process of generating the computing power demand fluctuation value for computer data processing and dividing the redundant computing power value into a first computing power value and a second computing power value includes the following steps: Obtain at least one historical data processing status, the duration of which is equal to a preset time. In the historical data processing status, obtain the historical maximum and minimum values ​​of computing power usage. The difference between the historical maximum and minimum values ​​is used to obtain the historical fluctuation value. The maximum value of the historical fluctuation value is used as the computing power demand fluctuation value. The value of computing power demand fluctuation is used as the first computing power value, and the result of subtracting the first computing power value from the redundant computing power value is used as the second computing power value.

3. The computer resource dynamic scheduling service system based on big data according to claim 2, characterized in that, The process of performing importance analysis on data in a local data block to obtain the importance coefficient of the local data block includes the following steps: Obtain at least one historical data call scenario, and use the data called in the historical data call scenario as a historical data packet; The historical data retrieval scenario corresponding to the historical data packet that overlaps with the local data block is used as the target data retrieval scenario; The overlapping portion of historical data packets and local data blocks in the target data retrieval scenario is taken as the target portion; The total frequency of all historical data call scenarios is taken as the first frequency, and the total frequency of the target data call scenarios is taken as the second frequency. Divide the second frequency by the first frequency to obtain the weight of the target data call scenario. Multiply the total cost of the project applied to the target data call scenario by the weight of the target data call scenario to obtain the application value of the target data call scenario. The proportion of the target portion to the historical data packets of the target data call scenario is taken as the first proportion. The application value of the target data call scenario is multiplied by the first proportion to obtain the data value of the target portion. The proportion of the target portion to the local data block is used as the second proportion. The data value of the target portion is multiplied by the second proportion to obtain the data contribution of the target portion. The data contributions of the target portion are accumulated to obtain the importance coefficient of the local data block.

4. The computer resource dynamic scheduling service system based on big data according to claim 3, characterized in that, The identification of at least one network attack affecting a local data block includes the following steps: In big data, at least one historical attack is obtained, and the attack on a local data block is compared with the historical attack. The historical attack that attacked the local data block is regarded as a network attack.

5. The computer resource dynamic scheduling service system based on big data according to claim 4, characterized in that, The network attack defense plan based on historical data includes the following steps: Historical data will be used to summarize defense measures against cyberattacks and form a cyberattack defense plan.

6. The computer resource dynamic scheduling service system based on big data according to claim 5, characterized in that, The analysis to obtain the computational cost of the defense scheme includes the following steps: The defense scheme is evenly divided into at least one defense block, and the data from the network attacks defended by the defense block are aggregated as the feature data set of the defense block. Defense blocks with the same feature data set are merged to obtain defense steps, and the computing power consumed by the defense steps is used as the computing power value of the defense steps. During the course of a cyberattack, at least one consecutive time interval of the defensive steps is combined to form the time range of the defensive steps. During the course of a cyberattack, at least one time point is taken evenly, and the time ranges containing these time points are summarized into a set of time ranges of action of the time points. The computing power values ​​of the defense steps corresponding to the effective time ranges in the set of effective time ranges are summed to obtain the total computing power value of the time points. The maximum value of the total computing power value of at least one time point is taken as the computing power consumption value of the defense scheme.

7. A computer resource dynamic scheduling service system based on big data according to claim 6, characterized in that, The analysis to determine the amount of data loss caused by the cyberattack includes the following steps: In big data, we obtain historical attack data of at least one network attack under undefended conditions, and take the average of the amount of data destroyed in the historical attack data to obtain the amount of data loss caused by the network attack.

8. The computer resource dynamic scheduling service system based on big data according to claim 7, characterized in that, The at least one computing power allocation scheme for forming local data blocks includes the following steps: The second computing power value is evenly divided into at least one local computing power value, and at least one local computing power value is randomly matched to a local data block. Each matching method of the local computing power value is recorded as at least one computing power allocation scheme.

9. A computer resource dynamic scheduling service system based on big data according to claim 8, characterized in that, The process of forming at least one secondary allocation scheme for the local computing power values ​​allocated to local data blocks includes the following steps: The defense steps in the defense scheme against network attacks are to evenly divide the local computing power value allocated to a local data block into at least one secondary computing power value and randomly allocate the secondary computing power value to the corresponding local data block. Each matching method of the secondary computing power value is recorded as at least one secondary allocation scheme.

10. A computer resource dynamic scheduling service system based on big data according to claim 9, characterized in that, The calculation of the data loss coefficient of the computing power allocation scheme includes the following steps: The total computing power consumption of the local data block is obtained by summing the computing power consumption values ​​of defense schemes against at least one network attack acting on the local data block. If the local computing power value allocated to a local data block in the computing power allocation scheme is not lower than the total computing power consumption value of the local data block, then the loss value of the local data block with respect to the computing power allocation scheme is 0. Otherwise, in the secondary allocation scheme, the secondary computing power values ​​allocated to the defense step are summed to obtain the total computing power allocation value for the defense step; If the total computing power allocation value of the defense step is less than the computing power value of the defense step, then the defense step will be regarded as an abnormal defense step. Subtracting the total computing power allocation of the anomaly defense steps from the computing power value of the anomaly defense steps yields the computing power gap of the anomaly defense steps. The computing power gaps of the anomaly defense steps for local data blocks are superimposed to obtain the total computing power gap for the local data blocks; The ratio of the total computing power gap of a local data block to the total computing power consumption of the local data block is obtained by dividing the total computing power gap of the local data block. The total loss of a local data block is obtained by summing up the data loss caused by network attacks on local data blocks. The data loss coefficient of the secondary allocation scheme is obtained by multiplying the importance coefficient of the local data block, the total loss of the local data block, and the gap ratio of the local data block. The average data loss coefficients of at least one secondary allocation scheme for local data blocks generated under the condition of implementing the computing power allocation scheme are taken to obtain the loss value of the local data block with respect to the computing power allocation scheme. The data loss coefficient of the computing power allocation scheme is obtained by superimposing the loss values ​​of at least one local data block with respect to the computing power allocation scheme.

Images