Cloud load balancing service method and system based on cloud platform

Abstract

The present invention provides a cloud load balancing service method and system based on a cloud platform. According to the working status information of the cloud load terminal, it distinguishes whether it is in a working saturation state, and further determines whether the cloud load terminal in the working saturation state needs to transfer the data processing task to other cloud load terminals for processing because it cannot process the new data processing task in time. It also determines the cloud load terminal suitable for receiving the transferred data processing task, so that the data processing task can be processed in time after being transferred, and the computing power of all cloud load terminals is fully deployed to try to ensure that the data processing tasks of the cloud load terminals can be processed in time and accurately, thereby improving the service balance and reliability of the entire cloud platform system.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud platform data processing, and in particular to a cloud load balancing service method and system based on a cloud platform. Background Art

[0002] The cloud platform is a distributed computing system built based on the cloud computing model. The cloud platform can be connected to multiple cloud load terminals at the same time. Each cloud load terminal can independently connect to external terminals for data exchange and independently process computing tasks. The cloud platform can realize the docking and processing of distributed computing tasks of different external terminals, and maximize the guarantee that the computing tasks sent by each external terminal can be processed in a timely and accurate manner. In actual work, the number of computing tasks and the computing workload received by each cloud load terminal connected to the cloud platform in real time are not the same, which makes some cloud load terminals busy while other cloud load terminals are idle, resulting in the inability to fully utilize the computing power of the entire distributed computing system, and some cloud load terminals have service imbalances such as untimely processing of computing tasks, which reduces the task processing efficiency and reliability of the entire distributed computing system. Summary of the invention

[0003] In view of the defects existing in the prior art, the present invention provides a cloud load balancing service method and system based on a cloud platform, which screens out cloud load terminals in a saturated working state according to the working status information of all cloud load terminals, and determines whether the cloud load terminals in a saturated working state need to transfer data processing tasks; selects one from all cloud load terminals that are not in a saturated working state to receive the data processing tasks that need to be transferred; determines to return the processing results of the data processing tasks to the cloud load terminals in a saturated working state according to the current external terminal connection status of the cloud load terminals in a saturated working state, distinguishes whether the cloud load terminals are in a saturated working state according to the working status information of the cloud load terminals, and further determines whether the cloud load terminals in a saturated working state need to transfer the data processing tasks to other cloud load terminals for processing because they cannot process new data processing tasks in time; and further determines the cloud load terminals suitable for receiving the transferred data processing tasks, so that the data processing tasks can be processed in time after being transferred out, fully deploys the computing power of all cloud load terminals, and tries to ensure that the data processing tasks of the cloud load terminals can be processed in time and accurately, thereby improving the service balance and reliability of the entire cloud platform system.

[0004] The present invention provides a cloud load balancing service method based on a cloud platform, comprising:

[0005] Step S1, obtaining the working status information of all connected cloud load terminals through the cloud platform, and screening the cloud load terminals in a saturated working state according to the working status information; judging whether the cloud load terminals in a saturated working state need to transfer data processing tasks according to the data interaction information between the cloud load terminals in a saturated working state and the outside world;

[0006] Step S2: if a cloud load terminal in a saturated state needs to transfer a data processing task, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task according to the data processing task to be transferred; and the data processing task is added to its corresponding task queue to be processed;

[0007] Step S3, determining whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state.

[0008] Further, in the step S1, the working status information of all connected cloud load terminals is obtained through the cloud platform, and cloud load terminals in a working saturation state are screened according to the working status information; and according to the data interaction information between the cloud load terminal in a working saturation state and the outside world, it is determined whether the cloud load terminal in a working saturation state needs to transfer the data processing task, including:

[0009] Extracting the current pending task queue information of all connected cloud load terminals from the operation log of the cloud platform; analyzing and processing the pending task queue information to estimate the total computing volume of the pending tasks of each cloud load terminal;

[0010] If the total computing amount of the tasks to be processed is greater than or equal to the preset computing amount threshold, it is determined that the corresponding cloud load terminal is in a saturated working state;

[0011] Acquire the data stream received by the cloud load terminal in the saturated state from the external terminal, analyze and process the data stream, and determine whether the external terminal sends a data processing task to the cloud load terminal in the saturated state; if so, determine that the cloud load terminal in the saturated state needs to transfer the data processing task sent by the external terminal.

[0012] Further, in step S2, if a cloud load terminal in a saturated working state needs to transfer a data processing task, according to the data processing task to be transferred, one cloud load terminal is selected from all cloud load terminals that are not in a saturated working state to receive the data processing task; and the data processing task is added to its corresponding to-be-processed task queue, including:

[0013] If a cloud load terminal in a saturated state needs to transfer a data processing task, based on the task attribute information of the data processing task to be transferred, and the remaining computing capacity information and the type information of the allowed processing tasks of all cloud load terminals that are not in a saturated state, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task that needs to be transferred;

[0014] A comparison process is performed on all pending tasks currently contained in the pending task queue of the selected cloud load terminal and the data processing task regarding the task calculation amount, and based on the result of the comparison process, the data processing task is added to the corresponding arrangement position of the pending task queue.

[0015] Further, in the step S3, determining whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state includes:

[0016] According to the terminal identity information of all external terminals currently connected to the cloud load terminal in a saturated working state, it is determined whether the cloud load terminal in a saturated working state is currently connected to the external terminal that sends the data processing task;

[0017] If so, the processing result of the data processing task is returned to the cloud load terminal that is in a saturated state; if not, the processing result of the data processing task is not returned to the cloud load terminal that is in a saturated state.

[0018] The present invention also provides a cloud load balancing service system based on a cloud platform, including:

[0019] A cloud load terminal identification and screening module is used to obtain the working status information of all connected cloud load terminals through the cloud platform, and screen out cloud load terminals in a saturated working state according to the working status information;

[0020] A data processing task transfer judgment module is used to judge whether the cloud load terminal in a saturated working state needs to transfer the data processing task according to the data interaction information between the cloud load terminal in a saturated working state and the outside world;

[0021] The data processing task transfer execution module is used to select a cloud load terminal from all cloud load terminals that are not in a saturated state to receive the data processing task according to the data processing task that needs to be transferred; and add the data processing task to its corresponding task queue to be processed;

[0022] The task processing result return control module is used to determine whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state.

[0023] Furthermore, the cloud load terminal identification and screening module is used to obtain the working status information of all connected cloud load terminals through the cloud platform, and screen out the cloud load terminals in a working saturation state according to the working status information, including:

[0024] Extracting the current pending task queue information of all connected cloud load terminals from the operation log of the cloud platform; analyzing and processing the pending task queue information to estimate the total computing volume of the pending tasks of each cloud load terminal;

[0025] If the total computing amount of the tasks to be processed is greater than or equal to the preset computing amount threshold, it is determined that the corresponding cloud load terminal is in a saturated working state;

[0026] The data processing task transfer judgment module is used to judge whether the cloud load terminal in the working saturation state needs to transfer the data processing task according to the data interaction information between the cloud load terminal in the working saturation state and the outside world, including:

[0027] Acquire the data stream received by the cloud load terminal in the saturated state from the external terminal, analyze and process the data stream, and determine whether the external terminal sends a data processing task to the cloud load terminal in the saturated state; if so, determine that the cloud load terminal in the saturated state needs to transfer the data processing task sent by the external terminal.

[0028] Further, the data processing task transfer execution module is used to select a cloud load terminal from all cloud load terminals that are not in a saturated state to receive the data processing task according to the data processing task to be transferred as needed; and add the data processing task to its corresponding task queue to be processed, including:

[0029] If a cloud load terminal in a saturated state needs to transfer a data processing task, based on the task attribute information of the data processing task to be transferred, and the remaining computing capacity information and the type information of the allowed processing tasks of all cloud load terminals that are not in a saturated state, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task that needs to be transferred;

[0030] A comparison process is performed on all pending tasks currently contained in the pending task queue of the selected cloud load terminal and the data processing task regarding the task calculation amount, and based on the result of the comparison process, the data processing task is added to the corresponding arrangement position of the pending task queue.

[0031] Further, according to the task attribute information of the data processing task to be transferred, and the remaining computing capacity information of all cloud load terminals that are not in a saturated working state and the type information of the tasks allowed to be processed, a cloud load terminal is selected from all cloud load terminals that are not in a saturated working state to receive the data processing task that needs to be transferred, including:

[0032] Step A1, using the following formula (1), based on the task attribute information of the data processing task to be transferred, the type information of the tasks allowed to be processed by all cloud load terminals that are not in a saturated state is screened, and the cloud load terminals that are not in a saturated state that meet the type of data processing task to be transferred are obtained.

[0033]

[0034] In the above formula (1), E(i) represents the primary screening control value of the i-th cloud load terminal that is not in a saturated working state; Z 2 Indicates the binary form of the data to be transferred; P(Z 2 ) represents the binary form of the task type information of the data to be transferred; P(i_k) represents the binary form of the kth type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process; n represents the total number of type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process;

[0035] If E(i) = 1, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out once;

[0036] If E(i) = 0, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out;

[0037] Step A2, first, the cloud load terminal in the saturated working state where the data to be transferred is currently located sends a string of test data to the cloud load terminals that are not in the saturated working state after the first screening. Each cloud load terminal that receives the test data returns the time when the data is currently received. The following formula (2) is used to obtain the data processing transfer speed of each cloud load terminal that is not in the saturated working state after the first screening according to the return status of each cloud load terminal that is not in the saturated working state after the first screening.

[0038]

[0039] In the above formula (2), V(a) represents the data processing transfer speed of the ath cloud load terminal that is not in a saturated working state. 2 (a) represents the binary form of the time data returned by the ath cloud load terminal that is not in a saturated working state after screening; [] 10 Indicates converting the data in brackets into decimal form; t 0 (a) represents the time when the cloud load terminal in the saturated state where the data to be transferred is currently located sends a string of verification data to the ath cloud load terminal that is not in the saturated state selected once; t e (a) represents the time when the cloud load terminal in the saturated state where the data to be transferred is currently located receives the return information of the ath cloud load terminal that is not in the saturated state after screening; J 2 Represents the binary form of a string of test data; len() and len[] both represent the number of bits of the binary data in brackets;

[0040] Step A3, using the following formula (3), a secondary screening is performed based on the data processing transfer speed of each cloud load terminal that is not in a saturated working state after the primary screening and the remaining computing capacity information of each cloud load terminal that is not in a saturated working state after the primary screening, and an optimal cloud load terminal is selected to receive the data processing task that needs to be transferred.

[0041]

[0042] In the above formula (3), A indicates that the optimal cloud load terminal is the Ath cloud load terminal that is not in a saturated state after screening; D(a) indicates the maximum data processing capacity that can be carried by the remaining computing capacity of the ath cloud load terminal that is not in a saturated state after screening; d(Z 2 ) represents the data volume of the data processing task that needs to be transferred; m represents the total number of cloud load terminals that are not in a saturated working state after screening; It means that D(a)≥d(Z 2 ), substitute the value of a from 1 to m into the brackets to get the value of a when the maximum value is obtained in the brackets;

[0043] The Ath cloud load terminal that is not in a saturated working state and is screened out in one screening receives the data processing task that needs to be transferred.

[0044] Further, the task processing result return control module is used to determine whether to return the processing result of the data processing task to the cloud load terminal in the working saturation state according to the current external terminal connection state of the cloud load terminal in the working saturation state, including:

[0045] According to the terminal identity information of all external terminals currently connected to the cloud load terminal in a saturated working state, it is determined whether the cloud load terminal in a saturated working state is currently connected to the external terminal that sends the data processing task;

[0046] If so, the processing result of the data processing task is returned to the cloud load terminal that is in a saturated state; if not, the processing result of the data processing task is not returned to the cloud load terminal that is in a saturated state.

[0047] Compared with the prior art, the cloud load balancing service method and system based on the cloud platform screens out cloud load terminals in a saturated working state according to the working status information of all cloud load terminals, and determines whether the cloud load terminals in a saturated working state need to transfer data processing tasks; selects one from all cloud load terminals that are not in a saturated working state to receive the data processing tasks that need to be transferred; determines to return the processing results of the data processing tasks to the cloud load terminals in a saturated working state according to the current external terminal connection status of the cloud load terminals in a saturated working state, distinguishes whether the cloud load terminals are in a saturated working state according to the working status information of the cloud load terminals, and further determines whether the cloud load terminals in a saturated working state need to transfer the data processing tasks to other cloud load terminals for processing because they cannot process new data processing tasks in time; and also determines the cloud load terminals suitable for receiving the transferred data processing tasks, so that the data processing tasks can be processed in time after being transferred out, fully deploys the computing power of all cloud load terminals, and tries to ensure that the data processing tasks of the cloud load terminals can be processed in time and accurately, thereby improving the service balance and reliability of the entire cloud platform system.

[0048] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0049] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A flow chart of a cloud load balancing service method based on a cloud platform provided by the present invention.

[0052] Figure 2 A schematic diagram of the structure of a cloud load balancing service system based on a cloud platform provided by the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] See also Figure 1 , is a flow chart of a cloud load balancing service method based on a cloud platform provided by an embodiment of the present invention. The cloud load balancing service method based on a cloud platform includes:

[0055] Step S1, obtaining the working status information of all connected cloud load terminals through the cloud platform, and filtering out cloud load terminals in a saturated working state according to the working status information; judging whether the cloud load terminals in a saturated working state need to transfer data processing tasks according to the data interaction information between the cloud load terminals in a saturated working state and the outside world;

[0056] Step S2: if a cloud load terminal in a saturated state needs to transfer a data processing task, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task according to the data processing task to be transferred; and the data processing task is added to its corresponding task queue to be processed;

[0057] Step S3, determining whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state.

[0058] The beneficial effects of the above technical scheme are as follows: the cloud load balancing service method based on the cloud platform screens out cloud load terminals in a saturated working state according to the working status information of all cloud load terminals, and determines whether the cloud load terminals in a saturated working state need to transfer data processing tasks; selects one from all cloud load terminals that are not in a saturated working state to receive the data processing tasks that need to be transferred; determines to return the processing results of the data processing tasks to the cloud load terminals in a saturated working state according to the current external terminal connection status of the cloud load terminals in a saturated working state, distinguishes whether the cloud load terminals are in a saturated working state according to the working status information of the cloud load terminals, and further determines whether the cloud load terminals in a saturated working state need to transfer the data processing tasks to other cloud load terminals for processing because they cannot process new data processing tasks in time; and also determines the cloud load terminals suitable for receiving the transferred data processing tasks, so that the data processing tasks can be processed in time after being transferred out, fully deploys the computing power of all cloud load terminals, and tries to ensure that the data processing tasks of the cloud load terminals can be processed in time and accurately, thereby improving the service balance and reliability of the entire cloud platform system.

[0059] Preferably, in step S1, the working status information of all connected cloud load terminals is obtained through the cloud platform, and cloud load terminals in a working saturation state are screened according to the working status information; and according to the data interaction information between the cloud load terminals in a working saturation state and the outside world, it is determined whether the cloud load terminals in a working saturation state need to transfer data processing tasks, including:

[0060] Extract the current pending task queue information of all connected cloud load terminals from the operation log of the cloud platform; analyze and process the pending task queue information to estimate the total computing volume of pending tasks of each cloud load terminal;

[0061] If the total computing amount of the task to be processed is greater than or equal to the preset computing amount threshold, it is determined that the corresponding cloud load terminal is in a saturated working state;

[0062] Obtain the data stream received by the cloud load terminal in the saturated state from the external terminal, analyze and process the data stream, and determine whether the external terminal sends a data processing task to the cloud load terminal in the saturated state; if so, determine that the cloud load terminal in the saturated state needs to transfer the data processing task sent by the external terminal.

[0063] The beneficial effects of the above technical solution are as follows: the distributed cloud platform system includes a cloud platform and several cloud load terminals connected to the cloud platform. Each cloud load terminal can independently connect to an external terminal, receive data processing tasks from the external terminal, and perform independent calculations on the data processing tasks. In addition, the cloud platform, as the central terminal of the entire distributed cloud platform system, can realize the communication connection between different cloud load terminals. The cloud platform monitors the operating status of all cloud load terminals to generate an operation log. By analyzing and processing the operation log of the cloud platform, it is possible to extract the current pending task queue information of each cloud load terminal, which includes the data computing volume information of all data processing tasks not processed by the cloud load terminal. In this way, the total computing volume of pending tasks of each cloud load terminal can be estimated based on the pending task queue information. By comparing the total computing volume of pending tasks with the preset computing volume threshold, it can be determined whether the cloud load terminal is currently in a saturated state, that is, it is unable to receive new data processing tasks. Among them, for different cloud load terminals, the preset computing threshold is different, and the preset computing threshold is related to the computing power of the cloud load terminal itself. The larger the computing rate of the cloud load terminal, the larger the corresponding preset computing threshold. When the cloud load terminal in the saturated working state is connected to the external terminal, the data stream sent by the external terminal is intercepted and analyzed to determine whether the data stream contains the corresponding data processing task identification information. If it does, it is determined that the external terminal sends a data processing task to the cloud load terminal in the saturated working state, so as to accurately identify whether the cloud load terminal in the saturated working state receives a new data processing task.

[0064] Preferably, in step S2, if a cloud load terminal in a saturated working state needs to transfer a data processing task, according to the data processing task to be transferred, one cloud load terminal is selected from all cloud load terminals that are not in a saturated working state to receive the data processing task; and the data processing task is added to its corresponding to-be-processed task queue, including:

[0065] If a cloud load terminal in a saturated state needs to transfer a data processing task, based on the task attribute information of the data processing task to be transferred, and the remaining computing capacity information and the type information of the allowed processing tasks of all cloud load terminals that are not in a saturated state, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task that needs to be transferred;

[0066] A comparison process is performed on all pending tasks currently contained in the pending task queue of the selected cloud load terminal and the data processing task regarding the task calculation amount. Based on the result of the comparison process, the data processing task is added to the corresponding arrangement position of the pending task queue.

[0067] The beneficial effect of the above technical solution is: if a cloud load terminal in a saturated state receives a new data processing task, it is necessary to transfer the received new data processing task to a cloud load terminal that is not in a saturated state, and identify the task attribute information of the data processing task that needs to be transferred, wherein the task attribute information may be but is not limited to the amount of data and the type of data that the data processing task needs to process. Then, the task attribute information of the data processing task that needs to be transferred, as well as the remaining computing capacity information of all cloud load terminals that are not in a saturated state and the type information of the allowed processing tasks are compared, and the cloud load terminal that is not in a saturated state and can be compatible with the data type of the data processing task that needs to be transferred and has the highest remaining computing capacity value is used as the cloud load terminal to which the data processing task needs to be transferred. Then, the task calculation amount of the data processing task that needs to be transferred is arranged from large to small with the task calculation amount of all the tasks to be processed currently contained in the queue of pending tasks of a selected cloud load terminal, so that the data processing task is added to the corresponding arrangement position of the queue of pending tasks, ensuring that the selected cloud load terminal can quickly process the data processing task that needs to be transferred.

[0068] Preferably, in step S3, determining whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state includes:

[0069] According to the terminal identity information of all external terminals currently connected to the cloud load terminal in the saturated working state, it is determined whether the cloud load terminal in the saturated working state is currently connected to the external terminal that sends the data processing task;

[0070] If so, the processing result of the data processing task is returned to the cloud load terminal that is in a saturated state; if not, the processing result of the data processing task is not returned to the cloud load terminal that is in a saturated state.

[0071] The beneficial effect of the above technical solution is: through the above method, it is determined whether the cloud load terminal in the work saturation state is currently connected to the external terminal that sends the data processing task. If so, the processing result of the data processing task is returned to the cloud load terminal in the work saturation state. In this way, the cloud load terminal in the work saturation state directly sends the processing result back to the corresponding external terminal, ensuring that the external terminal receives the processing result of the data processing task sent previously in time, thereby improving the service balance and reliability of the entire cloud platform system.

[0072] See also Figure 2 , is a schematic diagram of the structure of a cloud load balancing service system based on a cloud platform provided by an embodiment of the present invention. The cloud load balancing service method based on a cloud platform includes:

[0073] The cloud load terminal identification and screening module is used to obtain the working status information of all connected cloud load terminals through the cloud platform, and screen out the cloud load terminals in a saturated working state according to the working status information;

[0074] A data processing task transfer judgment module is used to judge whether the cloud load terminal in a saturated working state needs to transfer the data processing task according to the data interaction information between the cloud load terminal in a saturated working state and the outside world;

[0075] The data processing task transfer execution module is used to select a cloud load terminal from all cloud load terminals that are not in a saturated state to receive the data processing task according to the data processing task that needs to be transferred; and add the data processing task to its corresponding task queue to be processed;

[0076] The task processing result return control module is used to determine whether to return the processing result of the data processing task to the cloud load terminal in the work saturation state according to the current external terminal connection state of the cloud load terminal in the work saturation state.

[0077] The beneficial effects of the above technical solution are as follows: the cloud load balancing service system based on the cloud platform screens out cloud load terminals that are in a saturated state according to the working status information of all cloud load terminals, and determines whether the cloud load terminals in a saturated state need to transfer data processing tasks; selects one from all cloud load terminals that are not in a saturated state to receive the data processing tasks that need to be transferred; determines to return the processing results of the data processing tasks to the cloud load terminals that are in a saturated state according to the current external terminal connection status of the cloud load terminals that are in a saturated state, distinguishes whether they are in a saturated state according to the working status information of the cloud load terminals, and further determines whether the cloud load terminals in a saturated state need to transfer the data processing tasks to other cloud load terminals for processing because they cannot process new data processing tasks in time; and also determines the cloud load terminals that are suitable for receiving the transferred data processing tasks, so that the data processing tasks can be processed in time after being transferred out, fully deploys the computing power of all cloud load terminals, and tries to ensure that the data processing tasks of the cloud load terminals can be processed in a timely and accurate manner, thereby improving the service balance and reliability of the entire cloud platform system.

[0078] Preferably, the cloud load terminal identification and screening module is used to obtain the working status information of all connected cloud load terminals through the cloud platform, and screen out cloud load terminals in a working saturation state according to the working status information, including:

[0079] Extract the current pending task queue information of all connected cloud load terminals from the operation log of the cloud platform; analyze and process the pending task queue information to estimate the total computing volume of pending tasks of each cloud load terminal;

[0080] If the total computing amount of the task to be processed is greater than or equal to the preset computing amount threshold, it is determined that the corresponding cloud load terminal is in a saturated working state;

[0081] The data processing task transfer judgment module is used to judge whether the cloud load terminal in the working saturation state needs to transfer the data processing task according to the data interaction information between the cloud load terminal in the working saturation state and the outside world, including:

[0082] Obtain the data stream received by the cloud load terminal in the saturated state from the external terminal, analyze and process the data stream, and determine whether the external terminal sends a data processing task to the cloud load terminal in the saturated state; if so, determine that the cloud load terminal in the saturated state needs to transfer the data processing task sent by the external terminal.

[0083] The beneficial effects of the above technical solution are as follows: the distributed cloud platform system includes a cloud platform and several cloud load terminals connected to the cloud platform. Each cloud load terminal can independently connect to an external terminal, receive data processing tasks from the external terminal, and perform independent calculations on the data processing tasks. In addition, the cloud platform, as the central terminal of the entire distributed cloud platform system, can realize the communication connection between different cloud load terminals. The cloud platform monitors the operating status of all cloud load terminals to generate an operation log. By analyzing and processing the operation log of the cloud platform, it is possible to extract the current pending task queue information of each cloud load terminal, which includes the data computing volume information of all data processing tasks not processed by the cloud load terminal. In this way, the total computing volume of pending tasks of each cloud load terminal can be estimated based on the pending task queue information. By comparing the total computing volume of pending tasks with the preset computing volume threshold, it can be determined whether the cloud load terminal is currently in a saturated state, that is, it is unable to receive new data processing tasks. Among them, for different cloud load terminals, the preset computing threshold is different, and the preset computing threshold is related to the computing power of the cloud load terminal itself. The larger the computing rate of the cloud load terminal, the larger the corresponding preset computing threshold. When the cloud load terminal in the saturated working state is connected to the external terminal, the data stream sent by the external terminal is intercepted and analyzed to determine whether the data stream contains the corresponding data processing task identification information. If it does, it is determined that the external terminal sends a data processing task to the cloud load terminal in the saturated working state, so as to accurately identify whether the cloud load terminal in the saturated working state receives a new data processing task.

[0084] Preferably, the data processing task transfer execution module is used to select a cloud load terminal from all cloud load terminals that are not in a saturated working state to receive the data processing task according to the data processing task to be transferred as needed; and add the data processing task to its corresponding to-be-processed task queue, including:

[0085] If a cloud load terminal in a saturated state needs to transfer a data processing task, based on the task attribute information of the data processing task to be transferred, and the remaining computing capacity information and the type information of the allowed processing tasks of all cloud load terminals that are not in a saturated state, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task that needs to be transferred;

[0086] A comparison process is performed on all pending tasks currently contained in the pending task queue of the selected cloud load terminal and the data processing task regarding the task calculation amount. Based on the result of the comparison process, the data processing task is added to the corresponding arrangement position of the pending task queue.

[0087] The beneficial effect of the above technical solution is: if a cloud load terminal in a saturated state receives a new data processing task, it is necessary to transfer the received new data processing task to a cloud load terminal that is not in a saturated state, and identify the task attribute information of the data processing task that needs to be transferred, wherein the task attribute information may be but is not limited to the amount of data and the type of data that the data processing task needs to process. Then, the task attribute information of the data processing task that needs to be transferred, as well as the remaining computing capacity information of all cloud load terminals that are not in a saturated state and the type information of the allowed processing tasks are compared, and the cloud load terminal that is not in a saturated state and can be compatible with the data type of the data processing task that needs to be transferred and has the highest remaining computing capacity value is used as the cloud load terminal to which the data processing task needs to be transferred. Then, the task calculation amount of the data processing task that needs to be transferred is arranged from large to small with the task calculation amount of all the tasks to be processed currently contained in the queue of pending tasks of a selected cloud load terminal, so that the data processing task is added to the corresponding arrangement position of the queue of pending tasks, ensuring that the selected cloud load terminal can quickly process the data processing task that needs to be transferred.

[0088] Preferably, according to the task attribute information of the data processing task to be transferred, and the remaining computing capacity information of all cloud load terminals that are not in a saturated working state and the type information of the tasks allowed to be processed, selecting a cloud load terminal from all cloud load terminals that are not in a saturated working state to receive the data processing task that needs to be transferred includes:

[0089] Step A1, using the following formula (1), based on the task attribute information of the data processing task to be transferred, the type information of the tasks allowed to be processed by all cloud load terminals that are not in a saturated state is screened, and the cloud load terminals that are not in a saturated state that meet the type of data processing task to be transferred are obtained.

[0090]

[0091] In the above formula (1), E(i) represents the primary screening control value of the i-th cloud load terminal that is not in a saturated working state; Z 2 Indicates the binary form of the data to be transferred; P(Z 2 ) represents the binary form of the task type information of the data to be transferred; P(i_k) represents the binary form of the kth type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process; n represents the total number of type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process;

[0092] If E(i) = 1, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out once;

[0093] If E(i) = 0, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out;

[0094] Step A2, first, the cloud load terminal in the saturated working state where the data to be transferred is currently located sends a string of test data to the cloud load terminals that are not in the saturated working state after the first screening. Each cloud load terminal that receives the test data returns the time when the data is currently received. The following formula (2) is used to obtain the data processing transfer speed of each cloud load terminal that is not in the saturated working state after the first screening according to the return status of each cloud load terminal that is not in the saturated working state after the first screening.

[0095]

[0096] In the above formula (2), V(a) represents the data processing transfer speed of the ath cloud load terminal that is not in a saturated working state. 2 (a) represents the binary form of the time data returned by the ath cloud load terminal that is not in a saturated working state after screening; [] 10 Indicates converting the data in brackets into decimal form; t 0(a) represents the time when the cloud load terminal in the saturated state where the data to be transferred is currently located sends a string of verification data to the ath cloud load terminal that is not in the saturated state selected once; t e (a) represents the time when the cloud load terminal in the saturated state where the data to be transferred is currently located receives the return information of the ath cloud load terminal that is not in the saturated state after screening; J 2 Represents the binary form of a string of test data; len() and len[] both represent the number of bits of the binary data in brackets;

[0097] Step A3, using the following formula (3), a secondary screening is performed based on the data processing transfer speed of each cloud load terminal that is not in a saturated working state after the primary screening and the remaining computing capacity information of each cloud load terminal that is not in a saturated working state after the primary screening, and an optimal cloud load terminal is selected to receive the data processing task that needs to be transferred.

[0098]

[0099] In the above formula (3), A indicates that the optimal cloud load terminal is the Ath cloud load terminal that is not in a saturated state after screening; D(a) indicates the maximum data processing capacity that can be carried by the remaining computing capacity of the ath cloud load terminal that is not in a saturated state after screening; d(Z 2 ) represents the data volume of the data processing task that needs to be transferred; m represents the total number of cloud load terminals that are not in a saturated working state after screening; It means that D(a)≥d(Z 2 ), substitute the value of a from 1 to m into the brackets to get the value of a when the maximum value is obtained in the brackets;

[0100] The Ath cloud load terminal that is not in a saturated working state and is screened out in one screening receives the data processing task that needs to be transferred.

[0101] The beneficial effects of the above technical solution are as follows: using the above formula (1), based on the task attribute information of the data processing task to be transferred, the type information of the tasks allowed to be processed by all cloud load terminals that are not in a saturated state is screened once, and cloud load terminals that are not in a saturated state and meet the type of data processing tasks that need to be transferred are obtained, thereby shortening the screening range and improving the work efficiency of the system; then using the above formula (2), based on the return status of each cloud load terminal that is not in a saturated state after a screening, the data processing transfer speed of each cloud load terminal that is not in a saturated state after a screening is obtained, thereby knowing the data processing transfer speed of each cloud load terminal that is not in a saturated state after a screening. The situation of each cloud load terminal that is not in a saturated state during the transfer is convenient for further preferential screening; finally, the above formula (3) is used to perform a secondary screening based on the data processing transfer speed of each cloud load terminal that is not in a saturated state after the first screening and the remaining computing capacity information of each cloud load terminal that is not in a saturated state after the first screening, and an optimal cloud load terminal is selected to receive the data processing task that needs to be transferred, thereby selecting a cloud load terminal that is not in a saturated state and has sufficient space to process the data volume of the data processing task that needs to be transferred and has a faster transmission speed to receive the data, thereby ensuring the reliability of the system.

[0102] Preferably, the task processing result return control module is used to determine whether to return the processing result of the data processing task to the cloud load terminal in the working saturation state according to the current external terminal connection state of the cloud load terminal in the working saturation state, including:

[0103] According to the terminal identity information of all external terminals currently connected to the cloud load terminal in the saturated working state, it is determined whether the cloud load terminal in the saturated working state is currently connected to the external terminal that sends the data processing task;

[0104] If so, the processing result of the data processing task is returned to the cloud load terminal that is in a saturated state; if not, the processing result of the data processing task is not returned to the cloud load terminal that is in a saturated state.

[0105] The beneficial effect of the above technical solution is: through the above method, it is determined whether the cloud load terminal in the work saturation state is currently connected to the external terminal that sends the data processing task. If so, the processing result of the data processing task is returned to the cloud load terminal in the work saturation state. In this way, the cloud load terminal in the work saturation state directly sends the processing result back to the corresponding external terminal, ensuring that the external terminal receives the processing result of the data processing task sent previously in time, thereby improving the service balance and reliability of the entire cloud platform system.

[0106] It can be seen from the contents of the above embodiments that the cloud load balancing service method and system based on the cloud platform screens out cloud load terminals in a saturated working state according to the working status information of all cloud load terminals, and determines whether the cloud load terminals in a saturated working state need to transfer data processing tasks; selects one from all cloud load terminals that are not in a saturated working state to receive the data processing tasks that need to be transferred; determines to return the processing results of the data processing tasks to the cloud load terminals in a saturated working state according to the current external terminal connection status of the cloud load terminals in a saturated working state, distinguishes whether the cloud load terminals are in a saturated working state according to the working status information of the cloud load terminals, and further determines whether the cloud load terminals in a saturated working state need to transfer the data processing tasks to other cloud load terminals for processing because they cannot process new data processing tasks in time; and also determines the cloud load terminals suitable for receiving the transferred data processing tasks, so that the data processing tasks can be processed in time after being transferred out, fully deploys the computing power of all cloud load terminals, and tries to ensure that the data processing tasks of the cloud load terminals can be processed in time and accurately, thereby improving the service balance and reliability of the entire cloud platform system.

[0107] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A cloud load balancing service method based on a cloud platform, characterized in that: include: Step S1, obtaining the working status information of all connected cloud load terminals through the cloud platform, and screening the cloud load terminals in a saturated working state according to the working status information; judging whether the cloud load terminals in a saturated working state need to transfer data processing tasks according to the data interaction information between the cloud load terminals in a saturated working state and the outside world; Step S2: if a cloud load terminal in a saturated state needs to transfer a data processing task, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task according to the data processing task to be transferred; and the data processing task is added to its corresponding task queue to be processed; Step S3, determining whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state; Wherein, in the step S2, if the cloud load terminal in the saturated working state needs to transfer the data processing task, according to the data processing task to be transferred, one cloud load terminal is selected from all cloud load terminals that are not in the saturated working state to receive the data processing task; and the data processing task is added to its corresponding to-be-processed task queue, including: If a cloud load terminal in a saturated state needs to transfer a data processing task, based on the task attribute information of the data processing task to be transferred, and the remaining computing capacity information and the type information of the allowed processing tasks of all cloud load terminals that are not in a saturated state, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task that needs to be transferred; Performing a comparison process on the task calculation amount of all pending tasks currently contained in the pending task queue of the selected cloud load terminal and the data processing task, and adding the data processing task to the corresponding arrangement position of the pending task queue according to the result of the comparison process; According to the task attribute information of the data processing task to be transferred, and the remaining computing capacity information of all cloud load terminals that are not in a saturated working state and the type information of the tasks allowed to be processed, a cloud load terminal is selected from all cloud load terminals that are not in a saturated working state to receive the data processing task that needs to be transferred, including: Step A1, using the following formula (1), based on the task attribute information of the data processing task to be transferred, the type information of the tasks allowed to be processed by all cloud load terminals that are not in a saturated state is screened, and the cloud load terminals that are not in a saturated state that meet the type of data processing task to be transferred are obtained. In the above formula (1), E(i) represents the primary screening control value of the i-th cloud load terminal that is not in a saturated working state; Z2 represents the binary form of the data to be transferred; P(Z2) represents the binary form of the task type information of the data to be transferred; P(i_k) represents the binary form of the k-th type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process; n represents the total number of type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process; If E(i) = 1, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out once; If E(i) = 0, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out; Step A2, first, the cloud load terminal in the saturated working state where the data to be transferred is currently located sends a string of test data to the cloud load terminals that are not in the saturated working state after the first screening. Each cloud load terminal that receives the test data returns the time when the data is currently received. The following formula (2) is used to obtain the data processing transfer speed of each cloud load terminal that is not in the saturated working state after the first screening according to the return status of each cloud load terminal that is not in the saturated working state after the first screening. In the above formula (2), V(a) represents the data processing transfer speed of the a-th cloud load terminal that is not in a saturated working state selected by the first screening; t2(a) represents the binary form of the time data returned by the a-th cloud load terminal that is not in a saturated working state selected by the first screening; [] 10 Indicates that the data in the brackets are converted into decimal data; t0(a) indicates the time when the cloud load terminal in the saturated state where the data to be transferred is currently located sends a string of inspection data to the ath cloud load terminal that is not in the saturated state selected at one time; t e (a) indicates the time when the cloud load terminal in the saturated state where the data to be transferred is currently located receives the return information of the ath cloud load terminal that is not in the saturated state after screening; J2 indicates the binary form of a string of test data; len() and len[] both indicate the number of bits of the binary data in the brackets; Step A3, using the following formula (3), a secondary screening is performed based on the data processing transfer speed of each cloud load terminal that is not in a saturated working state after the primary screening and the remaining computing capacity information of each cloud load terminal that is not in a saturated working state after the primary screening, and an optimal cloud load terminal is selected to receive the data processing task that needs to be transferred. In the above formula (3), A indicates that the optimal cloud load terminal is the Ath cloud load terminal that is not in a saturated state after screening; D(a) indicates the maximum data processing capacity that can be carried by the remaining computing capacity of the ath cloud load terminal that is not in a saturated state after screening; d(Z2) indicates the data volume of the data processing task that needs to be transferred; m indicates the total number of cloud load terminals that are not in a saturated state after screening; It means that under the condition that a satisfies D(a)≥d(Z2), substitute the value of a from 1 to m into the brackets to obtain the value of a when the maximum value is obtained in the brackets; The Ath cloud load terminal that is not in a saturated working state and is screened out in one screening receives the data processing task that needs to be transferred.

2. The cloud load balancing service method based on a cloud platform according to claim 1, characterized in that: In the step S1, the working status information of all connected cloud load terminals is obtained through the cloud platform, and cloud load terminals in a saturated working state are screened according to the working status information; and according to the data interaction information between the cloud load terminal in a saturated working state and the outside world, it is determined whether the cloud load terminal in a saturated working state needs to transfer the data processing task, including: Extracting the current pending task queue information of all connected cloud load terminals from the operation log of the cloud platform; analyzing and processing the pending task queue information to estimate the total computing volume of the pending tasks of each cloud load terminal; If the total computing amount of the tasks to be processed is greater than or equal to the preset computing amount threshold, it is determined that the corresponding cloud load terminal is in a saturated working state; Acquire the data stream received by the cloud load terminal in the saturated state from the external terminal, analyze and process the data stream, and determine whether the external terminal sends a data processing task to the cloud load terminal in the saturated state; if so, determine that the cloud load terminal in the saturated state needs to transfer the data processing task sent by the external terminal.

3. The cloud load balancing service method based on a cloud platform according to claim 1, characterized in that: In the step S3, determining whether to return the processing result of the data processing task to the cloud load terminal in the saturated working state according to the current external terminal connection state of the cloud load terminal in the saturated working state includes: According to the terminal identity information of all external terminals currently connected to the cloud load terminal in a saturated working state, it is determined whether the cloud load terminal in a saturated working state is currently connected to the external terminal that sends the data processing task; If so, the processing result of the data processing task is returned to the cloud load terminal that is in a saturated state; if not, the processing result of the data processing task is not returned to the cloud load terminal that is in a saturated state.

4. A cloud load balancing service system based on a cloud platform, characterized in that: include: A cloud load terminal identification and screening module is used to obtain the working status information of all connected cloud load terminals through the cloud platform, and screen out cloud load terminals in a saturated working state according to the working status information; A data processing task transfer judgment module is used to judge whether the cloud load terminal in a saturated working state needs to transfer the data processing task according to the data interaction information between the cloud load terminal in a saturated working state and the outside world; The data processing task transfer execution module is used to select a cloud load terminal from all cloud load terminals that are not in a saturated working state to receive the data processing task according to the data processing task to be transferred as needed; and add the data processing task to its corresponding to-be-processed task queue; the task processing result return control module is used to determine whether to return the processing result of the data processing task to the cloud load terminal in a saturated working state according to the current external terminal connection state of the cloud load terminal in a saturated working state; The data processing task transfer execution module is used to select a cloud load terminal from all cloud load terminals that are not in a saturated working state to receive the data processing task according to the data processing task to be transferred as needed; and add the data processing task to its corresponding to-be-processed task queue, including: If a cloud load terminal in a saturated state needs to transfer a data processing task, based on the task attribute information of the data processing task to be transferred, and the remaining computing capacity information and the type information of the allowed processing tasks of all cloud load terminals that are not in a saturated state, a cloud load terminal is selected from all cloud load terminals that are not in a saturated state to receive the data processing task that needs to be transferred; Performing a comparison process on the task calculation amount of all pending tasks currently contained in the pending task queue of the selected cloud load terminal and the data processing task, and adding the data processing task to the corresponding arrangement position of the pending task queue according to the result of the comparison process; According to the task attribute information of the data processing task to be transferred, and the remaining computing capacity information of all cloud load terminals that are not in a saturated working state and the type information of the tasks allowed to be processed, a cloud load terminal is selected from all cloud load terminals that are not in a saturated working state to receive the data processing task that needs to be transferred, including: Step A1, using the following formula (1), based on the task attribute information of the data processing task to be transferred, the type information of the tasks allowed to be processed by all cloud load terminals that are not in a saturated state is screened, and the cloud load terminals that are not in a saturated state that meet the type of data processing task to be transferred are obtained. In the above formula (1), E(i) represents the primary screening control value of the i-th cloud load terminal that is not in a saturated working state; Z2 represents the binary form of the data to be transferred; P(Z2) represents the binary form of the task type information of the data to be transferred; P(i_k) represents the binary form of the k-th type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process; n represents the total number of type information of the task that the i-th cloud load terminal that is not in a saturated working state is allowed to process; If E(i) = 1, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out once; If E(i) = 0, it means that the i-th cloud load terminal that is not in a saturated working state needs to be screened out; Step A2, first, the cloud load terminal in the saturated working state where the data to be transferred is currently located sends a string of test data to the cloud load terminals that are not in the saturated working state after the first screening. Each cloud load terminal that receives the test data returns the time when the data is currently received. The following formula (2) is used to obtain the data processing transfer speed of each cloud load terminal that is not in the saturated working state after the first screening according to the return status of each cloud load terminal that is not in the saturated working state after the first screening. In the above formula (2), V(a) represents the data processing transfer speed of the a-th cloud load terminal that is not in a saturated working state selected by the first screening; t2(a) represents the binary form of the time data returned by the a-th cloud load terminal that is not in a saturated working state selected by the first screening; [] 10 Indicates that the data in the brackets are converted into decimal data; t0(a) indicates the time when the cloud load terminal in the saturated state where the data to be transferred is currently located sends a string of inspection data to the ath cloud load terminal that is not in the saturated state selected at one time; t e (a) indicates the time when the cloud load terminal in the saturated state where the data to be transferred is currently located receives the return information of the ath cloud load terminal that is not in the saturated state after screening; J2 indicates the binary form of a string of test data; len() and len[] both indicate the number of bits of the binary data in the brackets; Step A3, using the following formula (3), a secondary screening is performed based on the data processing transfer speed of each cloud load terminal that is not in a saturated working state after the primary screening and the remaining computing capacity information of each cloud load terminal that is not in a saturated working state after the primary screening, and an optimal cloud load terminal is selected to receive the data processing task that needs to be transferred. In the above formula (3), A indicates that the optimal cloud load terminal is the Ath cloud load terminal that is not in a saturated state after screening; D(a) indicates the maximum data processing capacity that can be carried by the remaining computing capacity of the ath cloud load terminal that is not in a saturated state after screening; d(Z2) indicates the data volume of the data processing task that needs to be transferred; m indicates the total number of cloud load terminals that are not in a saturated state after screening; It means that under the condition that a satisfies D(a)≥d(Z2), substitute the value of a from 1 to m into the brackets to obtain the value of a when the maximum value is obtained in the brackets; The Ath cloud load terminal that is not in a saturated working state and is screened out in one screening receives the data processing task that needs to be transferred.

5. The cloud load balancing service system based on the cloud platform according to claim 4, characterized in that: The cloud load terminal identification and screening module is used to obtain the working status information of all connected cloud load terminals through the cloud platform, and screen out cloud load terminals in a working saturation state according to the working status information, including: Extracting the current pending task queue information of all connected cloud load terminals from the operation log of the cloud platform; analyzing and processing the pending task queue information to estimate the total computing volume of the pending tasks of each cloud load terminal; If the total computing amount of the tasks to be processed is greater than or equal to the preset computing amount threshold, it is determined that the corresponding cloud load terminal is in a saturated working state; The data processing task transfer judgment module is used to judge whether the cloud load terminal in the working saturation state needs to transfer the data processing task according to the data interaction information between the cloud load terminal in the working saturation state and the outside world, including: Acquire the data stream received by the cloud load terminal in the saturated state from the external terminal, analyze and process the data stream, and determine whether the external terminal sends a data processing task to the cloud load terminal in the saturated state; if so, determine that the cloud load terminal in the saturated state needs to transfer the data processing task sent by the external terminal.

6. The cloud load balancing service system based on a cloud platform as claimed in claim 4, characterized in that: The task processing result return control module is used to determine whether to return the processing result of the data processing task to the cloud load terminal in the working saturation state according to the current external terminal connection state of the cloud load terminal in the working saturation state, including: According to the terminal identity information of all external terminals currently connected to the cloud load terminal in a saturated working state, it is determined whether the cloud load terminal in a saturated working state is currently connected to the external terminal that sends the data processing task; If so, the processing result of the data processing task is returned to the cloud load terminal that is in a saturated state; if not, the processing result of the data processing task is not returned to the cloud load terminal that is in a saturated state.

Images