A real-time response method, device and medium for massive and highly concurrent data of the Internet of Things

Abstract

The present application relates to the technical field of real-time response to massive high-concurrency data in the Internet of Things, and specifically to a real-time response method, device and medium for massive high-concurrency data in the Internet of Things, the method comprising: determining the data queuing rate of each server at the current moment based on the extreme distribution of all class static attributes in each server and the data volume of each process; determining the type adaptability of the waiting data task at the current moment under each server based on the data volume of the waiting data task at the current moment, all class static attributes of each server, and the task type characteristics and the process type characteristics, and responding to the waiting data task at the current moment in combination with the data queuing rate. The present application aims to reasonably allocate tasks to servers with corresponding hardware resources and improve the response speed of the server to massive concurrent data.

Description

Technical Field

[0001] The present application relates to the technical field of real-time response to massive and high-concurrency data in the Internet of Things, and specifically to a method, device and medium for real-time response to massive and high-concurrency data in the Internet of Things. Background Art

[0002] The Internet of Things is a network based on information carriers such as the Internet and traditional telecommunications networks, which allows all ordinary physical objects that can be independently addressed to be interconnected. During the use of the Internet of Things, a large amount of data transmitted from the Internet of Things nodes will be collected through the terminal. Due to the large size of the Internet of Things, when data transmission is at its peak, a large amount of concurrent data will appear, which can easily cause the terminal server to crash.

[0003] In order to prevent massive high-concurrency data from causing terminal servers to crash, multiple servers are usually deployed at the terminal, and the minimum number of connections strategy is used to allocate data tasks to multiple servers to handle massive concurrent data. Since the minimum number of connections strategy does not take into account the performance differences of data servers, simply allocating requests by the number of connections can easily aggravate the load imbalance problem of service nodes and reduce the processing efficiency of massive concurrent data. Therefore, when the IoT terminal responds to massive high-concurrency data, the performance differences of the servers are not taken into account when allocating data tasks, resulting in unreasonable task allocation, which in turn causes the problem of a decrease in the server's response speed to massive concurrent data due to uneven load. Summary of the invention

[0004] In a first aspect, an embodiment of the present application provides a method for real-time response to massive high-concurrency data in an Internet of Things, the method comprising the following steps:

[0005] Get all class static properties of each server, as well as the data volume of each process in each server at the current moment and the IoT node corresponding to each process, and get the processing start time, data volume and all class resource consumption data of each historical data task in each IoT node before the current moment, get the data volume of the data task waiting to be assigned to the server at the current moment, and record the data task waiting to be assigned to the server as the waiting data task;

[0006] Based on the difference between the data volume of each historical data task and the resource consumption data of all types, and the time interval between the processing start time of each historical data task and the current time, determine the task type characteristics of each IoT node at the current time;

[0007] Based on the extreme distribution of static attributes of all classes in each server and the amount of data in each process, determine the data queuing rate of each server at the current moment;

[0008] Based on the task type characteristics of each process in each server corresponding to the IoT node and the data volume of each process in each server, determine the process type characteristics of each server at the current moment;

[0009] Based on the data volume of the waiting data task at the current moment, all class static properties of each server, and the task type characteristics and the process type characteristics, the judgment vector of the waiting data task at the current moment under each process in each server is determined, based on the discrete degree of all elements in the judgment vector under each process, the type adaptability of the waiting data task at the current moment under each server is determined, and combined with the data queuing rate, the task adaptability of the waiting data task at the current moment under each server is determined, and a response is made to the waiting data task at the current moment.

[0010] Preferably, the method for determining the task type characteristics of each IoT node at the current moment is:

[0011] In each IoT node, the ratio of all resource consumption data of each historical data task to the data volume of the corresponding historical data task is calculated, and all the ratios are combined into the task feature vector of each historical data task;

[0012] Calculate the normalized value of the time interval between the processing start time of each historical data task and the current time, and multiply the task feature vector of each historical data task by the normalized value of the time interval as the feature vector of each historical data task;

[0013] The task type feature of each IoT node at the current moment is the sum vector of the feature vectors of all historical data tasks of each IoT node before the current moment.

[0014] Preferably, the expression of the data queuing rate of each server at the current moment is: ; In the formula, When represents the data queuing rate of server r at the previous moment; Indicates the data volume of the sth process in server r at the current moment; Represents the maximum value among all class static attributes of server r; Represents the number of all processes in server r at the current moment.

[0015] Preferably, the method for determining the process type characteristics of each server at the current moment is:

[0016] Number all processes in each server at the current moment, multiply the task type characteristics of the IoT node corresponding to each process in each server by the data volume of the process, and record it as the resource consumption vector of each process in each server at the current moment; multiply the resource consumption vector by the normalized value of the corresponding process number as the process feature vector of each process in each server at the current moment;

[0017] The process type feature of each server at the current moment is the sum vector of the process feature vectors of all processes in each server.

[0018] Preferably, the determination of the judgment vector of the waiting data task at the current moment under each process in each server includes:

[0019] The result of multiplying the data volume of the waiting data task at the current moment by the task type feature number of each IoT node is recorded as the current task feature of the waiting data task at the current moment under each IoT node;

[0020] The sum vector of the process type characteristics of each process in each server and the current task characteristics under the corresponding IoT node is recorded as the current task vector of the waiting data task under each process on each server at the current moment;

[0021] All elements in the current task vector correspond to all class static attributes one by one. The ratios of all elements in the current task vector to the corresponding static attributes in all class static attributes are calculated, and the judgment vectors of the waiting data tasks at the current moment under each process in each server are formed.

[0022] Preferably, the type adaptability of the waiting data task at the current moment in each server is the standard deviation of the normalized values ​​of all elements in the judgment vector of the waiting data task at the current moment in each server.

[0023] Preferably, the expression of the task adaptability of the waiting data task at the current moment in each server is: ; In the formula, Indicates the task adaptability of the waiting data task under server r at the current moment; Indicates the type adaptability of the waiting data task under server r at the current moment; represents the data queuing rate of server r.

[0024] Preferably, the waiting data task at the current moment is responded to:

[0025] At the current moment, for all servers, the server with the smallest task adaptability is selected as the server of the waiting data task at the current moment, and the waiting data task at the current moment is processed.

[0026] In a second aspect, an embodiment of the present application provides a real-time response device for massive and high-concurrency data of the Internet of Things, the device comprising:

[0027] The data acquisition module is used to obtain all class static properties of each server, as well as the data volume of each process in each server at the current moment and the IoT node corresponding to each process, and obtain the processing start time, data volume and all class resource consumption data of each historical data task in each IoT node before the current moment, obtain the data volume of the data task waiting to be assigned to the server at the current moment, and record the data task waiting to be assigned to the server as the waiting data task;

[0028] A data analysis and processing module is used to determine the task type characteristics of each IoT node at the current moment based on the difference between the data volume of each historical data task and the data of all resource consumption types, and the time interval between the processing start time of each historical data task and the current moment;

[0029] Based on the extreme distribution of static attributes of all classes in each server and the amount of data in each process, determine the data queuing rate of each server at the current moment;

[0030] Based on the task type characteristics of each process in each server corresponding to the IoT node and the data volume of each process in each server, determine the process type characteristics of each server at the current moment;

[0031] The data response module is used to determine the judgment vector of the waiting data task at the current moment under each process in each server based on the data volume of the waiting data task at the current moment, all class static properties of each server, and the task type characteristics and the process type characteristics, determine the type adaptability of the waiting data task at the current moment under each server based on the discrete degree of all elements in the judgment vector under each process, and determine the task adaptability of the waiting data task at the current moment under each server in combination with the data queuing rate, and respond to the waiting data task at the current moment.

[0032] In a third aspect, an embodiment of the present application also provides a real-time response medium for massive and high-concurrency data in the Internet of Things, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of a real-time response method for massive and high-concurrency data in the Internet of Things as described above are implemented.

[0033] It can be seen from the above embodiments that the real-time response method for massive and highly concurrent IoT data provided by the embodiments of the present application has at least the following beneficial effects:

[0034] This application constructs a task type feature by analyzing the difference between the data volume and resource consumption data of historical data tasks, and taking the time interval between the start time of processing the historical data tasks and the current time as a weight, which can more accurately predict the occupation of different hardware resources by the waiting data tasks at the current time, thereby improving the matching degree between the waiting data tasks at the current time and the server hardware resources, and accelerating the server's response speed to data tasks; further, by analyzing the static and dynamic properties of the server, a data queuing rate is constructed, which can avoid the problem of unbalanced server load and improve the server's response speed to massive concurrent data; further, by comprehensively considering the task type characteristics and the data volume of the process, a process queue rate is constructed. The process type characteristics can better predict the processing efficiency of the process in the server, thereby improving the parallel processing efficiency, more reasonably assigning tasks to servers with corresponding hardware resources, and improving the response speed to data tasks; further, by comprehensively considering the task type characteristics and process type characteristics, as well as the hardware performance of the server, type adaptability is constructed, which can improve the task processing speed and more reasonably assign the waiting data tasks at the current moment to servers with corresponding hardware resources, thereby improving the data response speed; finally, by comprehensively considering the data queuing rate and type adaptability, task adaptability is constructed, which can respond to the waiting data tasks at the current moment, which can more reasonably assign tasks and improve the response speed of the server to massive concurrent data. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of a method for real-time response to massive and highly concurrent data in the Internet of Things provided by one embodiment of the present application;

[0037] Figure 2 A schematic diagram of a concurrent data response process provided for an embodiment of the present application;

[0038] Figure 3 A block diagram of a real-time response device for massive and highly concurrent data in the Internet of Things provided by one embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the real-time response method, device and medium for massive and high-concurrency data of the Internet of Things proposed in accordance with the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0041] The following is a detailed description of a specific solution of a real-time response method, device and medium for massive and high-concurrency data in the Internet of Things provided by the present application in conjunction with the accompanying drawings.

[0042] See also Figure 1 , which shows a flowchart of a method for real-time response to massive and high-concurrency data in the Internet of Things provided by an embodiment of the present application, the method comprising the following steps:

[0043] Step S1: Obtain all class static properties of each server, as well as the data volume of each process in each server at the current moment and the IoT node corresponding to each process, and obtain the processing start time, data volume and all class resource consumption data of each historical data task in each IoT node before the current moment, obtain the data volume of data tasks waiting for allocation of servers at the current moment, and record the data tasks waiting for allocation of servers as waiting data tasks.

[0044] In the Internet of Things, data processing terminals use multiple servers to respond to concurrent data and obtain all class-static properties of each server. All class-static properties include: CPU processing power, IO disk read and write capabilities, and network transmission capabilities. These capabilities are directly obtained through the hardware parameters stored in the server system, and their dimensions are the same, which is bit / s.

[0045] Furthermore, the data volume of each process in each server at the current moment and the IoT node corresponding to each process are obtained, and the processing start time of multiple historical data tasks in each IoT node before the current moment, the data volume of each of the multiple historical data tasks and all types of resource consumption data of each historical data task are obtained, the data volume of the data tasks waiting to be assigned to the server at the current moment is obtained, and the data tasks waiting to be assigned to the server are recorded as waiting data tasks.

[0046] Among them, the waiting data tasks at the current moment refer to the data tasks that have not yet been assigned to the server, and the process refers to the data tasks that have been assigned to the server and are being processed or waiting to be processed in the server; and an IoT node may send multiple data tasks to the server at the same time, but the server processes the data tasks in sequence, so a process in the server and an IoT node are one-to-one corresponding.

[0047] It should be noted that all resource consumption data include the total number of bits calculated by the CPU when the server processes data, the total number of bits read and written by the IO disk, and the total number of bits transmitted by the network bandwidth.

[0048] In addition, it should be understood that the number of historical data tasks is manually set. In this embodiment, the 30 historical data tasks closest to the current moment in each IoT node before the current moment are obtained as analysis objects. The implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0049] Step S2: Based on the difference between the data volume of each historical data task and all resource consumption data, and the time interval between the processing start time of each historical data task and the current time, determine the task type characteristics of each IoT node at the current time.

[0050] In a server, different types of tasks use different hardware facilities when performing calculations. For example, some tasks occupy more CPU for calculations during processing, but occupy less other hardware. Therefore, these tasks are called CPU-intensive tasks. There are also IO-intensive tasks, network transmission-intensive tasks, etc.

[0051] In the Internet of Things, the function of a node is generally relatively fixed, which leads to a relatively fixed data format and thus a relatively fixed type of data task. For example, an Internet of Things node composed of speakers usually needs to process sound data, and the occupancy rate of this type of data task on the CPU, disk IO, and network bandwidth is usually relatively fixed.

[0052] Therefore, this embodiment analyzes the data volume of historical data tasks and the consumption of various resources by historical data tasks to infer the occupation of different hardware by the waiting data tasks at the current moment. At the same time, when processing historical data tasks, considering that the task type in the IoT node may change due to equipment updates and other issues, the task type characteristics of the waiting data task at the current moment are determined by analyzing the task types of historical data tasks that are closer to the current moment. Specifically:

[0053] (1) In each IoT node, the ratio of all resource consumption data of each historical data task to the data volume of the corresponding historical data task is calculated, and all the ratios are combined into the task feature vector of each historical data task;

[0054] (2) Further, a normalized value of the time interval between the processing start time of each historical data task and the current time is calculated, and a result of multiplying the task feature vector of each historical data task by the normalized value of the time interval is used as the feature vector of each historical data task;

[0055] (3) Furthermore, the sum vector of the feature vectors of all historical data tasks of each IoT node before the current moment is used as the task type feature of each IoT node at the current moment.

[0056] The sum vector refers to a new vector formed by adding the elements at the same position of two or more vectors of the same length.

[0057] In addition, it should be understood that the normalized value of the time interval between the start time of processing and the current time is used as the weight. This is because the IoT node may change its task type characteristics due to equipment updates and other issues. Therefore, the historical data task at the most recent moment is used to characterize the task type characteristics of the IoT node at the current moment. The task feature vector and the corresponding normalized value are further weighted and fused to obtain the task type characteristics of the IoT node at the current moment, where the task type characteristic is a vector of length 3, and the three elements correspond to CPU characteristics, IO disk characteristics, and network bandwidth characteristics, respectively. The task feature type characterizes the consumption of hardware resources by the IoT node at the current moment.

[0058] Step S3: Based on the extreme distribution of all class static attributes in each server and the data volume of each process, determine the data queuing rate of each server at the current moment.

[0059] In IoT terminals, different servers have different hardware performances. Traditional terminal server load balancing methods do not take into account the performance differences of servers, which results in a situation where one type of hardware is running at full capacity while another type of hardware is idle. This situation will reduce the efficiency of IoT terminals in responding to concurrent data.

[0060] And if you want to add the waiting data task at the current moment to the process processing queue of the server, since the server has parallel processing capabilities, the process that is queued on the server may respond at the same time as the current process. If the current process and the queued process on the server can complement each other, that is, the task type characteristics of the current process and the task type characteristics of the queued process are complementary, then you can make full use of the server hardware resources, improve the server's parallel response capabilities, and improve the response speed to massive concurrent data.

[0061] Therefore, this embodiment allocates data tasks to the servers by analyzing the performance differences of the servers, that is, by analyzing the quality of all class static properties in the servers and combining the data volume of each process, the data queuing rate of each server at the current moment is determined, specifically:

[0062] The data queuing rate of server r at the current moment The expression is: ; In the formula, Indicates the data volume of the sth process in server r at the current moment; Represents the maximum value among all class static attributes of server r; Represents the number of all processes in server r at the current moment.

[0063] According to the data queuing rate of each server at the current moment, it can be understood that the larger the total amount of processes queued in the server, that is, the larger the total data volume of all processes in the server, and the smaller the maximum value of all class static attributes of the server, the smaller the service capacity of the server, and the larger the data queuing rate, the more time the server will spend processing the queued processes, and the less tasks should be assigned to the server; conversely, the smaller the total amount of processes queued in the server, that is, the smaller the total data volume of all processes in the server, and the larger the maximum value of all class static attributes of the server, the greater the service capacity of the server and the smaller the data queuing rate, and the task can be assigned to the server.

[0064] Step S4: Based on the task type characteristics of each process in each server corresponding to the IoT node and the data volume of each process in each server, determine the process type characteristics of each server at the current moment.

[0065] If the waiting data task at the current moment is added to the server, the later the process is solved in the server, the more likely it is to be processed in parallel with the waiting data task at the current moment. For example: the rth server already has 10 processes, and the processes are processed in parallel in ascending order according to the process numbers 1, 2, 3...10. For example, processes 1, 2, and 3 are processed at the same time. At this time, the server is already fully loaded and cannot calculate more processes in parallel. When process No. 1 is solved, the server has surplus computing power, and processes No. 4 and 5 are combined with processes No. 2 and 3 for parallel calculation. Therefore, the current process should be numbered 11 after it is added. Since the server calculates processes in parallel according to the order of numbers, the processes with later numbers are more likely to be calculated in parallel with the current process.

[0066] For parallel processing processes, if the features of the processes that are solved later in the server are more complementary to the features of the current task, adding the waiting data tasks at the current moment to the server can avoid the situation where one type of hardware is running at full capacity while the other type of hardware is idle. Therefore, by analyzing the task type features of the IoT nodes corresponding to each process and the data volume of each process in each server, the process type features of each server at the current moment are determined, specifically:

[0067] Number all processes in each server at the current moment, multiply the task type characteristics of the IoT node corresponding to each process in each server by the data volume of the process, and record it as the resource consumption vector of each process in each server at the current moment; multiply the resource consumption vector by the normalized value of the corresponding process number as the process feature vector of each process in each server at the current moment;

[0068] The process type feature of each server at the current moment is the sum vector of the process feature vectors of all processes in each server.

[0069] It should be understood that the result of multiplying the task type characteristics of the IoT node corresponding to each process by the data volume of the process, i.e., the resource consumption vector, is used to characterize the amount of hardware resources occupied by the process predicted by log data; further, the normalized value of the process change is used as the weight of the corresponding resource consumption vector, indicating that if the waiting data task process at the current moment is added to the current server, the later processed processes in the server are more likely to be processed in parallel with the waiting data task at the current moment, and the later processed processes should be used to estimate the process type characteristics of the server. The process type characteristics reflect the server hardware that the current process in a server prefers to use. The greater the numerical difference of all elements in the process type characteristics, the more likely it is that the server will have a certain type of hardware running at full capacity while another type of hardware is idle, making the server's response speed to concurrent data slower.

[0070] Step S5: Based on the data volume of the waiting data task at the current moment, all class static properties of each server, and the task type characteristics and the process type characteristics, determine the judgment vector of the waiting data task at the current moment under each process in each server, based on the discrete degree of all elements in the judgment vector under each process, determine the type adaptability of the waiting data task at the current moment under each server, and combined with the data queuing rate, determine the task adaptability of the waiting data task at the current moment under each server, and respond to the waiting data task at the current moment.

[0071] In order to determine the matching degree between each server hardware and the waiting data task at the current moment, so as to respond to the waiting data task at the current moment more quickly, by comprehensively considering the performance differences of the server hardware and combining the processing efficiency of each process in the server at the current moment, a more suitable server is assigned to respond to the waiting data task at the current moment. Specifically:

[0072] (1) The result of multiplying the data volume of the waiting data task at the current moment by the task type feature number of each IoT node is recorded as the current task feature of the waiting data task at the current moment under each IoT node;

[0073] (2) Further, the sum vector of the process type characteristics of each process in each server and the current task characteristics under the corresponding IoT node is recorded as the current task vector of the waiting data task under each process in each server at the current moment;

[0074] (3) Further, the ratio of all elements in the current task vector to the corresponding static attributes in all class static attributes is calculated, and the judgment vector of the waiting data task at the current moment under each process in each server is formed.

[0075] It should be noted that the three elements in the task type feature correspond to the CPU feature, IO disk feature and network bandwidth feature respectively, so the current task vector obtained is also a vector composed of three elements, and these three elements still correspond to the CPU feature, IO disk feature and network feature. At the same time, all class static attributes include CPU processing power, IO disk read and write capability, network transmission capability, and all class static attributes correspond to CPU features, IO disk features and network features. Therefore, in the process of obtaining the judgment vector, the ratio between all elements in the current task vector and the corresponding static attributes in all class static attributes is calculated. In this process, there is a one-to-one correspondence between the elements in the current task vector and the static attributes.

[0076] (4) Furthermore, the standard deviation of the normalized values ​​of all elements in the judgment vector of the waiting data task at the current moment under each server is used as the type adaptability of the waiting data task at the current moment under each server.

[0077] It should be understood that the smaller the standard deviation of all elements in the judgment vector, the smaller the type adaptability value, which indicates that the waiting data task at the current moment is more complementary to the current process in the server in terms of the use of server hardware resources, and the more adapted the waiting data task at the current moment is to the hardware characteristics of the server after the waiting data task at the current moment is added to the server, and the faster the waiting data task at the current moment can be responded to; conversely, the larger the standard deviation of all elements in the judgment vector, the larger the type adaptability value, and the less adapted the waiting data task at the current moment is to the hardware characteristics of the server after the waiting data task at the current moment is added to the server, and the waiting data task at the current moment cannot be responded to quickly.

[0078] (5) Further, based on the type adaptability and the data queuing rate, the task adaptability of the waiting data task at the current moment on each server is determined, specifically:

[0079] The expression of the task adaptability of the waiting data task on each server at the current moment is: ; In the formula, Indicates the task adaptability of the waiting data task under server r at the current moment; Indicates the type adaptability of the waiting data task under server r at the current moment; represents the data queuing rate of server r.

[0080] According to the task adaptability of the data tasks at the previous moment under each server, it can be understood that the smaller the type adaptability, the smaller the data queuing rate, and the smaller the obtained task adaptability, which means that the server is more suitable for processing the waiting data tasks at the current moment in terms of computing resource type and total computing resources; conversely, the greater the type adaptability, the greater the data queuing rate, and the greater the obtained task adaptability, which means that the server is less suitable for processing the waiting data tasks at the current moment in terms of computing resource type and total computing resources.

[0081] (6) Furthermore, at the current moment, for all servers, the server with the smallest task adaptability is selected as the server for the waiting data task at the current moment, and the waiting data task at the current moment is processed.

[0082] At this point, this embodiment allocates a server to the current process based on the hardware performance of the server and the type characteristics of the process in the server, which can fully utilize the hardware performance advantages of different servers, thereby effectively avoiding the problem of server load imbalance and improving the real-time response efficiency of the Internet of Things terminal to massive high-concurrency data.

[0083] Preferably, the concurrent data response process diagram provided in this embodiment is as follows: Figure 2 shown.

[0084] Based on the same inventive concept as the above method, the embodiment of the present application also provides a real-time response device for massive and high-concurrency data of the Internet of Things, including:

[0085] The data acquisition module is used to obtain all class static properties of each server, as well as the data volume of each process in each server at the current moment and the IoT node corresponding to each process, and obtain the processing start time, data volume and all class resource consumption data of each historical data task in each IoT node before the current moment, obtain the data volume of the data task waiting to be assigned to the server at the current moment, and record the data task waiting to be assigned to the server as the waiting data task;

[0086] A data analysis and processing module is used to determine the task type characteristics of each IoT node at the current moment based on the difference between the data volume of each historical data task and the data of all resource consumption types, and the time interval between the processing start time of each historical data task and the current moment;

[0087] Based on the extreme distribution of static attributes of all classes in each server and the amount of data in each process, determine the data queuing rate of each server at the current moment;

[0088] Based on the task type characteristics of each process in each server corresponding to the IoT node and the data volume of each process in each server, determine the process type characteristics of each server at the current moment;

[0089] The data response module is used to determine the judgment vector of the waiting data task at the current moment under each process in each server based on the data volume of the waiting data task at the current moment, all class static properties of each server, and the task type characteristics and the process type characteristics, determine the type adaptability of the waiting data task at the current moment under each server based on the discrete degree of all elements in the judgment vector under each process, and determine the task adaptability of the waiting data task at the current moment under each server in combination with the data queuing rate, and respond to the waiting data task at the current moment.

[0090] The embodiment of the present application provides a block diagram of a real-time response device for massive and high-concurrency data in the Internet of Things, such as Figure 3 shown.

[0091] Based on the same inventive concept as the above method, an embodiment of the present application also provides a real-time response medium for massive and high-concurrency data in the Internet of Things, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned methods for real-time response to massive and high-concurrency data in the Internet of Things are implemented.

[0092] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0094] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A real-time response method for massive and highly concurrent data in the Internet of Things, characterized in that: The method comprises the following steps: Get all class static properties of each server, as well as the data volume of each process in each server at the current moment and the IoT node corresponding to each process, and get the processing start time, data volume and all class resource consumption data of each historical data task in each IoT node before the current moment, get the data volume of the data task waiting to be assigned to the server at the current moment, and record the data task waiting to be assigned to the server as the waiting data task; Based on the difference between the data volume of each historical data task and the resource consumption data of all types, and the time interval between the processing start time of each historical data task and the current time, determine the task type characteristics of each IoT node at the current time; Based on the extreme distribution of static attributes of all classes in each server and the amount of data in each process, determine the data queuing rate of each server at the current moment; Based on the task type characteristics of each process in each server corresponding to the IoT node and the data volume of each process in each server, determine the process type characteristics of each server at the current moment; Based on the data volume of the waiting data task at the current moment, all class static properties of each server, and the task type characteristics and the process type characteristics, the judgment vector of the waiting data task at the current moment under each process in each server is determined, based on the discrete degree of all elements in the judgment vector under each process, the type adaptability of the waiting data task at the current moment under each server is determined, and combined with the data queuing rate, the task adaptability of the waiting data task at the current moment under each server is determined, and a response is made to the waiting data task at the current moment.

2. A real-time response method for massive and highly concurrent data in the Internet of Things as claimed in claim 1, characterized in that: The method for determining the task type characteristics of each IoT node at the current moment is: In each IoT node, the ratio of all resource consumption data of each historical data task to the data volume of the corresponding historical data task is calculated, and all the ratios are combined into the task feature vector of each historical data task; Calculate the normalized value of the time interval between the processing start time of each historical data task and the current time, and multiply the task feature vector of each historical data task by the normalized value of the time interval as the feature vector of each historical data task; The task type feature of each IoT node at the current moment is the sum vector of the feature vectors of all historical data tasks of each IoT node before the current moment.

3. A real-time response method for massive and highly concurrent data in the Internet of Things as claimed in claim 1, characterized in that: The expression of the data queuing rate of each server at the current moment is: ; In the formula, When represents the data queuing rate of server r at the previous moment; Indicates the data volume of the sth process in server r at the current moment; Represents the maximum value among all class static attributes of server r; Represents the number of all processes in server r at the current moment.

4. A real-time response method for massive and highly concurrent data of the Internet of Things as claimed in claim 3, characterized in that: The method for determining the process type characteristics of each server at the current moment is: Number all processes in each server at the current moment, multiply the task type characteristics of the IoT node corresponding to each process in each server by the data volume of the process, and record it as the resource consumption vector of each process in each server at the current moment; multiply the resource consumption vector by the normalized value of the corresponding process number as the process feature vector of each process in each server at the current moment; The process type feature of each server at the current moment is the sum vector of the process feature vectors of all processes in each server.

5. A real-time response method for massive and highly concurrent data in the Internet of Things as claimed in claim 1, characterized in that: The determination of the judgment vector of the waiting data task at the current moment under each process in each server includes: The result of multiplying the data volume of the waiting data task at the current moment by the task type feature number of each IoT node is recorded as the current task feature of the waiting data task at the current moment under each IoT node; The sum vector of the process type characteristics of each process in each server and the current task characteristics under the corresponding IoT node is recorded as the current task vector of the waiting data task under each process on each server at the current moment; All elements in the current task vector correspond to all class static attributes one by one. The ratios of all elements in the current task vector to the corresponding static attributes in all class static attributes are calculated, and the judgment vectors of the waiting data tasks at the current moment under each process in each server are formed.

6. A real-time response method for massive and highly concurrent data in the Internet of Things as claimed in claim 1, characterized in that: The type adaptability of the waiting data task at the current moment in each server is the standard deviation of the normalized values ​​of all elements in the judgment vector of the waiting data task at the current moment in each server.

7. A real-time response method for massive and highly concurrent data of the Internet of Things as claimed in claim 1, characterized in that: The expression of the task adaptability of the waiting data task at the current moment on each server is: ; In the formula, Indicates the task adaptability of the waiting data task under server r at the current moment; Indicates the type adaptability of the waiting data task under server r at the current moment; represents the data queuing rate of server r.

8. A real-time response method for massive and high-concurrency data in the Internet of Things as claimed in claim 1, characterized in that: The response to the waiting data task at the current moment: At the current moment, for all servers, the server with the smallest task adaptability is selected as the server of the waiting data task at the current moment, and the waiting data task at the current moment is processed.

9. A real-time response device for massive and high-concurrency data of the Internet of Things, implementing a real-time response method for massive and high-concurrency data of the Internet of Things as claimed in claim 1, characterized in that: The device comprises: The data acquisition module is used to obtain all class static properties of each server, as well as the data volume of each process in each server at the current moment and the IoT node corresponding to each process, and obtain the processing start time, data volume and all class resource consumption data of each historical data task in each IoT node before the current moment, obtain the data volume of the data task waiting to be assigned to the server at the current moment, and record the data task waiting to be assigned to the server as the waiting data task; A data analysis and processing module is used to determine the task type characteristics of each IoT node at the current moment based on the difference between the data volume of each historical data task and the data of all resource consumption types, and the time interval between the processing start time of each historical data task and the current moment; Based on the extreme distribution of static attributes of all classes in each server and the amount of data in each process, determine the data queuing rate of each server at the current moment; Based on the task type characteristics of each process in each server corresponding to the IoT node and the data volume of each process in each server, determine the process type characteristics of each server at the current moment; The data response module is used to determine the judgment vector of the waiting data task at the current moment under each process in each server based on the data volume of the waiting data task at the current moment, all class static properties of each server, and the task type characteristics and the process type characteristics, determine the type adaptability of the waiting data task at the current moment under each server based on the discrete degree of all elements in the judgment vector under each process, and determine the task adaptability of the waiting data task at the current moment under each server in combination with the data queuing rate, and respond to the waiting data task at the current moment.

10. A real-time response medium for massive and highly concurrent data of the Internet of Things, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a real-time response method for massive and highly concurrent data of the Internet of Things as described in any one of claims 1-8 are implemented.

Images