Elastic expansion and contraction method, device and equipment for cloud service capacity of Internet of Vehicles and storage medium

Abstract

The invention relates to an elastic scaling method, device and equipment for Internet of Vehicles cloud service capacity and a storage medium. According to the main technical scheme, the method comprises the following steps: acquiring Internet of Vehicles historical data in a preset historical date uploaded by a vehicle-mounted terminal, and carrying out real-time statistics on the Internet of Vehicles historical data to obtain moment historical data corresponding to each historical date; inputting the historical data of the moments corresponding to the historical dates into an Internet of Vehicles prediction model for model training to obtain prediction data of the moments corresponding to the current date; according to the moment historical data corresponding to each historical date and the moment prediction data corresponding to the current date, obtaining a moment prediction data confidence interval corresponding to the current date; and acquiring actual data at the moment corresponding to the current date, and performing elastic scaling operation on the Internet of Vehicles cloud service capacity according to the actual data at the moment corresponding to the current date and the predicted data confidence interval at the moment corresponding to the current date. According to the application, the effect of improving the stability of the Internet of Vehicles cloud service can be achieved.

Description

Technical Field

[0001] The present application relates to the technical field of cloud service for Internet of Vehicles, and in particular to a method, device, equipment and storage medium for elastically scaling the capacity of cloud service for Internet of Vehicles. Background Art

[0002] With the rapid development of cloud computing technology, various regions are paying more and more attention to the cloud computing service industry, and have taken cloud computing services as a new opportunity for the rapid development of the national software industry. As an emerging resource utilization platform, the cloud computing platform has formed a relatively mature service model, and more users can share and use resources. As a result, it has also been widely used in various fields, such as the field of cloud services for Internet of Vehicles.

[0003] Due to the rapid development of electric vehicle intelligence, the scale of the corresponding Internet of Vehicles cloud service is also expanding, and the demand for cloud service resources is different in different periods. Therefore, the cloud service capacity can generally be elastically scaled based on the amount of Internet of Vehicles signal data required and policy adjustments. Elastic scaling here specifically refers to the capacity demand of cloud computing resources changing with the changes in business. When the demand for computing resources is large, the computing resources are tight and the capacity needs to be increased rapidly, which is referred to as expansion; when the demand for computing resources is small, a large number of resources are idle. In order to use resources efficiently, the capacity needs to be reduced quickly, which is referred to as reduction. Applied to the field of Internet of Vehicles cloud services, since the Internet of Vehicles cloud service load is positively correlated with the amount of Internet of Vehicles signal data, the more vehicle network signal data, the more computing resources are required for cloud services, and vice versa.

[0004] In traditional implementation methods, elastic scaling solutions for Internet of Vehicles cloud service capacity are basically implemented based on threshold strategies. The so-called threshold strategy triggers an alarm when the cloud service capacity reaches a certain threshold, so as to remind operation and maintenance personnel to manually adjust the cloud service resources.

[0005] However, in order to control costs and reduce the number of alarms, the above method generally sets a very high threshold. When the anomaly is identified, the problem has already occurred. At this time, the operation and maintenance operations are obviously delayed, and there is a defect that it cannot adapt to the stable growth of Internet of Vehicles cloud services and wastes costs. Summary of the invention

[0006] Based on this, the present application provides a method, device, equipment and storage medium for elastically scaling the capacity of an Internet of Vehicles cloud service. Based on the actual data at the time corresponding to the current date and the confidence interval of the predicted data at the time corresponding to the current date, the capacity of the Internet of Vehicles cloud service is elastically scaled, so that abnormal growth or decrease trends in Internet of Vehicles cloud service data can be discovered in advance, so as to timely trigger automatic scaling operations on the capacity of the Internet of Vehicles cloud service cluster, thereby achieving the effect of improving the stability of the Internet of Vehicles cloud service.

[0007] In a first aspect, a method for elastic scaling of the capacity of a vehicle networking cloud service is provided. The method includes:

[0008] Obtain the vehicle networking historical data within a preset historical date uploaded by a vehicle-mounted terminal, perform real-time statistics on the vehicle networking historical data, and obtain the historical data corresponding to each moment of each historical date;

[0009] Input the historical data corresponding to each moment of each historical date into a vehicle networking prediction model for model training to obtain the predicted data corresponding to the current date and the corresponding moment;

[0010] According to the historical data corresponding to each moment of each historical date and the predicted data corresponding to the current date and the corresponding moment, obtain the confidence interval of the predicted data corresponding to the current date and the corresponding moment;

[0011] Obtain the actual data corresponding to the current date and the corresponding moment, and perform an elastic scaling operation on the capacity of the vehicle networking cloud service according to the actual data corresponding to the current date and the corresponding moment and the confidence interval of the predicted data corresponding to the current date and the corresponding moment.

[0012] According to an implementable manner in the embodiments of the present application, performing real-time statistics on the vehicle networking historical data to obtain the historical data corresponding to each moment of each historical date includes:

[0013] Perform real-time statistics on the vehicle networking historical data with the date as the dimension and the time as the index to obtain a two-dimensional data matrix with the date as the column header and the moment as the row header;

[0014] Interchange the rows and columns of the two-dimensional data matrix to obtain the historical data corresponding to each moment of each historical date with the moment as the column header and the date as the row header.

[0015] According to an implementable manner in the embodiments of the present application, inputting the historical data corresponding to each moment of each historical date into a vehicle networking prediction model for model training to obtain the predicted data corresponding to the current date and the corresponding moment includes:

[0016] Use the historical data corresponding to each moment of each historical date as data samples;

[0017] Obtain the custom coefficients and corresponding orders of the vehicle networking prediction model and the custom white noise error term;

[0018] Perform model training according to the historical data samples corresponding to each moment of each historical date, the custom coefficients and corresponding orders, and the custom white noise error term to obtain the predicted data corresponding to the current date and the corresponding moment.

[0019] According to an implementable manner in the embodiments of the present application, obtaining the confidence interval of the predicted data corresponding to the current date and the corresponding moment according to the historical data corresponding to each moment of each historical date and the predicted data corresponding to the current date and the corresponding moment includes:

[0020] According to the historical data at corresponding moments of each historical date and a preset variance formula, variance data at corresponding moments of each historical date is obtained;

[0021] According to the predicted data at the corresponding moment of the current date and the variance data at corresponding moments of each historical date, a confidence interval of the predicted data at the corresponding moment of the current date is obtained.

[0022] According to an implementable manner in the embodiments of the present application, according to the actual data at the corresponding moment of the current date and the confidence interval of the predicted data at the corresponding moment of the current date, an elastic scaling operation is performed on the capacity of the vehicle networking cloud service, including:

[0023] Determine whether the actual data at the corresponding moment of the current date is within the confidence interval of the predicted data at the corresponding moment of the current date;

[0024] If it is, maintain the current state of the capacity of the vehicle networking cloud service;

[0025] If it is not, perform an elastic scaling operation on the capacity of the vehicle networking cloud service.

[0026] According to an implementable manner in the embodiments of the present application, performing an elastic scaling operation on the capacity of the vehicle networking cloud service includes:

[0027] Obtain the total number of current service nodes of the vehicle networking cloud service and the preset data processing amount of a single service node;

[0028] According to the actual data at the corresponding moment of the current date, the total number of current service nodes, and the preset data processing amount of a single service node, obtain the number of service nodes for elastic scaling;

[0029] Perform an elastic scaling operation on the capacity of the vehicle networking cloud service according to the number of service nodes for elastic scaling.

[0030] According to an implementable manner in the embodiments of the present application, the elastic scaling operation includes expansion and contraction. Performing an elastic scaling operation on the capacity of the vehicle networking cloud service according to the number of service nodes for elastic scaling includes:

[0031] Judge the positive or negative value of the number of service nodes for elastic scaling;

[0032] When the number of service nodes for elastic scaling is positive, perform an expansion operation on the capacity of the vehicle networking cloud service;

[0033] When the number of service nodes for elastic scaling is negative, perform a contraction operation on the capacity of the vehicle networking cloud service.

[0034] In a second aspect, an elastic scaling device for the capacity of a vehicle networking cloud service is provided. The device includes:

[0035] A historical data unit, configured to obtain the vehicle networking historical data within a preset historical date uploaded by a vehicle-mounted terminal, perform real-time statistics on the vehicle networking historical data, and obtain the historical data corresponding to each moment of each historical date;

[0036] A prediction data unit, configured to input the historical data corresponding to each moment of each historical date into a vehicle networking prediction model for model training, and obtain the prediction data corresponding to the current date and each moment;

[0037] A confidence interval unit, configured to obtain the confidence interval of the prediction data corresponding to the current date and each moment according to the historical data corresponding to each moment of each historical date and the prediction data corresponding to the current date and each moment;

[0038] An elastic scaling unit, configured to obtain the actual data corresponding to the current date and each moment, and perform an elastic scaling operation on the vehicle networking cloud service capacity according to the actual data corresponding to the current date and each moment and the confidence interval of the prediction data corresponding to the current date and each moment.

[0039] In a third aspect, a computer device is provided, including:

[0040] At least one processor; and

[0041] A memory communicatively connected to the at least one processor; wherein,

[0042] The memory stores computer instructions executable by the at least one processor, and the computer instructions are executed by the at least one processor so that the at least one processor can execute the method involved in the first aspect above.

[0043] In a fourth aspect, a computer-readable storage medium is provided, on which computer instructions are stored, wherein the computer instructions are used to cause a computer to execute the method involved in the first aspect above.

[0044] According to the technical content provided by the embodiments of the present application, obtain the vehicle networking historical data within a preset historical date uploaded by the vehicle-mounted terminal, perform real-time statistics on the vehicle networking historical data to obtain the historical data corresponding to each moment of each historical date; input the historical data corresponding to each moment of each historical date into the vehicle networking prediction model for model training to obtain the predicted data corresponding to the current date and moment; according to the historical data corresponding to each moment of each historical date and the predicted data corresponding to the current date and moment, obtain the confidence interval of the predicted data corresponding to the current date and moment; obtain the actual data corresponding to the current date and moment, and perform elastic scaling operations on the vehicle networking cloud service capacity according to the actual data corresponding to the current date and moment and the confidence interval of the predicted data corresponding to the current date and moment. The above operations can perform elastic scaling operations on the vehicle networking cloud service capacity based on the actual data corresponding to the current date and moment and the confidence interval of the predicted data corresponding to the current date and moment, realizing the ability to detect abnormal growth or decline trends in vehicle networking cloud service data in advance, so as to timely trigger automatic scaling operations on the capacity of the vehicle networking cloud service cluster, achieving the effect of improving the stability of the vehicle networking cloud service. Description of the Drawings

[0045] Figure 1 It is a system architecture diagram of an elastic scaling method for vehicle networking cloud service capacity in an embodiment;

[0046] Figure 2 It is a schematic flowchart of an elastic scaling method for vehicle networking cloud service capacity in an embodiment;

[0047] Figure 3 It is a schematic diagram of the process of performing real-time statistics on vehicle networking historical data in an elastic scaling method for vehicle networking cloud service capacity in an embodiment;

[0048] Figure 4 It is a preferred schematic flowchart of an elastic scaling method for vehicle networking cloud service capacity in an embodiment;

[0049] Figure 5 It is a structural block diagram of an elastic scaling device for vehicle networking cloud service capacity in an embodiment;

[0050] Figure 6 It is a schematic structural diagram of a computer device in an embodiment. Detailed Embodiments

[0051] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] For the convenience of understanding, the system applicable to this application is first described. An elastic scaling method for the capacity of a vehicle networking cloud service provided by this application can be applied to, for example, Figure 1 the system architecture shown. This system architecture includes: an in-vehicle terminal 101 - a cloud server 103. Specifically, the cloud server 103 obtains the vehicle networking historical data within a preset historical date uploaded by the in-vehicle terminal 101, performs real-time statistics on the vehicle networking historical data to obtain the historical data corresponding to each moment of each historical date; inputs the historical data corresponding to each moment of each historical date into a vehicle networking prediction model for model training to obtain the predicted data corresponding to the current date and moment; obtains the confidence interval of the predicted data corresponding to the current date and moment according to the historical data corresponding to each moment of each historical date and the predicted data corresponding to the current date and moment; obtains the actual data corresponding to the current date and moment, and performs an elastic scaling operation on the capacity of the vehicle networking cloud service according to the actual data corresponding to the current date and moment and the confidence interval of the predicted data corresponding to the current date and moment. Among them, the cloud server 103 can be implemented by an independent server or a server cluster composed of multiple servers.

[0053] Figure 2 The flowchart of an elastic scaling method for the capacity of a vehicle networking cloud service provided by an embodiment of this application. This method can be executed by the cloud server 103 in the system architecture shown in Figure 1 as follows. As shown in Figure 2 the following steps may be included:

[0054] Step S201: Obtain the vehicle networking historical data within a preset historical date uploaded by the in-vehicle terminal, and perform real-time statistics on the vehicle networking historical data to obtain the historical data corresponding to each moment of each historical date.

[0055] Among them, the preset historical date can be set as T - n days.

[0056] Here, the in-vehicle terminal uploads the vehicle networking historical data of T - n days according to the data transmission protocol. The cloud server receives the data packets of the vehicle networking historical data based on the data transmission protocol, and performs real-time statistics on the vehicle networking historical data, thereby obtaining the historical data corresponding to each moment of each historical date, that is, the historical data of each moment from 0:00 to 24:00 every day in T - n days, and persists the historical data in the real-time data warehouse in chronological order. Among them, the data transmission protocol can be the protocol buffer protocol. Protobuf is a lightweight and efficient structured data storage format under Google, which is independent of the platform and language, extensible, can be used in fields such as communication protocols and data storage, and its encoding and decoding speed is extremely fast.

[0057] Step S203: Input the historical data corresponding to each moment of each historical date into a vehicle networking prediction model for model training to obtain the predicted data corresponding to the current date and moment.

[0058] Among them, the vehicle networking prediction model can be a time series model.

[0059] Here, according to the user's driving habits, the staff samples and analyzes the vehicle networking data of recent years and finds that the vehicle networking data shows a characteristic trend of normal distribution, and the vehicle networking data volume on weekdays is significantly higher than that on weekends, indicating that there are obvious differences in vehicle networking data between different time periods, that is, the vehicle networking data volumes on weekdays, weekends, and holidays respectively have periodicity. Therefore, the ARIMA model in the time series model can be used as the vehicle networking preset model. Among them, the ARIMA model (Autoregressive Integrated Moving Average Model) is also denoted as ARIMA(p, d, q), which is one of the most common statistical models used for time series prediction. Therefore, the cloud server can input the historical data corresponding to each historical date at each moment into the ARIMA model for model training, and then obtain the predicted data corresponding to the current date at each moment. That is, assuming that the current date is day T, the predicted data from 0:00 to 24:00 on day T can be obtained.

[0060] Step S205: Obtain the confidence interval of the predicted data corresponding to the current date at each moment according to the historical data corresponding to each historical date at each moment and the predicted data corresponding to the current date at each moment.

[0061] Here, since the historical data corresponding to each historical date at each moment and the predicted data corresponding to the current date at each moment are both known, therefore, the confidence interval of the predicted data corresponding to the current date at each moment can be obtained based on the historical data of each moment from 0:00 to 24:00 every day in T - n days and the predicted data of each moment from 0:00 to 24:00 on the current date T, that is, the confidence interval of the predicted data of each moment from 0:00 to 24:00 on day T.

[0062] Step S207: Obtain the actual data corresponding to the current date at each moment, and perform elastic scaling operations on the vehicle networking cloud service capacity according to the actual data corresponding to the current date at each moment and the confidence interval of the predicted data corresponding to the current date at each moment.

[0063] Here, obtain the actual data corresponding to the current date at each moment, that is, the actual data of each moment from 0:00 to 24:00 on day T, and perform elastic scaling operations on the vehicle networking cloud service capacity according to the actual data corresponding to the current date at each moment and the confidence interval of the predicted data corresponding to the current date at each moment. Still assuming that the current date is day T and the current moment is 12:00, then according to the actual data at 12:00 on day T and the confidence interval of the predicted data at 12:00 on day T, it is convenient to perform elastic scaling operations on the vehicle networking cloud service capacity, that is, the future trend of the vehicle networking cloud server load status can be predicted through the time series model, and the elastic scaling operation of the vehicle networking cloud service capacity can be realized.

[0064] It can be seen that in the embodiment of the present application, by obtaining the vehicle networking historical data within a preset historical date uploaded by the vehicle-mounted terminal, performing real-time statistics on the vehicle networking historical data, and obtaining the historical data corresponding to each historical date at corresponding moments; inputting the historical data corresponding to each historical date at corresponding moments into the vehicle networking prediction model for model training to obtain the predicted data corresponding to the current date at corresponding moments; according to the historical data corresponding to each historical date at corresponding moments and the predicted data corresponding to the current date at corresponding moments, obtaining the confidence interval of the predicted data corresponding to the current date at corresponding moments; obtaining the actual data corresponding to the current date at corresponding moments, and performing elastic scaling operations on the vehicle networking cloud service capacity according to the actual data corresponding to the current date at corresponding moments and the confidence interval of the predicted data corresponding to the current date at corresponding moments. The above operations can perform elastic scaling operations on the vehicle networking cloud service capacity based on the actual data corresponding to the current date at corresponding moments and the confidence interval of the predicted data corresponding to the current date at corresponding moments, so as to be able to detect the abnormal growth or decreasing trend of the vehicle networking cloud service data in advance, facilitating timely triggering of automatic scaling operations on the capacity of the vehicle networking cloud service cluster, and achieving the effect of improving the stability of the vehicle networking cloud service.

[0065] The following will describe each step in the above method process in detail. First, in combination with the embodiment, the step of "performing real-time statistics on the vehicle networking historical data to obtain the historical data corresponding to each historical date at corresponding moments" in the above step 201 will be described in detail.

[0066] Perform real-time statistics on the vehicle networking historical data with the date as the dimension and the time as the index to obtain a two-dimensional data matrix with the date as the column header and the time as the row header; perform row-column swapping on the two-dimensional data matrix to obtain the historical data corresponding to each historical date at corresponding moments with the time as the column header and the date as the row header.

[0067] Specifically, after the cloud server obtains the vehicle networking historical data within a preset historical date uploaded by the vehicle-mounted terminal, it persists the vehicle networking historical data into the real-time data warehouse in chronological order to form a data model of a two-dimensional matrix, with two columns of time and data volume respectively. As Figure 3 shown, here, assuming that the vehicle networking historical data within the preset historical date is the vehicle networking data from August 27, 2023 to September 5, 2023, then perform real-time statistics on the vehicle networking historical data with the date as the dimension and the time as the index to obtain a two-dimensional data matrix with the date as the column header and the time as the row header. And for the convenience of predicting the data corresponding to the current date at corresponding moments, the two-dimensional data matrix can also be subjected to row-column swapping to obtain the historical data corresponding to each historical date at corresponding moments with the time as the column header and the date as the row header.

[0068] Through the above operations, by performing real-time statistics on the historical data of the vehicle networking with the date as the dimension and the time as the indicator, the historical data corresponding to each historical date at each moment with the moment as the column header and the date as the row header can be obtained, so as to facilitate the prediction of the data corresponding to the current date at each moment in the later stage, achieving the effect of improving the prediction accuracy and efficiency.

[0069] The following describes in detail the above step S203, that is, "input the historical data corresponding to each historical date at each moment into the vehicle networking prediction model for model training to obtain the prediction data corresponding to the current date at each moment".

[0070] Take the historical data corresponding to each historical date at each moment as data samples; obtain the custom coefficients and corresponding orders of the vehicle networking prediction model and the custom white noise error term; perform model training according to the historical data samples corresponding to each historical date at each moment, the custom coefficients and corresponding orders, and the custom white noise error term to obtain the prediction data corresponding to the current date at each moment.

[0071] Among them, the custom coefficients include autoregressive coefficients and moving average coefficients; the corresponding orders include the autoregressive part order and the moving average part order.

[0072] Here, since the historical data volume of the vehicle networking shows an autoregressive relationship, an ARIMA model can be used as the vehicle networking prediction model, that is, K T equals K T-1 、K T-2 、…、K T-n regression.

[0073] Specifically, take the historical data corresponding to each historical date at each moment as data samples, and at the same time obtain the autoregressive coefficients, moving average coefficients, autoregressive part order, moving average part order, and custom white noise error term of the vehicle networking prediction model. Based on the foregoing variables, perform model training to obtain the prediction data corresponding to the current date at each moment, that is, use the ARIMA model to Figure 3 predict the vehicle networking data in each row of

[0074] K Δt(T) =α1K Δt(T-1) +α2K Δt(T-1) +…+α n K Δt(Y-T-n) +∈ Δt +β1∈ Δt(T-1) +…+β q ∈ Δy(T-q)

[0075] Among them, α1, α2… represent autoregressive coefficients; ∈ ΔtIt represents a custom white noise error term; β1, β2... represent moving average coefficients; n represents the order of the autoregressive part; q represents the order of the moving average part. It should be noted that the autoregressive coefficient, the custom white noise error term, and the moving average coefficient are all constants preset by the staff based on experience. The order of the autoregressive part is generally set based on the statistical historical dates, and the order of the moving average part is generally set based on the dates with relatively large fluctuations in the historical dates, so as to achieve the adjustment of the time series.

[0076] Through the above operations, by using the historical data at the corresponding moments of each historical date as data samples and then obtaining the predicted data at the corresponding moment of the current date based on the ARIMA model, it realizes the effect of accurately predicting the predicted data at the corresponding moment of the current date based on the historical data at the corresponding moments of each historical date relatively stably.

[0077] The following combines the embodiments to describe in detail the above step S205, that is, "obtaining the confidence interval of the predicted data at the corresponding moment of the current date according to the historical data at the corresponding moments of each historical date and the predicted data at the corresponding moment of the current date".

[0078] According to the historical data at the corresponding moments of each historical date and the preset variance formula, the variance data at the corresponding moments of each historical date is obtained; according to the predicted data at the corresponding moment of the current date and the variance data at the corresponding moments of each historical date, the confidence interval of the predicted data at the corresponding moment of the current date is obtained.

[0079] Here, according to the historical data at the corresponding moments of each historical date and the preset variance formula, the variance data at the corresponding moments of each historical date is obtained, and the specific expression is as follows:

[0080]

[0081] Performing an addition operation on the predicted data at the corresponding moment of the current date and the variance data at the corresponding moments of each historical date to obtain the upper boundary of the confidence interval of the predicted data at the corresponding moment of the current date, and performing a subtraction operation on the predicted data at the corresponding moment of the current date and the variance data at the corresponding moments of each historical date to obtain the lower boundary of the confidence interval of the predicted data at the corresponding moment of the current date. The specific expression is as follows:

[0082] Upper boundary:

[0083] Lower boundary:

[0084] Through the above expressions, the confidence interval of the predicted data at the corresponding moment of the current date is obtained as Between.

[0085] The above operations predict data corresponding to the current moment based on the current date and variance data corresponding to the moments of each historical date, obtain the confidence interval of the predicted data corresponding to the current moment of the current date, and then facilitate elastic scaling operations on the capacity of the vehicle networking cloud service based on the confidence interval to achieve the warning function for high / low loads of the vehicle networking cloud service capacity.

[0086] Finally, in combination with the embodiments, the above step S207, that is, "perform elastic scaling operations on the capacity of the vehicle networking cloud service according to the actual data corresponding to the current moment of the current date and the confidence interval of the predicted data corresponding to the current moment of the current date", is described in detail.

[0087] Judge whether the actual data corresponding to the current moment of the current date is within the confidence interval of the predicted data corresponding to the current moment of the current date; if so, keep the current state of the vehicle networking cloud service capacity; if not, perform elastic scaling operations on the vehicle networking cloud service capacity.

[0088] Here, obtain the actual data corresponding to the current moment of the current date, and assume it is represented by K Δt(AT) Based on the actual data corresponding to the current moment of the current date, judge whether it is within the confidence interval of the predicted data corresponding to the current moment of the current date, that is, compare K Δt(AT) with and respectively. When , it means that the actual data corresponding to the current moment of the current date is within the confidence interval of the predicted data corresponding to the current moment of the current date, and then the current state of the vehicle networking cloud service capacity can be kept without any additional operations; when or , an alarm is triggered, indicating that the actual data corresponding to the current moment of the current date is not within the confidence interval of the predicted data corresponding to the current moment of the current date, that is, elastic scaling operations can be performed on the vehicle networking cloud service capacity to achieve real-time dynamic scaling of vehicle networking data during peak or trough periods. It should be noted that the capacity of the vehicle networking cloud service here generally refers to the capacity of the vehicle networking cloud service cluster.

[0089] In one embodiment, performing elastic scaling operations on the vehicle networking cloud service capacity includes: obtaining the total number of current service nodes of the vehicle networking cloud service and the preset data processing amount of a single service node; obtaining the number of elastic scaling service nodes according to the actual data corresponding to the current moment of the current date, the total number of current service nodes, and the preset data processing amount of a single service node; and performing elastic scaling operations on the vehicle networking cloud service capacity according to the number of elastic scaling service nodes.

[0090] Here, obtain the total number of current service nodes of the vehicle networking cloud service, assume it is represented by x, and the preset data processing amount of a single service node, assume it is represented by a. Obtain the number of elastic scaling service nodes according to the actual data corresponding to the current moment of the current date, the total number of current service nodes, and the preset data processing amount of a single service node. The specific expression is as follows:

[0091]

[0092] Among them, y represents the number of elastic scaling service nodes; a represents the preset data processing volume of a single service node, and x represents the total number of current service nodes. It should be noted that x and a are preset in advance. x is the total number of service nodes in the current system, and a represents the data processing capacity of a single service node. Based on different service operations, the setting of a is also different. After obtaining the number of elastic scaling service nodes, elastic scaling operations can be performed on the vehicle networking cloud service capacity according to the number of elastic scaling service nodes.

[0093] Based on the actual data at the corresponding moment of the current date, the total number of current service nodes, and the preset data processing volume of a single service node, the above operations obtain the number of elastic scaling service nodes, so that the staff can perform elastic scaling operations on the vehicle networking cloud service capacity based on the number of elastic scaling service nodes, achieving the effect of reducing the operation and maintenance and computing costs of the vehicle networking cloud service.

[0094] In one embodiment, performing elastic scaling operations on the vehicle networking cloud service capacity according to the number of elastic scaling service nodes includes: judging the positive and negative values of the number of elastic scaling service nodes; when the number of elastic scaling service nodes is positive, performing an expansion operation on the vehicle networking cloud service capacity; when the number of elastic scaling service nodes is negative, performing a reduction operation on the vehicle networking cloud service capacity.

[0095] Among them, the elastic scaling operation includes expansion and reduction.

[0096] Here, judging the positive and negative values of the number of elastic scaling service nodes. When the number of elastic scaling service nodes is positive, it indicates that an expansion operation needs to be performed on the vehicle networking cloud service capacity to achieve the effect of improving the job processing ability. The specific expression is as follows:

[0097] Nae = x + y

[0098] When the number of elastic scaling service nodes is negative, it indicates that a reduction operation needs to be performed on the vehicle networking cloud service capacity to avoid wasting operating costs. The specific expression is as follows:

[0099] Nae = max(x + y, 1)

[0100] Among them, Nae represents the adjusted vehicle networking cloud service capacity; x represents the total number of current service nodes; y represents the number of elastic scaling service nodes. It should be noted that when the number of elastic scaling service nodes is negative, taking the maximum value is to ensure that the minimum number of currently used service nodes is 1.

[0101] Through the above operations, by judging the positive and negative values of the number of elastic scaling service nodes, the capacity of the vehicle networking cloud service is expanded and reduced, so as to be able to detect in advance the abnormal growth or decreasing trend of the vehicle networking data volume, so as to trigger in time the expansion and contraction operations of the vehicle networking cloud service cluster capacity, and achieve the effect of early identification and automatic scaling.

[0102] Combined with the implementation manners in the above embodiments, the following Figure 4 is an example description of a preferred method flow provided by the embodiments of the present application. As Figure 4 shown, the method may include the following steps:

[0103] Step S401: Obtain the vehicle networking historical data within a preset historical date uploaded by the in-vehicle terminal.

[0104] Step S402: Perform real-time statistics on the vehicle networking historical data with the date as the dimension and the time as the index to obtain a two-dimensional data matrix with the date as the column header and the time as the row header.

[0105] Step S403: Swap the rows and columns of the two-dimensional data matrix to obtain the historical data corresponding to each historical date with the time as the column header and the date as the row header.

[0106] Step S404: Use the historical data corresponding to each historical date at the corresponding time as data samples.

[0107] Step S405: Obtain the custom coefficients, corresponding orders, and custom white noise error terms of the vehicle networking prediction model.

[0108] Step S406: Perform model training according to the historical data samples corresponding to each historical date at the corresponding time, custom coefficients, corresponding orders, and custom white noise error terms to obtain the prediction data corresponding to the current date at the corresponding time.

[0109] Step S407: Obtain the variance data corresponding to each historical date at the corresponding time according to the historical data corresponding to each historical date at the corresponding time and the preset variance formula.

[0110] Step S408: Obtain the confidence interval of the prediction data corresponding to the current date at the corresponding time according to the prediction data corresponding to the current date at the corresponding time and the variance data corresponding to each historical date at the corresponding time.

[0111] Step S409: Obtain the actual data corresponding to the current date at the corresponding time.

[0112] Step S410: Judge whether the actual data corresponding to the current date at the corresponding time is within the confidence interval of the prediction data corresponding to the current date at the corresponding time. If it is, execute Step S411; if not, execute Step S412.

[0113] Step S411: Maintain the current state of the vehicle networking cloud service capacity.

[0114] Step S412: Obtain the total number of current service nodes of the vehicle networking cloud service and the preset data processing volume per service node.

[0115] Step S413: Obtain the elastic scaling service node number according to the actual data at the corresponding moment of the current date, the total number of current service nodes, and the preset data processing volume per service node.

[0116] Step S414: Judge the positive or negative value of the elastic scaling service node number. When the elastic scaling service node number is positive, execute Step S415; when the elastic scaling service node number is negative, execute Step S416.

[0117] Step S415: Perform an operation to expand the capacity of the vehicle networking cloud service.

[0118] Step S416: Perform an operation to reduce the capacity of the vehicle networking cloud service.

[0119] It should be understood that although Figure 2 、 Figure 4 each step in the flowchart is shown in sequence according to the indication of the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless there is a clear description in this application, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, Figure 2 、 Figure 4 at least a part of the steps in

[0120] Figure 5 is a schematic structural diagram of an elastic scaling device for the capacity of a vehicle networking cloud service provided by an embodiment of this application. This device can be set in Figure 1 the cloud server in the system architecture shown in Figure 2 、 Figure 4 to execute the method flow shown in Figure 5 、

[0121] The historical data unit 501 is used to obtain the vehicle networking historical data within a preset historical date uploaded by the vehicle-mounted terminal, perform real-time statistics on the vehicle networking historical data, and obtain the historical data at the corresponding moment of each historical date;

[0122] A prediction data unit 503, configured to input historical data at corresponding times of each historical date into a vehicle networking prediction model for model training to obtain prediction data at the corresponding time of the current date;

[0123] A confidence interval unit 505, configured to obtain a confidence interval of the prediction data at the corresponding time of the current date according to the historical data at the corresponding times of each historical date and the prediction data at the corresponding time of the current date;

[0124] An elastic scaling unit 507, configured to obtain actual data at the corresponding time of the current date, and perform an elastic scaling operation on the vehicle networking cloud service capacity according to the actual data at the corresponding time of the current date and the confidence interval of the prediction data at the corresponding time of the current date.

[0125] In one embodiment, the historical data unit 501 is further configured to:

[0126] Perform real-time statistics on the vehicle networking historical data with the date as the dimension and the time as the index to obtain a two-dimensional data matrix with the date as the column header and the time as the row header;

[0127] Interchange the rows and columns of the two-dimensional data matrix to obtain historical data at the corresponding times of each historical date with the time as the column header and the date as the row header.

[0128] In one embodiment, the prediction data unit 503 is further configured to:

[0129] Use the historical data at the corresponding times of each historical date as data samples;

[0130] Obtain the custom coefficients and corresponding orders of the vehicle networking prediction model and the custom white noise error term;

[0131] Perform model training according to the historical data samples at the corresponding times of each historical date, the custom coefficients and corresponding orders, and the custom white noise error term to obtain prediction data at the corresponding time of the current date.

[0132] In one embodiment, the confidence interval unit 505 is further configured to:

[0133] Obtain variance data at the corresponding times of each historical date according to the historical data at the corresponding times of each historical date and a preset variance formula;

[0134] Obtain a confidence interval of the prediction data at the corresponding time of the current date according to the prediction data at the corresponding time of the current date and the variance data at the corresponding times of each historical date.

[0135] In one embodiment, the elastic scaling unit 507 is further configured to:

[0136] Determine whether the actual data at the corresponding time of the current date is within the confidence interval of the prediction data at the corresponding time of the current date;

[0137] If it exists, maintain the current state of the vehicle networking cloud service capacity;

[0138] If it does not exist, perform elastic scaling on the vehicle networking cloud service capacity.

[0139] In one embodiment, the elastic scaling unit 507 is further configured to:

[0140] Obtain the current total number of service nodes of the vehicle networking cloud service and the preset data processing amount per service node;

[0141] Based on the actual data at the corresponding moment of the current date, the current total number of service nodes, and the preset data processing amount per service node, obtain the number of service nodes for elastic scaling;

[0142] Perform elastic scaling on the vehicle networking cloud service capacity according to the number of service nodes for elastic scaling.

[0143] In one embodiment, the elastic scaling operation includes capacity expansion and reduction. The elastic scaling unit 507 is further configured to:

[0144] Judge the positive or negative value of the number of service nodes for elastic scaling;

[0145] When the number of service nodes for elastic scaling is positive, perform capacity expansion on the vehicle networking cloud service capacity;

[0146] When the number of service nodes for elastic scaling is negative, perform capacity reduction on the vehicle networking cloud service capacity.

[0147] For the same or similar parts among the above embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0148] It should be noted that the embodiments of the present application may involve the use of user data. In actual applications, user-specific personal data can be used in the solutions described herein within the scope permitted by applicable laws and regulations (such as when the user clearly consents, is effectively notified to the user, and the user clearly authorizes, etc.) in compliance with the requirements of applicable laws and regulations in the country where it is located.

[0149] According to the embodiments of the present application, the present application also provides a computer device and a computer-readable storage medium.

[0150] Such as Figure 6As shown, it is a block diagram of a computer device according to an embodiment of the present application. The computer device is intended to represent various forms of digital computers or mobile devices. Among them, the digital computer may include a desktop computer, a portable computer, a workbench, a personal digital assistant, a server, a mainframe computer, and other suitable computers. The mobile device may include a tablet computer, a smart phone, a wearable device, etc.

[0151] As Figure 6 shown, the device 600 includes a computing unit 601, a ROM 602, a RAM 603, a bus 604, and an input / output (I / O) interface 605. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via the bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0152] The computing unit 601 can execute various processes in the method embodiments of the present application according to the computer instructions stored in the read-only memory (ROM) 602 or the computer instructions loaded from the storage unit 608 into the random access memory (RAM) 603. The computing unit 601 can be various general-purpose and / or dedicated processing components with processing and computing capabilities. The computing unit 601 may include, but is not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. In some embodiments, the method provided by the embodiments of the present application can be implemented as a computer software program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 608.

[0153] The RAM 603 can also store various programs and data required for the operation of the device 600. Part or all of the computer programs can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609.

[0154] The input unit 606, the output unit 607, the storage unit 608, and the communication unit 609 in the device 600 can be connected to the I / O interface 605. Among them, the input unit 606 can be such as a keyboard, a mouse, a touch screen, a microphone, etc.; the output unit 607 can be such as a display, a speaker, an indicator light, etc. The device 600 can exchange information, data, etc. with other devices through the communication unit 609.

[0155] It should be noted that the device may also include other components necessary for normal operation. It may also only include the components necessary to implement the solution of the present application, and does not necessarily include all the components shown in the figure.

[0156] The various embodiments of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof.

[0157] The computer instructions for implementing the methods of this application can be written in any combination of one or more programming languages. These computer instructions can be provided to a computing unit 601 such that when the computer instructions are executed by a computing unit 601 such as a processor, the steps involved in the method embodiments of this application are performed.

[0158] The computer-readable storage medium provided by this application can be a tangible medium that can contain or store computer instructions for performing the steps involved in the method embodiments of this application. The computer-readable storage medium can include, but is not limited to, storage media in the form of electronic, magnetic, optical, electromagnetic, etc.

[0159] The above specific embodiments do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims

1. An elastic scaling method for the capacity of a vehicle networking cloud service, characterized in that, The method includes: Obtaining the historical data of the Internet of Vehicles within a preset historical date uploaded by the vehicle-mounted terminal, and performing real-time statistics on the historical data of the Internet of Vehicles to obtain the historical data corresponding to each moment of each historical date; Inputting the historical data corresponding to each moment of each historical date into the Internet of Vehicles prediction model for model training to obtain the predicted data corresponding to the current date and moment; Obtaining the confidence interval of the predicted data corresponding to the current date and moment according to the historical data corresponding to each moment of each historical date and the predicted data corresponding to the current date and moment; Obtaining the actual data corresponding to the current date and moment, and performing elastic scaling operations on the Internet of Vehicles cloud service capacity according to the actual data corresponding to the current date and moment and the confidence interval of the predicted data corresponding to the current date and moment.

2. The method according to claim 1, wherein The performing real-time statistics on the historical data of the Internet of Vehicles to obtain the historical data corresponding to each moment of each historical date includes: Performing real-time statistics on the historical data of the Internet of Vehicles with the date as the dimension and the time as the index to obtain a two-dimensional data matrix with the date as the column header and the moment as the row header; Interchanging the rows and columns of the two-dimensional data matrix to obtain the historical data corresponding to each moment of each historical date with the moment as the column header and the date as the row header.

3. The method according to claim 2, wherein The inputting the historical data corresponding to each moment of each historical date into the Internet of Vehicles prediction model for model training to obtain the predicted data corresponding to the current date and moment includes: Taking the historical data corresponding to each moment of each historical date as data samples; Obtaining the custom coefficients and corresponding orders of the Internet of Vehicles prediction model and the custom white noise error term; Performing model training according to the historical data samples corresponding to each moment of each historical date, the custom coefficients and corresponding orders, and the custom white noise error term to obtain the predicted data corresponding to the current date and moment.

4. The method according to claim 1, wherein The obtaining the confidence interval of the predicted data corresponding to the current date and moment according to the historical data corresponding to each moment of each historical date and the predicted data corresponding to the current date and moment includes: Obtaining the variance data corresponding to each moment of each historical date according to the historical data corresponding to each moment of each historical date and the preset variance formula; Obtaining the confidence interval of the predicted data corresponding to the current date and moment according to the predicted data corresponding to the current date and moment and the variance data corresponding to each moment of each historical date.

5. The method according to claim 1, wherein The performing elastic scaling operations on the Internet of Vehicles cloud service capacity according to the actual data corresponding to the current date and moment and the confidence interval of the predicted data corresponding to the current date and moment includes: Judging whether the actual data corresponding to the current date and moment is within the confidence interval of the predicted data corresponding to the current date and moment; If it is, maintaining the current state of the Internet of Vehicles cloud service capacity; If it is not, performing elastic scaling operations on the Internet of Vehicles cloud service capacity.

6. The method according to claim 5, wherein The performing elastic scaling operations on the Internet of Vehicles cloud service capacity includes: Obtaining the total number of current service nodes of the Internet of Vehicles cloud service and the preset data processing amount of a single service node; Obtaining the number of elastic scaling service nodes according to the actual data corresponding to the current date and moment, the total number of current service nodes, and the preset data processing amount of a single service node; Performing elastic scaling operations on the Internet of Vehicles cloud service capacity according to the number of elastic scaling service nodes.

7. The method according to claim 6, characterized in that The elastic scaling operation includes scaling up and scaling down. The elastic scaling operation on the vehicle networking cloud service capacity according to the number of elastic scaling service nodes includes: Judging the positive or negative value of the number of elastic scaling service nodes; When the number of elastic scaling service nodes is positive, performing a scaling-up operation on the vehicle networking cloud service capacity; When the number of elastic scaling service nodes is negative, performing a scaling-down operation on the vehicle networking cloud service capacity.

8. An elastic scaling device for the capacity of a vehicle networking cloud service, characterized in that, The device includes: A historical data unit, configured to obtain the vehicle networking historical data within a preset historical date uploaded by the vehicle-mounted terminal, perform real-time statistics on the vehicle networking historical data, and obtain the historical data at corresponding moments of each historical date; A prediction data unit, configured to input the historical data at corresponding moments of each historical date into a vehicle networking prediction model for model training, and obtain the prediction data at the corresponding moment of the current date; A confidence interval unit, configured to obtain the confidence interval of the prediction data at the corresponding moment of the current date according to the historical data at the corresponding moments of each historical date and the prediction data at the corresponding moment of the current date; An elastic scaling unit, configured to obtain the actual data at the corresponding moment of the current date, and perform an elastic scaling operation on the vehicle networking cloud service capacity according to the actual data at the corresponding moment of the current date and the confidence interval of the prediction data at the corresponding moment of the current date.

9. A computer device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores computer instructions executable by the at least one processor, and the computer instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-7.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that, The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.

Images