Methods for operating a backend server, backend server and system with such a

Abstract

In backend servers that receive and / or send data packets to a plurality of participants (preferably vehicles (1, 2, 3, 4, 5)), a prediction is made of the number of data packets to be processed during operation, and the capacity is adjusted accordingly. This capacity can be provided by an external data processing unit (20).

Description

[0001] The invention relates to a method for operating a backend server that receives data packets, in particular containers, from and / or sends data packets to a plurality of participants. The invention also relates to a backend server that can be used in the method, as well as a system that includes, among other things, several such backend servers.

[0002] The underlying principle is that the aforementioned participants should be vehicles that are primarily in motion. Data packets can be exchanged within the vehicle, for example, during internet use ("user login," downloading data from the internet). Vehicles can also transmit location data to services.

[0003] Overall, the invention primarily relates to the application layer in the OSI model (Open Systems Interconnection model). This means that the data packets must be handled by the backend server, utilizing processor capacity, for example, from external processors (data processing units). The available transmission bandwidth (i.e., relating to the transport layer in the OSI model) is initially assumed to be given. Processor capacity represents a bottleneck: particularly with a large fleet of vehicles managed by the backend server, it can happen that too many data packets are exchanged and thus require processing by the backend server. Therefore, there is a tendency to provide more data processing capacity than necessary, but this is inefficient in the long run.

[0004] From US 10 187 327 B2, it is known to attempt to predict when an individual user will need network services. This optimizes network bandwidth.

[0005] US 11 665 531 B2 concerns a prediction of “network health”, meaning that a deterioration in network quality should be predicted in a timely manner.

[0006] According to US 10 158 534 B2, the transmission of data packets over a network should be managed in order to predict a network failure early and to be able to react if necessary.

[0007] According to US 11 223 945 B2, a network outage should be detected and an alarm issued. The network outage is detected by comparing the current state with an expected state. The expected state specifically refers to the number of requests that were previously sent under comparable conditions in the more distant past.

[0008] US 10 409 649 B1 deals with load forecasting for web services and a corresponding response. US 2022 / 0 385 579 A1 concerns dynamic load balancing in data traffic.

[0009] The object of the invention is to improve the utilization of resources during the exchange of data packets between a backend server and participants. This object is achieved by the method with the features according to claim 1, the backend server with the features according to claim 8, and the system according to claim 9.

[0010] The inventive method for operating a backend server, which receives data packets, in particular containers, from and / or sends data packets to a plurality of participants provided by motor vehicles, and which has access to at least partially external data processing equipment and has at least one communication interface for wireless communication with the vehicles, wherein the backend server further comprises a prediction device designed to predict the number of data packets to be transmitted in the future via the at least one communication interface, and which finally comprises a capacity planning device designed to predict a future data processing capacity based on the predicted number, this method comprises: - Predictions of the number of data packets to be processed (i.e., received and / or sent) by the prediction device during ongoing operation at a later time interval; - In the event of a predicted number of data packets that does not correspond (according to a predetermined criterion) to the current processor capacity of the external data processing facility, effect a change, during operation, in the capacity required by the data processing facility by requesting such capacity in advance via access to the external data processing facility.

[0011] The invention thus allocates a flexible capacity to the backend server. There can be, for example, a base capacity that is always available, and expansion levels where at least partial access to an external data processing facility is made. The backend server essentially acts as a broker for capacity. This makes it possible to keep the total capacity provided in a system lower than previously required.

[0012] The participants are provided by motor vehicles. The participants can therefore, in a sense, be the motor vehicle itself and / or a unit located within the motor vehicle. The unit can also be installed in the motor vehicle (mobile communication device such as a smartphone, smartwatch, and the like), in the latter case preferably being connected to a data processing unit of the motor vehicle.

[0013] According to the invention, the capacity is the processor capacity of the data processing device external to the backend server. Thus, the backend server can handle a base load if necessary, but can always access the external data processing device when needed.

[0014] According to a preferred embodiment of the invention, the prediction relates to the number of data packets expected to be processed in a time interval that begins from 10 s to 1 min later and preferably lasts between 5 s and 1 min.

[0015] This involves a relatively short-term prediction, which can therefore be quite precise. For example, when a vehicle is switched on, it can be assumed that there is a certain probability that internet access will be desired shortly afterward, requiring the sending and receiving of data packets. Particularly when statistically averaged, this prediction can be very accurate. By varying the capacity accessed by the backend server relatively quickly, the overall capacity of a system can be kept particularly low, as there is no long-term capacity commitment.

[0016] According to another preferred embodiment, predictions are continuously remade at intervals of between 5 s and 1 min, preferably between 10 s and 30 s. Thus, the predictions can be made at fixed time intervals that are either 5 s long, 1 s long, or an intermediate value. Alternatively, a prediction can be triggered by events that change the number of data packets. In this case, a first interval between two predictions can be approximately 5 s, another 1 min, and a third an intermediate value between 5 s and 1 min.

[0017] According to a further preferred embodiment, if the predicted number is greater than the current capacity allows, the capacity is increased. Alternatively or additionally, it is provided that if the predicted number is less than the current capacity by a predetermined first ("too small") number, the capacity is increased. Alternatively or additionally, it is provided that if the predicted number is less than the current capacity by a predetermined second number, the capacity is decreased.

[0018] Overall, a kind of hysteresis curve can be traversed: A threshold to expand the capacity can be higher than the threshold to decrease it again.

[0019] Only in rare cases is it possible that the transport layer is actually affected: According to a preferred embodiment, an alarm is issued if a predicted number is too large for the existing bandwidth for data transmission.

[0020] According to another preferred embodiment, prediction is performed using a least squares method, a Kalman filter, and / or an artificial intelligence provision. These methods can be designed to be very efficient.

[0021] The backend server according to the invention serves to receive data packets from one vehicle and to provide data packets to a plurality of vehicles and comprises: - Access (interface) to an at least partially external data processing facility; - at least one communication interface for wireless communication with the vehicles; - a predictive device designed to predict a number of data packets to be transmitted in the future via at least one communication interface; - a capacity planning facility designed to predict, based on the predicted number, a future data processing capacity required by the data processing facility and to request such capacity in advance via access to the external data processing facility.

[0022] The backend server according to the invention enables the implementation of the method according to one embodiment.

[0023] The system according to the invention comprises a plurality of backend servers according to the invention and at least one data processing facility which is accessible to several backend servers.

[0024] Here the advantage of this measure becomes obvious: Since multiple backend servers can access one data processing facility, the maximum capacity of the data processing facility can be lower than if there were a dedicated data processing facility for each backend server, each of which would then have to have a maximum capacity, resulting in all maximum capacities of the individual data processing facilities being very high, as has been the practice so far.

[0025] The invention also includes the control device for the motor vehicle. The control device can comprise a data processing device or a processor circuit configured to carry out an embodiment of the method according to the invention. For this purpose, the processor circuit can comprise at least one microprocessor and / or at least one microcontroller and / or at least one FPGA (Field Programmable Gate Array) and / or at least one DSP (Digital Signal Processor). In particular, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an NPU (Neural Processing Unit) can be used as the microprocessor. Furthermore, the processor circuit can comprise program code configured to carry out the embodiment of the method according to the invention when executed by the processor circuit.The program code can be stored in a data memory of the processor device. The processor device can be based, for example, on at least one circuit board and / or on at least one SoC (System on Chip).

[0026] The motor vehicle provided according to the invention is preferably designed as a motor vehicle, in particular as a passenger car or truck, or as a passenger bus or motorcycle.

[0027] As a further solution, the invention also includes a computer-readable storage medium comprising program code which, when executed by a computer or a computer network, causes it to execute an embodiment of the method according to the invention. The storage medium can be provided at least partially as a non-volatile data storage medium (e.g., as flash memory and / or as an SSD - solid state drive) and / or at least partially as a volatile data storage medium (e.g., as RAM - random access memory). The storage medium can be located within the computer or computer network. However, the storage medium can also be operated, for example, as an app store server and / or cloud server on the internet. The computer or computer network can provide a processor circuit with, for example, at least one microprocessor.The program code can be provided as binary code, assembly code, source code in a programming language (e.g., C), or a program script (e.g., Python). Alternatively, the computer-readable storage medium can be implemented as a signal containing computer-readable data, such as a time-varying voltage signal or a radio signal.

[0028] The invention also includes combinations of the features of the described embodiments. The invention therefore also includes realizations that each exhibit a combination of the features of several of the described embodiments, provided that the embodiments have not been described as mutually exclusive.

[0029] A preferred embodiment of the invention is described in more detail below with reference to the drawing, in which: Fig. 1 schematically illustrates the components used in an embodiment of the method according to the invention and the steps they perform; Fig. 2 shows the components of a system according to the invention.

[0030] The exemplary embodiments described below are preferred embodiments of the invention. In these exemplary embodiments, the described components each represent individual features of the invention, which can be considered independently and each further develops the invention independently. Therefore, the disclosure is intended to include combinations of features of the embodiments other than those shown. Furthermore, the described embodiments can also be supplemented by further features of the invention already described.

[0031] In the figures, identical reference symbols denote functionally equivalent elements.

[0032] In the present case, it was assumed that there is a backend server 10 for a plurality of vehicles, which are in Fig. Figure 1 schematically illustrates the entire fleet 100, with the backend server 10 being able to communicate with the individual vehicles via a communication interface 11. The backend server 10 includes a forecasting device 12 and a capacity planning device 14, as well as a data processing device 20, which, as shown here, can be part of the backend server 10, but is preferably external to it. There is also a network failure detector 16 and an alarm device 18.

[0033] The vehicles in fleet 100 transmit their requests according to step S10, which are counted in backend server 10. In predictive function 12, a forecast of the expected future number of requests from fleet 100 is generated in step S12 using the least squares method, a Kalman filter, and / or an artificial intelligence (especially a neural network). The forecast can cover a period that begins between 10 seconds and 1 minute later and then lasts between 5 seconds and 1 minute after its start. The forecast is made continuously, for example, at fixed intervals of 5 seconds, 10 seconds, 15 seconds, 20 seconds, or even higher intervals, or as needed at flexible individual intervals, so that total time periods of between 5 seconds and 10 minutes are covered.

[0034] In step S14a, the predicted number is transmitted to the capacity planning unit 14, and simultaneously, in step S14b, to the network failure detector 16. The capacity planning unit 14 compares the predicted number with predetermined thresholds or reacts flexibly, and in step S16a, it sends a corresponding request to the data processing unit 20 to determine the capacity that is expected to be needed shortly. In step S16b, corresponding information is also sent to a central unit of the backend server 10. In this way, when the prediction is received, the backend server 10 has suitable capacity available in the data processing unit 20. This capacity is primarily computing capacity (processor capacity).In addition to this measure concerning the application layer, the network failure detector 16 checks whether the transport layer is still able to handle the predicted number of failures, whereby, according to step S18, the alarm output device 18 is then activated if necessary. This device can be located in a central location (operating personnel at the backend server), at the network operator's premises, or at a third location.

[0035] Out of Fig. Figure 2 shows the entire system 1. Fleet 100 comprises several vehicles 1, 2, 3, 4, and 5. System 1 also includes several backend servers 10a, 10b, and 10c, as well as a central data processing unit 20. Alternatively, there could be several data processing units 20. The essential point here is that at least two backend servers 10a, 10b, and, in this case, 10c, can access a data processing unit 20.

[0036] Depending on their geographical proximity, requests from vehicles 1, 2, 3, 4 and 5 can be directed to backend server 10a (as with vehicle 1), backend server 10b (as with vehicles 3, 4 and 5) and backend server 10c (as with vehicle 2), with vehicle 2 possibly moving in the direction of backend server 10a and already playing a role in predictions made by prediction facility 12.

[0037] It is evident that backend server 10b might require more capacity, which it queries from data processing unit 20, than, for example, backend server 10c. However, because all three backend servers 10a, 10b, and 10c can access data processing unit 20 jointly, the total capacity of this data processing unit 20 does not need to be as high as it would be in the prior art if each backend server 10a, 10b, and 10c were assigned a data processing unit 20 with a fixed capacity.

[0038] Overall, the examples show how a network failure procedure / system (measures to prevent failure; measures to take in case of failure) can be provided.

Claims

[1] Method for operating a backend server (10) which receives and / or sends data packets, in particular containers, from a plurality (100) of participants provided by motor vehicles (1, 2, 3, 4, 5), wherein the backend server (10) has access to at least partially external data processing equipment (20) and has at least one communication interface (11) for wireless communication with the vehicles, wherein the backend server (10) further comprises a prediction device (12) designed to predict the number of data packets to be transmitted in the future via the at least one communication interface (11), and finally a capacity planning device (14) designed to predict a future data processing capacity based on the predicted number, wherein the method comprises: - Predictions of the number of data packets to be processed by the prediction device during operation at a later time interval (12); - when a predicted number of data packets does not correspond to the current processor capacity of the external data processing facility (20), effect, during operation, a change in the processor capacity required by the data processing facility (20) by requesting such capacity in advance via access to the external data processing facility (20). [2] Method according to claim 1, wherein the prediction relates to the number of data packets expected to be processed in a time interval starting from 10 s to 1 min later, wherein preferably the prediction relates to the number of data packets expected to be processed in a time interval starting from 10 s to 1 min later and lasting between 5 s and 1 min. [3] Method according to claim 1 or 2, wherein predictions are continuously made anew at intervals of between 5 s and 1 min, preferably between 10 s and 30 s. [4] Method according to any one of claims 1 to 3, wherein the capacity is increased when the predicted number is greater or less than a predetermined first number than the current capacity allows. [5] Method according to any one of claims 1 to 4, wherein, when a predicted number is smaller than the current capacity is allowed by a predetermined second number, the capacity is reduced. [6] Method according to any one of claims 1 to 5, wherein an alarm is issued when a predicted number is too large for the existing bandwidth for data transmission. [7] Method according to any one of claims 1 to 6, wherein the prediction is carried out using a least squares method, a Kalman filter and / or a device for providing artificial intelligence. [8] Backend server (10) for receiving data packets from one and providing data packets to a plurality (100) of vehicles, with: - access to a data processing facility that is at least partially external (20); - at least one communication interface (11) for wireless communication with the vehicles; - a predictive device (12) designed to predict a number of data packets to be transmitted in the future via the at least one communication interface (11); - a capacity planning facility (14) designed to predict, based on the predicted number, a future data processing capacity required by the data processing facility (20) and to request such capacity in advance via access to the external data processing facility (20). [9] System (1) comprising a plurality of backend servers (10a, 10b, 10c) according to claim 8 and at least one data processing unit (20) which is accessible to multiple backend servers.

Images