Device and method for simulating a computer system in order to design, check and / or produce the computer system

Abstract

The invention relates to a method for simulating a computer system (100) in order to design, check and / or produce the computer system (100), wherein the computer system (100) comprises a first component (102) and a communication link (104); the first component (102) is designed to carry out a first function for determining an output variable of the first component (102); the first component (102) is designed to output the output variable to the communication link (104); in the simulation, the runtime of the execution of the first function is estimated, and the output variable or information on the output variable is output to the communication link (104) at the end of the estimated runtime; and the output variable is specified or estimated or the information is specified or estimated or the information is determined on the basis of the output variable.

Description

[0001] Description

[0002] title

[0003] Device and method for simulating a for the design and / or manufacture of the computer system

[0004] State of the art

[0005] The invention relates to a device and a method for simulating a computer system for the design, verification and / or manufacture of the computer system.

[0006] Conventional methods for simulating a computer system simulate the components of the computer system. This is very computationally intensive and requires a great deal of computing power and time for the design and manufacture of complex components and accurate simulation.

[0007] Disclosure of the invention

[0008] The method and apparatus for simulating a computer system according to the independent claims enable a simulation of the computer system with fewer computing resources and in a shorter time than is required for an exact simulation of the components.

[0009] The method, in particular a computer-implemented method, for simulating a computer system for the design, verification and / or manufacture of the computer system provides that the computer system comprises a first component and a communication link, wherein the first component is configured to execute a first function to determine an output variable of the first component, wherein the first component is configured to output the output variable to the communication link, wherein in the simulation the runtime of the execution of the first function is estimated and the output variable or information about the output variable is output to the communication link at the end of the estimated runtime, wherein the output variable is specified or estimated, or the information is specified or estimated or determined depending on the output variable.In simulation, estimating the runtime of the first function means that the first function is neither executed by the first component nor is its execution on the first component simulated by a model of the first component. In one example, the runtime of the first function is estimated using an estimation model, specifically artificial intelligence. The method provides a simulation interface for using the estimation model and for communication via the communication link, depending on the estimated runtime. The design or parameterization of the estimation model is not part of the method. The method assumes, for example, that an estimation model for estimating the runtime of the first function is already defined.

[0010] For example, the first component comprises a piece of hardware, where the runtime of the first function is estimated as a function of a first hardware parameter, which characterizes the hardware of the first component. Hardware parameters are variable parameters of the hardware architecture. Changing these hardware parameters of a component also changes the runtime when the software is executed on the component, or the energy consumption.

[0011] For example, the runtime of the first function without executing the first function is estimated depending on an input variable to be processed by the first function or information about an input variable of the first component to be processed by the first function.

[0012] It may be planned that a result of the simulation is determined and the computer system is designed, checked and / or manufactured depending on the result of the simulation.

[0013] The computer system may include a second component, where the first component is configured to communicate with the second component via a communication link, and the second component is configured to execute a second function. The output of the first component is transmitted as the input of the second component via the communication link at the end of the estimated runtime of the first function. Alternatively, information about the output of the first component is transmitted as the input of the second component via the communication link at the end of the estimated runtime of the first function. This means that the second function is executed after the first function and depends on the result of the first function.

[0014] It can be provided that the computer system includes a second component, wherein the first component is configured to communicate with the second component via the communication link, wherein the second component is configured to execute a second function to determine an output variable of the second component, wherein the first function to determine the output variable of the first component is configured as a function of an input variable of the first component, wherein the output variable of the second component is transmitted as an input variable of the first component from the second component to the first component via the communication link, and the runtime of the first function is estimated as a function of the input variable.or where information about the output of the second component is transmitted as information about the input of the first component via the communication link from the second component to the first component, and the runtime of the first function is estimated depending on the information about the input. This means that the first function is executed after the second function and depends on the result of the second function.

[0015] The execution of the second function by the second component is simulated in an example. This means that the runtime of the second function is not estimated, but rather results from the execution of the second function itself. The method thus provides a simulation interface for using the second component and for communication via the communication link, depending on the actual runtime. Therefore, the method does not require an estimation model for estimating the runtime of the second function.

[0016] In one example, the simulation estimates the execution time of the second function and outputs the output of the second component, or information about an output of the second component, to the communication link at the end of the estimated execution time. The output of the second component can be predetermined or estimated, or the information about the output of the second component can be predetermined or estimated, or determined depending on the output of the second component. This means that both the execution time of the first function and the execution time of the second function are estimated. For example, the execution time of the second function is estimated using an estimation model, particularly artificial intelligence.The procedure provides the simulation interface for using the estimation model and for communication via the communication link, depending on the estimated runtime. The setup or parameterization of the estimation model is not part of the procedure. For example, the procedure assumes that an estimation model for estimating the runtime of the second function is already defined.

[0017] In one example, the runtime without executing the second function is estimated based on the input of the second component or based on information about the input of the second component. Executing the second function is unnecessary because the runtime is estimated. The runtime depends on the input because the function processes the input. If the input is provided to the second component in the simulation, the runtime is estimated based on the input. The input is provided, for example, by a component of computer system 102 that is executed in the simulation. It is not necessary for the input itself to be calculated in the simulation.If the input variable in the computing system is provided by a component whose runtime is estimated in the simulation, information about the input variable, rather than the input variable itself, can be used to estimate the runtime of the second function. For example, the second component comprises hardware, and the runtime of the second function is estimated as a function of a second hardware parameter, where the second hardware parameter characterizes the hardware of the second component. Hardware parameters are variable parameters of the hardware architecture. Changing these hardware parameters of a component also changes the runtime when the software is executed on the component, or the energy consumption.

[0018] It can be provided that the computer system includes a third component, wherein the first component is configured to communicate with the third component via the communication link, wherein the third component is configured to execute a third function to determine an output variable of the third component, wherein the first function to determine the output variable of the first component is configured as a function of an input variable of the first component, wherein the output variable of the third component is transmitted as an input variable of the first component from the third component to the first component via the communication link, and the runtime of the first function is estimated as a function of the input variable.or wherein information about the output of the third component is transmitted as information about the input of the first component via the communication link from the third component to the first component, and the runtime of the first function is estimated as a function of the information about the input, wherein the execution of the third function by the third component is simulated in the simulation.

[0019] The device for simulating the computer system for the design, verification and / or manufacture of the computer system, which includes the first component and the communication link, includes the communication link and is configured to execute the procedure for the first component.

[0020] The device for simulating the computer system, which comprises the first component, the second component, and the communication link, is configured to execute the procedure for the first component and the second component. The device for simulating the computer system, which comprises the first component, the second component, the third component, and the communication link, is configured to execute the procedure for the first component, the second component, and the third component.

[0021] A computer program may be provided that includes instructions executable by a computer, and when these instructions are executed by the computer, the computer performs the procedure.

[0022] Further advantageous embodiments can be found in the following description and the drawing. The drawing shows:

[0023] Fig. 1 shows a schematic representation of a computer system.

[0024] Fig. 2 shows a schematic representation of a device for determining the runtime of the computer system for the design, verification and / or manufacture of the computer system.

[0025] Fig. 3 shows a flowchart with steps of a procedure for determining the runtime of the computer system for the design, verification and / or manufacture of the computer system.

[0026] Figure 1 shows a schematic representation of a computer system 100.

[0027] The Computer System 100 comprises hardware and software. The Computer System 100 can include a chiplet, a multi-core processor, and / or a system-on-a-chip. The Computer System can also be an embedded system.

[0028] The hardware in this example comprises 102 different components. In this example, software is provided for each of the 102 components.

[0029] For example, the computer system 100 comprises a component 102 configured for processing input signals, in particular a sensor configured for processing input signals. For example, the computer system 100 comprises a component 102, in particular a power amplifier configured for outputting output signals, in particular for controlling an actuator.

[0030] For example, the computer system 100 includes a component 102, in particular a central processing unit, CPU, which is designed to determine the output signals depending on the input signals.

[0031] For example, the computer system 100 includes a component 102, in particular a graphics processing unit, GPU, which can be used to perform calculations, especially in parallel, to determine the output signals depending on the input signals.

[0032] For example, the computer system 100 includes a component 102 that can be used to perform operations on the hardware, particularly in parallel, to determine the output signals depending on the input signals. An example of the component 102 that can perform operations on the hardware is a hardware accelerator, which is configured, for example, to perform a convolution operation on values ​​of the input signal or a part of the input signal.

[0033] For example, the computer system 100 comprises a component 102 configured for the permanent storage of instructions for determining the output signals depending on the input signals, in particular a non-volatile memory configured for the permanent storage of the instructions. It may be provided that the component 102 configured for permanent storage is also configured for storing and providing the results of the execution of the instructions.

[0034] For example, the computer system 100 includes a component 102 configured for the volatile storage of instructions for determining the output signals depending on the input signals, in particular a non-volatile memory configured for the permanent storage of the instructions. It may be provided that the component 102 configured for volatile storage is also configured for storing and providing the results of the execution of the instructions. The instructions define one or more functions that can be executed by a respective component 102 of the computer system 100, e.g., to determine the output signals depending on the input signals. A respective function is configured to determine an output value of the function depending on an input value of the function. The function may, for example, store an output value of another function on the hardware of component 102, e.g.,The function may involve storing data in permanent or volatile memory. For example, it may involve providing an input value to another function from the hardware of component 102, such as permanent or volatile memory. Alternatively, the function may involve calculating an output value on the hardware of component 102, such as the CPU, GPU, or hardware accelerator, depending on an input value of the function.

[0035] Determining the output signals based on the input signals is one example of a task that Computer System 100 can perform. Computer System 100 can also be configured to perform other tasks. Further examples include processing pixels of a digital image or file, and saving a digital image or file.

[0036] Depending on the task, computer system 100 can comprise various components 102. In this example, computer system 100 comprises the functions or components 102 required to perform the task. It may also be provided that computer system 100 includes additional functions or components 102 that are not required to perform the task. Furthermore, computer system 100 may be designed to perform various tasks.

[0037] In this example, the instructions are executed in computer system 100, with software of components 102 running on the respective hardware of components 102. The instructions can be defined in the hardware and / or implemented in the software. In this example, the hardware includes a communication link 104 between the components 102. This communication link 104 is configured to enable communication between the components 102. The components 102 are configured to communicate with each other via this communication link 104.

[0038] Figure 2 schematically depicts a device 200 for determining the runtime of the computer system 100 for the design, verification, and / or manufacture of the computer system 100. In this example, the device 200 comprises a computing unit 202 configured to perform a simulation of the computer system 100, in which part of the components 102 are simulated, and another part of the components 102 executes the software.

[0039] The device 200 is designed to carry out a method for determining the runtime of the computer system 100 for the design, verification and / or manufacture of the computer system 100.

[0040] The method and device 200 are described using the example of a given hardware and a given software.

[0041] The procedure for determining the runtime of computer system 100 is carried out in the simulation.

[0042] In the procedure for determining the runtime of computer system 100, the software of a first part of the components 102 and the communication link 104 are executed on the hardware of the respective components 102. In the procedure for determining the runtime of computer system 100, the runtime of a second part of the components 102 is estimated by an estimation model 204 of the respective component 102. This means that neither the hardware nor the software of the components 102 of the second part are used in the procedure for determining the runtime of the computer system.

[0043] Estimation model 204 of the respective component 102 estimates the runtime on the respective component 102. Estimation model 204 of the respective component 102 includes an artificial intelligence trained to estimate the runtime on the respective component 102.

[0044] The estimation model 204 is described using an exemplary component 102 and is designed for the other components 102 of the second part of the components 102, which are modeled by a respective estimation model 204, as described for the exemplary component 102.

[0045] Component 102 is configured to execute the software. The software is defined by code. In this example, the code includes at least one function configured to process an input variable transmitted to component 102 via communication link 104. This function is configured to determine an output variable, to be transmitted to another component 102 via communication link 104, depending on the input variable.

[0046] Estimation model 204 is designed to estimate the runtime as a function of a hardware parameter that characterizes the hardware of component 102. The hardware parameter is, for example, a latency, energy consumption, or the maximum possible throughput of a component of the hardware.

[0047] Estimation model 204 includes, for example, a trained mathematical model. This trained mathematical model is capable of estimating the runtime as a function of the hardware parameter.

[0048] The estimation model 204 is defined in an example by parameter 206 of the artificial intelligence.

[0049] The artificial intelligence 204 is trained, for example, with training data that includes an input variable and a runtime reference. The estimating model 204 includes parameters 206. These parameters 206 of the model 204 are learned on the training data in such a way that the runtime estimated by the estimating model 204 for the input variable approximates the runtime reference as closely as possible. The runtime reference is determined, for example, by measuring the runtime during the execution of the function for determining the output variable for the input variable on the hardware of component 102.

[0050] The reference for the runtime is determined, for example, by measuring the runtime during a simulation of the execution of the function to determine the output variable for the input variable on the hardware of component 102.

[0051] When the hardware changes, the hardware parameter changes. When the hardware changes, it is sufficient to train the estimation model 204 with a few new training data points or a small portion of the same training data, depending on the changed parameter.

[0052] The software is interchangeable after training the estimation model 204, and a corresponding estimation using the unchanged estimation model 204 is possible for the replaced software.

[0053] The estimation model 204, which models the exemplary component 102, is trained in the example independently of the other components 102 of the computer system.

[0054] The estimation model 204 is designed to output the output variable when the runtime estimated by the artificial intelligence 204 has been reached since the arrival of the input variable at the artificial intelligence 204.

[0055] In this example, the input variable contains information about data to be processed by the function. The output variable in this example contains information about data to be processed by a function of another component 102.

[0056] The data includes, for example, the pixels of an image. The function includes, for example, processing the pixels of the image. The information that includes the input size includes, for example, the number of pixels in the image. The artificial intelligence 204 is, for example, trained to estimate the runtime for processing the pixels depending on the number of pixels in the image. The artificial intelligence 204 is, for example, trained to estimate the runtime for processing the pixels for images with different numbers of pixels, depending on the number of pixels in each image. The function reduces, for example, the number of pixels in the image. The information that includes the output size includes, for example, the reduced number of pixels in the image.

[0057] The data can also be of a different type than pixels. For example, the data could represent a file or a time series containing values ​​of a digital signal. The data that the function processes and the data that the function generates by processing the data can be of different types.

[0058] The estimation model 204 is surrounded by a simulation interface 208. The simulation interface 208 is configured for communication with components 102 of the first part of the components 102 via the communication link 104. The simulation interface 208 is configured for communication with a respective simulation interface 208 that encompasses the estimation model 204 of a component 102 of the second part of the components 102.

[0059] In this example, the device 200 includes the simulation interface 208. The computing unit 202 is configured to communicate via the simulation interface 208 and via the communication link 104. In this example, the computing unit 202 provides the simulation interface 208.

[0060] In this example, the simulation interface 208 of component 102 is configured to send information about the output variable, or the output variable itself, via communication link 104, depending on whether the output variable is received. In this example, the simulation interface 208 of component 102 is configured to send information about the output variable via communication link 104 to the simulation interface 208 of another component 102 if that other component 102 is the receiver of the output variable. In this example, the simulation interface 208 of each component 102 is configured to generate a substitute variable for the function's output variable and send it via communication link 104 to another component 102 on which the software is running, if that other component 102 is the receiver of the output variable.It can be provided that the artificial intelligence 204 is trained to estimate the substitute quantity. It can be provided that the simulation interface 208 is trained to receive the input quantity. In the example, the simulation interface 208 is trained to receive the input quantity from a component 102 on which the software runs. In the example, the simulation interface 208 is trained to receive information about the input quantity from the simulation interface 208 of another component 102. It can be provided that the artificial intelligence 204 is trained to estimate the output quantity depending on the input quantity. This eliminates the need to actually calculate the output quantity. This speeds up the simulation.

[0061] In this example, the method is performed on the device 200 to determine the runtime of the computer system 100. The device 200 is configured to communicate with the first part of the components 102, on which the software is executed, via the communication link 104. The device 200 is configured to simulate the second part of the components 102.

[0062] The components 102 are designed to each provide at least one function of the computer system 100.

[0063] In the procedure, the runtime of the respective function or functions of the components 102 of the second part is estimated by a respective estimation model 204.

[0064] The respective component 102 of the first part is integrated into the simulation. The runtime of the respective function(s) of component 102 of the first part is determined by executing the software on the respective component 102 of the first part.

[0065] Figure 3 shows a flowchart with steps of the procedure.

[0066] The procedure includes a step 302. In step 302, the components 102 of the first part of the components 102 are made available for simulation.

[0067] The hardware of each component 102 and the software of each component 102 are designed to perform the function or functions of each component 102.

[0068] It may be planned that no component 102, or only one, is provided on which the software is actually supposed to run.

[0069] The procedure includes step 304.

[0070] In step 304, the respective estimation model 204 is provided for the simulation of the components 102 of the second part of components 102. In the example, the respective artificial intelligence is provided.

[0071] The estimation model 204, which is provided for the respective component 102, is designed to estimate the runtime of the function or the runtimes of the functions of the respective component 102.

[0072] It may be provided that only the runtime for one component 102 is estimated and the estimation model 204, e.g. the artificial intelligence, is provided only for this component 102.

[0073] The procedure includes step 306.

[0074] In step 306, the computer system 100 is simulated.

[0075] In this example, the simulation is started at a first point in time.

[0076] The simulation simulates the completion of a task by the computer system 100.

[0077] Instead of actually performing the task with the computer system 100, the communication via the communication link 104 is simulated to determine the runtime of the computer system 100 in the order in which the communication would be executed by the respective software on the respective components 102.

[0078] During the simulation, the components 102 of the first part of the components 102 communicate via communication link 104. During the simulation, the components 102 of the second part of the components 102 communicate via their respective simulation interface 208 and communication link 104. During the simulation, the execution of the respective function on the respective component 102 of the first part of the components 102 is simulated.

[0079] The respective estimation model 204 does not fully simulate the execution of the respective function of the respective component 102 of the second part of the components 102, but rather estimates the runtime required for the respective function to be executed by the respective component 102. Therefore, for a function whose complete simulation takes longer than the execution of the function by the component 102, the method is faster than if the function were fully simulated.

[0080] This means that the execution of a function by a component 102 is either simulated in the simulation or replaced by the result of an estimate using the estimation model 204. With multiple components 102, it may be provided that a component 102 from the first part of components 102 communicates with one or more components 102 from the first part 102 or with one or more components 102 from the second part.

[0081] The simulation provides, for example, real-world execution time values, especially with small deviations. The simulation itself does not run in the same amount of time as the actual execution would take on the hardware.

[0082] The procedure provides, for example, that the estimation model 204 estimates the runtime for the execution of the function upon arrival of the input variable or information about the input variable at the simulation interface 208. The procedure also provides, for example, that the output variable or information about the output variable is output via the simulation interface 208 when the estimated runtime has been reached.

[0083] The procedure provides, for example, that the output variable is estimated by estimation model 204, or that the information about the output variable is estimated by the model. It can be provided that the output variable or the information about the output variable is estimated by estimation model 204 independently of the input variable or independently of the information about the input variable. It can also be provided that a predefined output variable or predefined information about the output variable is output independently of the input variable or independently of the information about the input variable.

[0084] The procedure includes step 308.

[0085] In step 308, the runtime of computer system 100 is determined.

[0086] In the example, a second point in time is determined at which the simulation of the task completion is finished.

[0087] In the example, the runtime of computer system 100 is determined as the difference between the second time point and the first time point.

[0088] The procedure includes step 310.

[0089] Step 310 outputs the runtime of the computer system 100.

[0090] The method for determining runtime is used, for example, in the design of computer system 100. The design of computer system 100 may stipulate that the runtime is estimated on different hardware using this method. The design of computer system 100 may also stipulate that the runtime is estimated using this method with different software. The different software differs, for example, in the code that implements the function. The design of computer system 100 may further stipulate that the runtime is estimated using this method with different software on different hardware.

[0091] The method for determining runtime is used, for example, instead of measuring the runtime using different software on the same hardware. It may be stipulated that the method be repeated after a change in hardware and / or software.

[0092] The procedure for determining the runtime of computer system 100 is carried out in a simulation. Simulation is used for the design, verification, and / or manufacture of computer system 100. This means that, for the design of computer system 100, the procedure additionally stipulates that the computer system 100 is designed based on the simulation results. This means that, for the verification of computer system 100, the procedure additionally stipulates that the computer system 100 is verified based on the simulation results. This means that, for the manufacture of computer system 100, the procedure additionally stipulates that the computer system 100 is manufactured based on the simulation results.

[0093] For example, one of the hardware parameters of computer system 100 is designed or verified in the simulation. For instance, computer system 100 is built with the hardware parameter provided by the simulation result.

Claims

Claims 1. A method, in particular a computer-implemented method, for simulating a computer system (100) for the design, verification and / or manufacture of the computer system (100), characterized in that the computer system (100) comprises a first component (102) and a communication link (104), wherein the first component (102) is configured to execute a first function for determining an output variable of the first component (102), wherein the first component (102) is configured to output the output variable to the communication link (104), wherein in the simulation the runtime of the execution of the first function is estimated and the output variable or information about the output variable is output to the communication link (104) at the end of the estimated runtime (306), wherein the output variable is specified or estimated, or the information is specified or estimated or determined depending on the output variable.

2. Method according to claim 1, characterized in that the first component (102) comprises hardware, wherein the runtime of the first function is estimated depending on a first hardware parameter (306), wherein the first hardware parameter characterizes the hardware of the first component (102).

3. Method according to one of the preceding claims, characterized in that the runtime of the first function without executing the first function is estimated depending on an input variable to be processed by the first function or information about an input variable of the first component (102) to be processed by the first function (306).

4. Method according to one of the preceding claims, characterized in that a result of the simulation is determined and the computer system (100) depending on the result of the simulation, it is designed, checked and / or manufactured.

5. A method according to one of the preceding claims, characterized in that the computer system (100) comprises a second component (102), wherein the first component (102) is configured to communicate with the second component (102) via the communication link (104), wherein the second component (102) is configured to execute a second function, wherein the output variable of the first component (102) is transmitted from the first component (102) to the second component (102) via the communication link (104) at the end of the estimated runtime of the first function as an input variable (102) of the second component (102) (306), or wherein the information about the output variable of the first component (102) is transmitted from the first component (102) to the second component (102) via the communication link (104) at the end of the estimated runtime of the first function as information about the input variable (102) of the second component (102) (306).

6. A method according to any one of claims 1 to 4, characterized in that the computer system (100) comprises a second component (102), wherein the first component (102) is configured to communicate with the second component (102) via the communication link (104), wherein the second component (102) is configured to execute a second function for determining an output variable of the second component (102), wherein the first function for determining the output variable of the first component is configured as a function of an input variable of the first component (102), wherein the output variable of the second component (102) is transmitted as an input variable (102) of the first component (102) from the second component (102) to the first component (102) via the communication link (104), and the runtime of the first function is estimated as a function of the input variable (306).or wherein information about the output of the second component (102) is transmitted as information about the input (102) of the first component (102) via the communication link (104) from the second component (102) to the first component (102) and the runtime of the first function is estimated as a function of the information about the input (306).

7. Method according to one of claims 5 or 6, characterized in that the execution of the second function by the second component (102) is simulated in the simulation (306).

8. Method according to one of claims 5 or 6, characterized in that in the simulation the runtime of the execution of the second function is estimated and the output variable of the second component (102) or the information about an output variable of the second component (102) is output to the communication link (104) at the end of the estimated runtime of the execution of the second function (306), wherein the output variable of the second component is specified or estimated, or the information about the output variable of the second component is specified or estimated or determined depending on the output variable of the second component (306).

9. Method according to claim 8, characterized in that the runtime without executing the second function is estimated depending on the input variable of the second component (102) or depending on the information about the input variable of the second component (102) (306).

10. Method according to claim 8 or 9, characterized in that the second component (102) comprises hardware, wherein the runtime of the second function is estimated depending on a second hardware parameter (306), wherein the second hardware parameter characterizes the hardware of the second component (102).

11. Method according to one of claims 8 to 10, characterized in that the computer system (100) comprises a third component (102), wherein the first component (102) is configured to communicate with the third component (102) via the communication link (104), wherein the third component (102) is configured to execute a third function for determining an output variable of the third component (102), wherein the first function for determining the output variable of the first component depends on an input variable of the first component (102). is formed, wherein the output variable of the third component (102) is transmitted as the input variable (102) of the first component (102) via the communication link (104) from the third component (102) to the first component (102) and the runtime of the first function is estimated as a function of the input variable (306), or wherein information about the output variable of the third component (102) is transmitted as information about the input variable (102) of the first component (102) via the communication link (104) from the third component (102) to the first component (102) and the runtime of the first function is estimated as a function of the information about the input variable (306), wherein the execution of the third function by the third component (102) is simulated in the simulation (306).

12. Device (200) for simulating a computer system (100) for the design, verification and / or manufacture of the computer system (100), characterized in that the computer system (100) comprises a first component (102) and a communication link (104), wherein the device (200) comprises the communication link (104), and wherein the device (200) is configured to perform the method according to one of claims 1 to 4.

13. Device (200) according to claim 12, characterized in that the computer system (100) comprises a second component (102), and wherein the device (200) is configured to perform the method according to any one of claims 4 to 10.

14. Device (200) according to claim 12 or 13, characterized in that the computer system (100) comprises a third component (102), and wherein the device (200) is configured to perform the method according to claim 11.

15. Computer program, characterized in that the computer program comprises instructions executable by a computer, the execution of which by the computer performs the method according to one of claims 1 to 11.

Images