Power supply method, task execution system and device, electronic equipment and medium

Abstract

The invention provides a power supply method, a task execution system, a power supply device, electronic equipment, a storage medium and a program product, and relates to the technical field of chips, in particular to the technical field of power supply, the technical field of artificial intelligence and the like. According to the specific implementation scheme, based on task attribute information of a to-be-processed task sequence, initial task configuration information for executing the task sequence on task execution equipment is determined, and the task sequence comprises a plurality of tasks with an execution sequence; on the basis of the initial task configuration information, under the condition that it is determined that a power consumption hopping phenomenon exists in the task sequence, the initial task configuration information is updated, and task configuration information is obtained; and sending the task configuration information to the power supply management equipment, so that the power supply management equipment outputs electric energy matched with the task configuration information to the task execution equipment.

Description

Technical Field

[0001] The present disclosure relates to the field of chip technologies, in particular to the fields of power supply technologies, artificial intelligence technologies, etc., and specifically relates to a power supply method, a task execution system, a power supply device, an electronic device, a storage medium, and a program product. Background Art

[0002] With the development of artificial intelligence technologies, the computing power requirements for chips used to execute artificial intelligence tasks are continuously increasing. As a result, the power consumption of a single chip is soaring continuously. Especially during the process of using the multi-core processor of a chip to execute parallel computing tasks, the working state of the chip will continuously switch between light load and heavy load, which easily makes it difficult to maintain the stable state of the power supply end and easily causes voltage oscillation. Summary of the Invention

[0003] The present disclosure provides a power supply method, a task execution system, a power supply device, an electronic device, a storage medium, and a program product.

[0004] According to one aspect of the present disclosure, there is provided a power supply method, including: determining initial task configuration information for executing the task sequence on a task execution device based on task attribute information of a task sequence to be processed, where the task sequence includes a plurality of tasks with an execution order; updating the initial task configuration information to obtain task configuration information in the case that it is determined based on the initial task configuration information that there is a power consumption jump phenomenon in the task sequence; and sending the task configuration information to a power supply management device so that the power supply management device outputs electric energy matching the task configuration information to the task execution device.

[0005] According to another aspect of the present disclosure, there is provided a power supply method, including: receiving task configuration information from the task execution device, where the task configuration information is obtained by determining initial task configuration information for executing the task sequence on the task execution device based on task attribute information of the task sequence to be processed, where the task sequence includes a plurality of tasks with an execution order, and updating the initial task configuration information in the case that it is determined based on the initial task configuration information that there is a power consumption jump phenomenon in the task sequence; and outputting electric energy matching the task configuration information to the task execution device.

[0006] According to another aspect of the present disclosure, a task execution system is provided, including: a task execution device configured to determine initial task configuration information for executing the task sequence based on task attribute information of the task sequence to be processed, where the task sequence includes a plurality of tasks with an execution order; in a case where it is determined based on the initial task configuration information that there is a power consumption jump phenomenon in the task sequence, updating the initial task configuration information to obtain task configuration information; and a power supply management device configured to receive the task configuration information from the task execution device and output electric energy matching the task configuration information to the task execution device.

[0007] According to another aspect of the present disclosure, a power supply device is provided, including: a configuration module configured to determine initial task configuration information for executing the task sequence on a task execution device based on task attribute information of the task sequence to be processed, where the task sequence includes a plurality of tasks with an execution order; an update module configured to update the initial task configuration information to obtain task configuration information in a case where it is determined based on the initial task configuration information that there is a power consumption jump phenomenon in the task sequence; and a sending module configured to send the task configuration information to a power supply management device so that the power supply management device outputs electric energy matching the task configuration information to the task execution device.

[0008] According to another aspect of the present disclosure, a power supply device is provided, including: a receiving module configured to receive task configuration information from the task execution device, where the task configuration information is determined by determining initial task configuration information for executing the task sequence on the task execution device based on task attribute information of the task sequence to be processed, where the task sequence includes a plurality of tasks with an execution order, and updating the initial task configuration information in a case where it is determined based on the initial task configuration information that there is a power consumption jump phenomenon in the task sequence; and a power supply module configured to output electric energy matching the task configuration information to the task execution device.

[0009] According to another aspect of the present disclosure, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; where the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method as described above.

[0010] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause the computer to execute the method as described above.

[0011] According to another aspect of the present disclosure, there is provided a computer program product including a computer program which, when executed by a processor, implements the method as described above.

[0012] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. Description of the Drawings

[0013] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0014] Figure 1A Schematically shows a block diagram of a task execution system according to an embodiment of the present disclosure;

[0015] Figure 1B Schematically shows a block diagram of a task execution system according to a related example of the present disclosure;

[0016] Figure 1C Schematically shows a schematic diagram of steady-state regulation of a task execution system according to a related example of the present disclosure;

[0017] Figure 1D Schematically shows a schematic diagram of steady-state regulation of a task execution system according to another related example of the present disclosure;

[0018] Figure 2 Schematically shows a block diagram of a task execution system according to another embodiment of the present disclosure;

[0019] Figure 3 Schematically shows a flowchart of a power supply method according to an embodiment of the present disclosure;

[0020] Figure 4A Schematically shows a schematic diagram of a task sequence with a power consumption jump phenomenon according to an embodiment of the present disclosure;

[0021] Figure 4B Schematically shows a schematic diagram of a task sequence without a power consumption jump phenomenon according to an embodiment of the present disclosure;

[0022] Figure 5 Schematically shows a block diagram of a task execution device according to an embodiment of the present disclosure;

[0023] Figure 6 Schematically shows a flowchart of a power supply method according to another embodiment of the present disclosure;

[0024] Figure 7 Schematically shows a block diagram of a power supply device according to an embodiment of the present disclosure;

[0025] Figure 8A block diagram schematically shows a power supply device according to another embodiment of the present disclosure; and

[0026] Figure 9 A block diagram of an electronic device suitable for implementing a power supply method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0027] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0028] Figure 1A The block diagram of the task execution system according to an embodiment of the present disclosure is schematically shown.

[0029] like Figure 1A As shown, the task execution system may include a task execution device 110, such as a chip, and a power management device 120 (also known as a voltage regulator module, VRM) that provides power to the chip. The task execution system may also include a power distribution network (PDN) 130 for electrically connecting the task execution device and the power management device.

[0030] The power management device 120 is connected to a power source and is used to receive high-voltage direct current and convert it into low-voltage direct current that meets the application requirements of the task execution device.

[0031] The power distribution network 130 is used to transmit the power output by the power management device 120 to the task execution device 110. The goal is to control the power supply noise within a very small tolerance range and respond to the rapid changes in current caused by the load in real time, thereby providing clean and stable voltage for the chip and a low-impedance return path for other signals.

[0032] With the development of artificial intelligence (AI) technology, the demand for chip computing power continues to grow, leading to a surge in the power consumption of individual chips. This is especially true when using a chip's multi-core processor to execute parallel computing tasks. The chip's operating state constantly switches between light and heavy loads. This can cause transient changes in the current rate of change (dI / dt) caused by short-term high current behavior, which can cause large voltage fluctuations in power management equipment. If improperly controlled, the chip voltage can easily drop below the minimum voltage required for logic circuit operation, causing malfunctions in the task execution system.

[0033] The following will analyze from multiple embodiments how to solve the problem of abnormal steady state of the task execution system caused by the unknown working state of the chip.

[0034] Figure 1B A block diagram of a task execution system according to related examples of the present disclosure is schematically shown.

[0035] As Figure 1B shown, the task execution system includes a task execution device 110 and a power supply management device 120.

[0036] Different from the task execution system as Figure 1A shown, the task execution system as Figure 1B shown further includes a voltage feedback circuit 140. One end of the voltage feedback circuit 140 is connected to the transmission path, and the other end of the voltage feedback circuit 140 is connected to the power supply management device 120, and is used to transmit the detected load feedback signal of the task execution device 110 to the power supply management device 120, so that the power supply management device 120 determines the load information based on the load feedback information. Thus, the output voltage of the power supply management device 120 to the task execution device 110 is dynamically adjusted.

[0037] As Figure 1B shown, for the task execution system as

[0038] Figure 1C shown and the adjustment method of the output voltage, the existing problem is that: there are intermediate devices such as inductors and capacitors in the power distribution network and the like on the transmission path between the task execution device and the power supply management device. The existence of the intermediate devices causes the load feedback signal to be transmitted to the power supply management device relatively late after the load current changes, and there will be a delay in making the output voltage adjustment action based on the received load feedback signal. Moreover, the effective loop control bandwidth of the power supply management device is at most a few hundred KHz, and it cannot respond in time to some rapidly changing current demands.

[0039] As Figure 1C shown, the number of cores that can work simultaneously in the multi-core processor in the task execution device can be limited to limit the increased second load power consumption to be less than the full load power consumption, thereby avoiding the occurrence of power consumption jump phenomenon.

[0040] Figure 1D A schematic diagram of steady state regulation of a task execution system according to another related example of the present disclosure is schematically shown.

[0041] As Figure 1DAs shown, multiple cores in a multi-core processor in a task execution device can be started with a staged delay. For example, when multiple cores start simultaneously, there will be a power consumption jump phenomenon from the first load power consumption to the second load power consumption. By starting multiple cores in stages, the first load power consumption slowly changes to the second load power consumption, ..., the Nth load power consumption, where N is an integer greater than or equal to 2. This slows down the change in load power consumption and avoids the occurrence of power consumption jump phenomenon.

[0042] Such as Figure 1C and 1D The steady-state regulation methods shown, although they can avoid the occurrence of power consumption jump phenomenon and regulate the task execution system to maintain a stable state, all have problems including: a great loss to the performance of the chip and failure to exert the maximum performance of the chip.

[0043] In view of this, the present disclosure provides a power supply method, device, system, electronic device, storage medium and program product, in order to solve the problem of exerting the maximum performance of the chip while the task execution system is in a stable state during task execution.

[0044] Figure 2 Schematically shows a block diagram of a task execution system according to an embodiment of the present disclosure.

[0045] Such as Figure 2 As shown, the task execution system 200 may include: a task execution device 210 and a power supply management device 220, and also includes a power distribution network 230 and a communication bus 240.

[0046] The task execution device 210 may include a power consumption manager 211 and a multi-core processor 212. The power consumption manager 211 is used to determine the initial task configuration information for executing the task sequence based on the task attribute information of the task sequence to be processed, where the task sequence includes multiple tasks with an execution order; and update the initial task configuration information to obtain the task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information. The multi-core processor 212 can perform the task processing operations corresponding to the task sequence based on the task configuration information.

[0047] Optionally, the task execution device 210 may be an artificial intelligence (AI) chip, including one or a combination of processors such as a central processing unit (CPU), a graphics processing unit (GPU), a neural network processing unit (NPU), a deep learning processing unit (DPU), a tensor processing unit (TPU), and an intelligent processing unit (IPU).

[0048] The power supply management device 220 may include a voltage regulation circuit 221 and an intelligent decision maker 222. The intelligent decision maker 222 may be configured to receive task configuration information from the task execution device 210 and generate a voltage output policy based on the task configuration information. The intelligent decision maker 222 transmits the voltage output policy to the voltage regulation circuit 221 to output electrical energy matching the task configuration information to the task execution device 210 through the power distribution network 230 when the multi-core processor 212 performs task processing operations.

[0049] The power distribution network 230 may be configured to distribute the electrical energy output by the power supply management device 220 to the task execution device 210.

[0050] The communication bus 240 is used for transmitting operation instructions and task configuration information between the power supply management device 220 and the task execution device 210. Preferably, transmitting the task configuration information or other operation instructions through the communication bus 240 can achieve time synchronization between the power supply management device 220 and the task execution device 210.

[0051] Sending the task configuration information to the power supply management device so that the power supply management device outputs electrical energy matching the task configuration information to the task execution device can predict in advance the load condition of the task execution device in a future period by predicting the initial task configuration information based on the task attribute information, avoiding the problem of lag adjustment. In addition, based on the identification of the power consumption jump phenomenon in the task sequence and updating the initial task configuration information when the task sequence has a power consumption jump phenomenon, the running stability of the task execution device can be improved, while improving the power supply timeliness of the power supply management device and enhancing the stability and intelligence of power supply regulation.

[0052] The following will further explain how to achieve the above effects through the flowchart of the power supply method as Figure 3 shown.

[0053] It should be noted that the serial numbers of the respective operations in the following methods are only used as representations of the operations for description purposes and should not be regarded as indicating the execution order of the respective operations. Unless explicitly stated, the method does not need to be executed exactly in the order shown.

[0054] Figure 3 A flowchart of a power supply method according to an embodiment of the present disclosure is schematically shown.

[0055] As Figure 3 shown, the method includes operations S310 to S330.

[0056] In operation S310, based on the task attribute information of the task sequence to be processed, the initial task configuration information for executing the task sequence on the task execution device is determined.

[0057] In operation S320, when it is determined, based on the initial task configuration information, that there is a power consumption jump phenomenon in the task sequence, the initial task configuration information is updated to obtain the task configuration information.

[0058] In operation S330, the task configuration information is sent to the power supply management device so that the power supply management device outputs electric energy matching the task configuration information to the task execution device.

[0059] As Figure 3 shown, the execution subject of the power supply method may include a task execution device. Optionally, the task execution device may include a multi-core processor and a power consumption manager. The multi-core processor is used to execute the task sequence to be processed. The power consumption manager is used to execute the above operations S310 to operation S330.

[0060] The task sequence includes multiple tasks with an execution order. For example, the execution order is successively task A executed first, task B executed second, and task C executed third.

[0061] The task attribute information may include multiple task attributes having an order relationship matching the task sequence. For example, task attribute A of task A, task attribute B of task B, and task attribute C of task C. The task attributes may include information such as task type, task name, operator used, tool called, etc. As long as it is information for describing the task to be processed.

[0062] The initial task configuration information can be the operating information required to execute a task on a task execution device, predicted based on task attribute information. For example, the initial task configuration information can include load information, running duration, kernel information, etc., as long as it is used to describe the information required to execute the task. Optionally, the initial task configuration information can include multiple initial task configuration sub-information with an order relationship that matches the task sequence. For example, the initial task configuration sub-information A of task A, the initial task configuration sub-information B of task B, and the initial task configuration sub-information C of task C.

[0063] The initial task configuration information can be determined based on the task attribute information of the task sequence. The task sequence is a plurality of tasks to be processed. By predicting the initial task configuration information in advance, the initial task configuration information can be sent to the power supply management device so that the power supply management device can be aware of the load information required by the task execution device in advance, and then provide the adapted electric energy, thereby avoiding the problem of lag in power supply regulation.

[0064] Optionally, based on the initial task configuration information, the task sequence can be identified for power consumption jump phenomenon to obtain an identification result. In the case where the identification result indicates that the task sequence has a power consumption jump phenomenon, the initial task configuration information can be updated to obtain the task configuration information.

[0065] The power consumption jump phenomenon can refer to a large mutation in power consumption in a transient state. For example, when task A is completed and task B is about to be executed, since the load information required by task A and task B changes from light load to heavy load, a power consumption jump occurs, then there is a power consumption jump phenomenon in the task sequence.

[0066] The initial task configuration information can be updated. For example, the execution order of task B and task C can be updated so that the power consumption change during the task transition gap when task A, task C, and task B complete one task and switch to execute the next task is slowed down, and the initial task configuration information with a power consumption jump phenomenon is updated to the task configuration information that avoids the occurrence of a power consumption jump phenomenon. However, it is not limited to this. The load demand of task B can also be updated so that the power consumption of the updated task B does not mutate with that of task A. As long as it is an update method that can slow down the power consumption change.

[0067] Send the task configuration information to the power supply management device so that the power supply management device can output electrical energy matching the task configuration information to the task execution device. By predicting the initial task configuration information based on the task attribute information, the load condition of the task execution device in the future time period can be predicted in advance, avoiding the problem of lag adjustment. In addition, based on the recognition of the power consumption jump phenomenon in the task sequence and updating the initial task configuration information when there is a power consumption jump phenomenon in the task sequence, the running steady state of the task execution device can be improved, while improving the power supply timeliness of the power supply management device and enhancing the stability and intelligence of power supply regulation.

[0068] According to an embodiment of the present disclosure, before performing operation S320 as shown in Figure 3 the power supply method may further include: based on the initial task configuration information, identifying the power consumption jump phenomenon in the task sequence to determine whether there is a power consumption jump phenomenon in the task sequence.

[0069] Specifically, when it is determined based on the initial task configuration information that the power consumption difference between two adjacent tasks is greater than a predetermined threshold, it is determined that there is a power consumption jump phenomenon in the task sequence.

[0070] The initial task configuration information may include the task execution duration of each task and the power consumption required to run the task, but is not limited thereto, and may also include the execution order of multiple tasks and the power consumption required for each task.

[0071] The power consumption difference between two adjacent tasks can be calculated to obtain the power consumption difference. When the power consumption difference is greater than a predetermined threshold, such as the power consumption predetermined threshold, it is determined that there is a power consumption jump phenomenon in the task sequence.

[0072] Figure 4A Schematically shows a schematic diagram of a task sequence with a power consumption jump phenomenon according to an embodiment of the present disclosure.

[0073] As Figure 4A shown, the task sequence includes 6 tasks, which respectively include task A, task B, task C, task D, task E, and task F in the execution order. Based on the task attribute information, it is determined that the initial task configuration information may include the power consumption of task A being 300W, the power consumption of task B being 600W, the power consumption of task C being 200W, the power consumption of task D being 700W, the power consumption of task E being 100W, and the power consumption of task F being 400W. The task execution duration of task A on the task execution device is t1, the task execution duration of task B on the task execution device is t2, the task execution duration of task C on the task execution device is t3, the task execution duration of task D on the task execution device is t4, the task execution duration of task E on the task execution device is t5, and the task execution duration of task F on the task execution device is t6.

[0074] The power consumption difference of 400W between adjacent tasks B and C is greater than the predetermined threshold of 350W. The power consumption difference of 500W between adjacent tasks D and C is greater than the predetermined threshold of 350W. Also, the power consumption difference of 600W between adjacent tasks D and E is greater than the predetermined threshold of 350W. Then, it is determined that there is a power consumption jump phenomenon in the task sequence.

[0075] There is no limitation on the information type of the initial task configuration information. For example, it can adopt the table form shown on the left, but it is not limited to this. It can also adopt the curve graph form shown on the right, or a combination of table and curve graph. Figure 4A The form of the curve graph shown on the right, or a combination of table and curve graph. Figure 4A The form of the curve graph shown on the right, or a combination of table and curve graph.

[0076] The power consumption jump phenomenon can be directly identified by taking the difference in power consumption, but it is not limited to this. The power difference between two adjacent tasks can also be calculated to obtain the power difference value. When the power difference is greater than the predetermined power threshold, it is determined that there is a power consumption jump phenomenon in the task sequence. It is also possible to determine that there is a power consumption jump phenomenon in the task sequence when the current difference is compared with the predetermined current threshold.

[0077] Compared with the methods using power difference and current difference, the method using power consumption difference for identification is accurate and effective. In addition, compared with the methods using power and current, etc., which reflect the effective power consumption of the task execution device, using power consumption for identification can accurately reflect the entire power consumption on the task execution device, which is close to, intuitive, and effective for the data required by the actual power supply management device.

[0078] When it is determined that there is a power consumption jump phenomenon in the task sequence, the power consumption in the initial task configuration information can be updated. For example, the power consumption of at least one task can be reduced. For example, the power consumption of Figure 4A Tasks 2 and 4 shown is reduced so that the power consumption difference between any two adjacent tasks is less than the predetermined threshold. However, it is not limited to this. The execution order of multiple tasks can also be updated, so that the power consumption difference between any two adjacent tasks is less than the predetermined threshold. Compared with the method of reducing power consumption, the method of updating the execution order can improve the performance of the task execution device to the optimal level while ensuring the stability of the task execution system.

[0079] Next, an embodiment will be combined with Figure 4B to further illustrate how to update the execution order.

[0080] According to the embodiment of the present disclosure, the initial task configuration information includes the execution order and the power consumption of each of the multiple tasks.

[0081] For Figure 3The operation S320 shown updates the initial task configuration information to obtain task configuration information, which may include: updating the execution order of multiple tasks in the initial task configuration information, and obtaining the target execution order of the multiple tasks when it is determined that the power consumption difference between any two adjacent tasks is less than a predetermined threshold based on the power consumption of each of the multiple tasks. Based on the target execution order, target task configuration information is obtained.

[0082] The execution order between multiple tasks can be adjusted until the power consumption difference between any two adjacent tasks is less than the predetermined threshold, then the update operation is completed to obtain the target execution order. Based on the target execution order, target task configuration information is obtained.

[0083] Figure 4B Schematically shows a schematic diagram of a task sequence without power consumption jump phenomenon according to an embodiment of the present disclosure.

[0084] Compared with Figure 4A the schematic diagram of the task sequence shown, as Figure 4B shown in the schematic diagram of the task sequence, the task sequence after the execution order rearrangement includes 6 tasks, such as task C, task A, task B, task D, task F, and task E. As Figure 4B shown, the power consumption difference between any two adjacent tasks in the task sequence after the execution order rearrangement is less than the predetermined threshold of 350W, and the power consumption of each task has not decreased.

[0085] Compared with Figure 4A the initial task configuration information shown, as Figure 4B shown in the task configuration information, the execution order of multiple tasks changes, then the task execution period changes. The task execution period of each task can be determined based on the target execution order and the task execution duration.

[0086] Sending the task configuration information with the power consumption of the task and the task execution period to the power supply management device can enable the power supply management device to deliver electric energy matching the power consumption to the task execution device during the corresponding task execution period based on the task configuration information, so as to achieve a dynamic, flexible and timely voltage regulation effect while meeting the optimal performance requirements of the task execution device.

[0087] The power consumption jump phenomenon is likely to cause an impact on the power supply management device. Therefore, by updating the initial task configuration information, such as re - arranging the execution order, the occurrence of the power consumption jump phenomenon can be reduced or avoided. However, the change in the execution order will change the task execution period. For example, task A was executed first before the change in the execution order, but after the change in the execution order, it will be postponed. This causes some tasks not to be completed in time, and may even lead to a decline in the user experience due to the task execution delay.

[0088] Preferably, a delay threshold can be set, and based on the delay threshold, a restrictive rearrangement of the execution order of the task queue is performed.

[0089] For example, the initial task configuration information further includes the task execution duration of each of the multiple tasks. The delay threshold can be used as a time window, and the time window is slid. Based on the task execution duration of each of the multiple tasks, the power consumption difference of the task group within the time window in the task sequence is detected. If it is found that the power consumption difference between two adjacent tasks exceeds a predetermined threshold, the task group within the time window is rearranged, tasks with similar power consumption levels are arranged adjacent to each other in the queue, and at the same time, the jump of the power consumption value of consecutive tasks is controlled to be stepped, minimizing the impact of dI / dt on the power supply management device.

[0090] Optionally, updating the execution order of the multiple tasks in the initial task configuration information to obtain the target execution order may further include: determining a task group to be adjusted from the task sequence based on the task execution duration of each of the multiple tasks, the delay threshold, and multiple jump tasks where power consumption jumps occur. Updating the execution order of the multiple tasks in the task group to be adjusted, and obtaining the target execution order when it is determined that the power consumption difference between any two adjacent tasks in the task group to be adjusted is less than a predetermined threshold.

[0091] Take the task sequence as shown Figure 4A as an example. The multiple tasks where power consumption jumps occur are determined from the multiple tasks as: between task B and task C, between task C and task D, and between task D and task E. Based on the delay threshold, with task B and task C as the center, the duration is extended outward, and multiple tasks adjacent to task B and task C and with task execution duration within the delay threshold are determined as the task group to be adjusted. For example, task A, task B, and task C form a task group to be adjusted. Task D, task E, and task F form a task group to be adjusted. Update the execution order of the multiple tasks in the task group to be adjusted. When it is determined that the power consumption difference between any two adjacent tasks in the task group to be adjusted is less than a predetermined threshold, the target execution order is obtained.

[0092] Using the above method for updating can flexibly adjust the sliding mode of the time window, improve the update efficiency while reducing the task execution delay, ensure the stability of the task execution system while improving the user experience.

[0093] Preferably, the execution order of the multiple tasks can be adjusted, and when it is determined that the power consumption difference between any two adjacent tasks in the task sequence is less than a predetermined threshold and the power consumption of the multiple tasks changes in a stepped manner, the target execution order is obtained. This avoids the occurrence of power consumption jump phenomena, makes the power consumption change relatively smoothly, and reduces the repeated impact on the power supply management device caused by the repeated rise and fall of power consumption.

[0094] The above text describes how to adjust the execution order using the predicted power consumption. The following will explain how to predict the power consumption.

[0095] According to an embodiment of the present disclosure, for the operation S310 as Figure 3 shown, based on the task attribute information of the task sequence to be processed, determine the initial task configuration information for executing the task sequence on the task execution device, including at least one of the following: Process the task attribute information using an evaluation model to obtain the initial task configuration information. Process the task attribute information using an evaluation function to obtain the initial task configuration information. Determine the initial task configuration information based on the task attribute information and the mapping relationship. The mapping relationship represents the mapping relationship between the task attribute information and the task configuration information.

[0096] The task attribute information may include task types, such as image processing tasks, text processing tasks, etc., but is not limited thereto. It may also include calculation types, such as convolution calculation, normalization calculation, etc., as long as it is information related to the task for evaluating the initial task configuration information. The initial task configuration information may include information such as task execution duration, power consumption, etc.

[0097] A deep learning model can be trained using training samples to obtain an evaluation model. The evaluation model is used to process the task attribute information to obtain the initial task configuration information. The model architecture of the deep learning model is not limited. For example, it may include an encoder-decoder, but is not limited thereto. It may also include an open-source pre-trained large model. As long as it can be trained to output the initial task configuration information based on the task attribute information.

[0098] The task attribute information can be processed using an evaluation function to obtain the initial task configuration information. For example, the task attribute information can be represented as computable parameter data. Substitute the parameter data into the evaluation function to obtain the initial task configuration information including power consumption and task execution duration.

[0099] It is also possible to use the mapping relationship, such as a look-up table (LUT) including the correspondence between the task attribute information and the initial task configuration information such as task execution duration, power consumption, etc., to determine the initial task configuration information from the look-up table based on the task attribute information.

[0100] The simulation method can be used to simulate typical tasks to determine the simulation power consumption and simulation execution duration required to run the task on the task execution device. However, it is not limited thereto. It is also possible to actually run typical tasks to determine the power consumption and task execution duration required to run the task on the task execution device.

[0101] Based on the task attribute information obtained from simulation or actual occurrence and the initial task configuration information, use them as training samples to train a model to obtain an evaluation model; use them as fitting data to generate an evaluation function; use them as a mapping relationship to establish a lookup table.

[0102] Optionally, the above three methods can be used for weighted summation to obtain the initial task configuration information, which can improve the accuracy and effectiveness of the initial task configuration information. In addition, setting multiple determination methods can use the evaluation model to evaluate and obtain the initial task configuration information in the case where the evaluation function does not match the task or there is no data corresponding to the task in the mapping relationship, thereby expanding the application scope and avoiding task types that cannot be determined.

[0103] Preferably, it is possible to receive constraint information such as the maximum static current and the highest dynamic current that the power supply management device can carry sent by the power supply management device. Based on the power consumption in the initially predicted task configuration information and the constraint information, determine whether the predicted power consumption meets the actual requirements and whether it is the power consumption that the power supply management device can carry. In the case where the power consumption is greater than the power consumption that the power supply management device can carry, the power consumption in the initial task configuration information can be updated based on the constraint information, so that the power consumption in the initial task configuration information meets the actual output capacity.

[0104] According to an embodiment of the present disclosure, for the operation S330 as shown in Figure 3 , the task configuration information can be directly sent to the power supply management device, but it is not limited thereto. It is also possible to determine the path loss information required to deliver the electrical energy matching the task configuration information. Combine the path loss information and the task configuration information and send them to the power supply management device.

[0105] According to an embodiment of the present disclosure, the path loss information may include the path loss power consumption. The current can be determined based on the power consumption of the task, the task execution duration, and the rated voltage of the task execution device in the task configuration information. Based on the current and the impedance existing in the path, determine the path loss power consumption when supplying the appropriate electrical energy to the task execution device.

[0106] Optionally, the impedance of the path can be generated by components such as circuit boards and decoupling capacitors. The impedance can be determined by simulation, but it is not limited thereto. It can also be obtained based on actual operation data, as long as it can represent the impedance of the path.

[0107] Combine the path loss power consumption and the power consumption of the task in the task configuration information to determine the electrical energy that the power supply management device needs to output. This improves the combination of the electrical energy output by the power supply management device with the actual situation, improves the precise control of the power supply, and avoids the problem that the electrical energy output does not match the electrical energy required by the task execution device due to path loss.

[0108] Figure 5 A block diagram of a task execution device according to an embodiment of the present disclosure is schematically shown.

[0109] As Figure 5 shown, the task execution device 500 may include a cache 510, a multi-core processor 520, a scheduler 530, and a power consumption manager 540.

[0110] The scheduler 530 may receive task instructions and store the task attribute information of the task sequence in the cache 510 in the form of a queue.

[0111] The power consumption manager 540 reads the task attribute information of the task sequence from the cache 510 and determines the initial task configuration information based on the task attribute information. In the case where it is determined that there is a power consumption jump phenomenon in the task sequence, the initial task configuration information is updated to obtain the task configuration information. The power consumption manager 540 may transmit the task configuration information to the power management device through a bus and send the updated task sequence with the target execution order to the cache 510.

[0112] The scheduler 530 reads the task-related information from the updated task sequence in the cache 510 and distributes it to the multi-core processor 520 to execute the task using the multi-core processor 520.

[0113] Optionally, before performing the operation S310 as Figure 3 shown, determining the execution order of the task sequence may include: determining the execution order of multiple tasks based on the task reception time information of each of the multiple tasks.

[0114] Optionally, it may be that the scheduler stores the task reception time information in the task queue according to the task reception time information, and determines the execution order of the multiple tasks according to the storage order of the tasks in the task queue.

[0115] Determining the task sequence in the above manner can improve the adaptability to the actual application. The tasks received first are processed first, thereby avoiding delays and improving the efficiency and response ability of task execution.

[0116] According to an embodiment of the present disclosure, after performing the operation S330 as Figure 3 shown, the power supply method may further include: sending the updated task sequence determined based on the task configuration information to the cache so that the task execution device can perform task processing based on the updated task sequence.

[0117] Sending the updated task sequence to the cache may be sending the sequence identifier of the task sequence with the target execution order to the cache, but it is not limited thereto. It may also be sending the task attribute information of the task sequence with the target execution order to the cache. As long as it can facilitate the task execution device to execute the task corresponding to the updated task sequence.

[0118] According to an embodiment of the present disclosure, the updated task sequence is stored using a cache, which facilitates subsequent task processing by the task execution device and improves overall adaptability.

[0119] Figure 6 A flowchart of a power supply method according to another embodiment of the present disclosure is schematically shown.

[0120] As Figure 6 shown, the power supply method includes operation S610 to operation S620.

[0121] In operation S610, task configuration information from the task execution device is received.

[0122] The task configuration information is the initial task configuration information determined for executing the task sequence on the task execution device based on the task attribute information of the task sequence to be processed. The task sequence includes multiple tasks with an execution order. It is obtained by updating the initial task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information.

[0123] In operation S620, electrical energy matching the task configuration information is output to the task execution device.

[0124] As Figure 6 shown, the power supply method can be applied to a power supply management device.

[0125] The task configuration information can receive the task configuration information transmitted by the execution device and form a control logic.

[0126] For example, before it is indicated in the task configuration information that a large load current is about to arrive, the output voltage is raised in advance to offset the subsequent voltage drop, so as to ensure that the voltage on the task execution device side is controlled within the range required for the processor core to work. If it is indicated in the task configuration information that the current of the load circuit is about to become smaller, the output voltage can be lowered in advance.

[0127] According to an embodiment of the present disclosure, the task configuration information fed back by the task execution device side can be used to sense in advance the power consumption requirements of the task execution device in the next period of time and accordingly adjust and control the output voltage, realizing the intervention in the output electrical energy in advance to cope with the upcoming current transient change situation, and improving the steady state of the task execution system and the operating performance of the task execution device.

[0128] Figure 7 A block diagram of a power supply device according to an embodiment of the present disclosure is schematically shown.

[0129] As Figure 7As shown, the power supply device 700 includes: a configuration module 710, an update module 720, and a sending module 730.

[0130] The configuration module 710 is configured to determine initial task configuration information for executing a task sequence on a task execution device based on task attribute information of the task sequence to be processed, where the task sequence includes multiple tasks with an execution order.

[0131] The update module 720 is configured to update the initial task configuration information to obtain task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information; and

[0132] The sending module 730 is configured to send the task configuration information to a power supply management device so that the power supply management device outputs electric energy matching the task configuration information to the task execution device.

[0133] According to an embodiment of the present disclosure, the power supply device further includes: a jump identification module.

[0134] The jump identification module is configured to determine that there is a power consumption jump phenomenon in the task sequence when it is determined that the power consumption difference between two adjacent tasks is greater than a predetermined threshold based on the initial task configuration information.

[0135] According to an embodiment of the present disclosure, the initial task configuration information includes an execution order and the power consumption of each of the multiple tasks.

[0136] According to an embodiment of the present disclosure, the update module includes: a sequence update sub-module and an information determination sub-module.

[0137] The sequence update sub-module is configured to update the execution order of the multiple tasks in the initial task configuration information, and obtain the target execution order of the multiple tasks when it is determined that the power consumption difference between any two adjacent tasks is less than a predetermined threshold based on the power consumption of each of the multiple tasks.

[0138] The information determination sub-module is configured to obtain target task configuration information based on the target execution order.

[0139] According to an embodiment of the present disclosure, the initial task configuration information further includes the task execution duration of each of the multiple tasks.

[0140] According to an embodiment of the present disclosure, the sequence update sub-module includes: a task group identification unit and an update unit.

[0141] The task group identification unit is configured to determine a task group to be adjusted from the task sequence based on the task execution duration of each of the multiple tasks, a time delay threshold, and multiple jump tasks where power consumption jumps occur.

[0142] An update unit is configured to update the execution order of multiple tasks in a task group to be adjusted, and obtain a target execution order when it is determined that the power consumption difference between any two adjacent tasks in the task group to be adjusted is less than a predetermined threshold.

[0143] According to an embodiment of the present disclosure, the sending module includes: a loss determination sub-module and a combination sub-module.

[0144] The loss determination sub-module is configured to determine path loss information required for delivering electric energy matching the task configuration information.

[0145] The combination sub-module is configured to combine the path loss information and the task configuration information and send the combination to a power supply management device.

[0146] According to an embodiment of the present disclosure, the configuration module includes at least one of the following: a model configuration sub-module, a function configuration sub-module, and a mapping configuration sub-module.

[0147] The model configuration sub-module is configured to process task attribute information using an evaluation model to obtain initial task configuration information.

[0148] The function configuration sub-module is configured to process task attribute information using an evaluation function to obtain initial task configuration information.

[0149] The mapping configuration sub-module is configured to determine initial task configuration information based on task attribute information and a mapping relationship. The mapping relationship represents the mapping relationship between task attribute information and task configuration information.

[0150] According to an embodiment of the present disclosure, the power supply device further includes: an order determination module.

[0151] The order determination module is configured to determine the execution order of multiple tasks based on the task reception time information of each of the multiple tasks.

[0152] According to an embodiment of the present disclosure, the power supply device further includes: a cache module.

[0153] The cache module is configured to send the updated task sequence determined based on the task configuration information to a cache, so that a task execution device executes task processing based on the updated task sequence.

[0154] Figure 8 Schematically shows a block diagram of a power supply device according to another embodiment of the present disclosure.

[0155] As Figure 8 shown, the power supply device 800 includes: a receiving module 810 and a power supply module 820.

[0156] A receiving module 810 is configured to receive task configuration information from a task execution device. The task configuration information is an initial task configuration information for executing a task sequence on the task execution device determined based on task attribute information of the task sequence to be processed. The task sequence includes multiple tasks with an execution order, and is obtained by updating the initial task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information.

[0157] A power supply module 820 is configured to output electric energy matching the task configuration information to the task execution device.

[0158] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program product.

[0159] According to an embodiment of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method as described above.

[0160] According to an embodiment of the present disclosure, a non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions are used to cause a computer to execute the method as described above.

[0161] According to an embodiment of the present disclosure, a computer program product includes a computer program which, when executed by a processor, implements the method as described above.

[0162] Figure 9 FIG. shows a schematic block diagram of an exemplary electronic device 900 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0163] As Figure 9As shown, device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0164] Multiple components in the device 900 are connected to the input / output (I / O) interface 905, including: an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disc, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0165] The computing unit 901 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are 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 appropriate processor, controller, microcontroller, etc. The computing unit 901 executes the various methods and processes described above, such as the power supply method. For example, in some embodiments, the power supply method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the power supply method described above can be executed. Alternatively, in other embodiments, the computing unit 901 can be configured to execute the power supply method in any other appropriate manner (e.g., by means of firmware).

[0166] The various embodiments of the systems and techniques described above in this specification 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. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0167] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0168] In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0169] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0170] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0171] A computer system can include a client and a server. The client and the server are generally far from each other and usually interact through a communication network. The client - server relationship is created by computer programs running on the respective computers and having a client - server relationship with each other. The server can be a cloud server, or a server of a distributed system, or a server incorporating a blockchain.

[0172] It should be understood that the various forms of processes shown above can be reordered, added to, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in this disclosure can be achieved, and no limitation is imposed herein.

[0173] The above - mentioned specific implementation manners do not constitute a limitation on the protection scope of this disclosure. 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 substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A power supply method, comprising: Determining initial task configuration information for executing the task sequence on a task execution device based on task attribute information of the task sequence to be processed, where the task sequence includes a plurality of tasks with an execution order; Updating the initial task configuration information to obtain task configuration information when it is determined, based on the initial task configuration information, that there is a power consumption jump phenomenon in the task sequence; and Sending the task configuration information to a power supply management device so that the power supply management device outputs electric energy matching the task configuration information to the task execution device.

2. The method according to claim 1, further comprising: Determining that there is a power consumption jump phenomenon in the task sequence when it is determined, based on the initial task configuration information, that the power consumption difference between two adjacent tasks is greater than a predetermined threshold.

3. The method according to claim 1 or 2, wherein The initial task configuration information includes an execution order and the power consumption of each of the plurality of tasks; The updating the initial task configuration information to obtain task configuration information includes: Updating the execution order of the plurality of tasks in the initial task configuration information, and obtaining a target execution order of the plurality of tasks when it is determined, based on the power consumption of each of the plurality of tasks, that the power consumption difference between any two adjacent tasks is less than the predetermined threshold; and Obtaining the target task configuration information based on the target execution order.

4. The method according to claim 3, wherein The initial task configuration information further includes the task execution duration of each of the plurality of tasks; The updating the execution order of the plurality of tasks in the initial task configuration information, and obtaining a target execution order when it is determined, based on the power consumption of each of the plurality of tasks, that the power consumption difference between any two adjacent tasks is less than the predetermined threshold, includes: Determining a task group to be adjusted from the task sequence based on the task execution duration of each of the plurality of tasks, a time delay threshold, and a plurality of jump tasks where power consumption jumps occur; and Updating the execution order of the plurality of tasks in the task group to be adjusted, and obtaining the target execution order when it is determined that the power consumption difference between any two adjacent tasks in the task group to be adjusted is less than the predetermined threshold.

5. The method according to any one of claims 1 to 4, wherein, The sending the task configuration information to the power supply management device includes: Determining path loss information required for delivering electric energy matching the task configuration information; and Combining the path loss information and the task configuration information and sending them to the power supply management device.

6. The method according to any one of claims 1 to 5, wherein, The determining initial task configuration information for executing the task sequence on a task execution device based on task attribute information of the task sequence to be processed includes at least one of the following: Processing the task attribute information using an evaluation model to obtain the initial task configuration information; Processing the task attribute information using an evaluation function to obtain the initial task configuration information; And Determining the initial task configuration information based on the task attribute information and a mapping relationship, where the mapping relationship represents the mapping relationship between the task attribute information and the task configuration information.

7. The method according to any one of claims 1 to 6, further comprising: Determine the execution order of multiple tasks based on the task reception time information of each of the multiple tasks.

8. The method according to any one of claims 1 to 7, further comprising: Send the updated task sequence determined based on the task configuration information to the cache, so that the task execution device performs task processing based on the updated task sequence.

9. A power supply method, comprising: Receiving task configuration information from the task execution device, where the task configuration information is the initial task configuration information determined for executing the task sequence on the task execution device based on the task attribute information of the task sequence to be processed, where the task sequence includes multiple tasks with an execution order, and is obtained by updating the initial task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information; and Outputting electrical energy matching the task configuration information to the task execution device.

10. A task execution system, comprising: A task execution device, configured to determine the initial task configuration information for executing the task sequence based on the task attribute information of the task sequence to be processed, where the task sequence includes multiple tasks with an execution order; and update the initial task configuration information to obtain task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information; and A power supply management device, configured to receive the task configuration information from the task execution device and output electrical energy matching the task configuration information to the task execution device.

11. A power supply device, comprising: A configuration module, configured to determine the initial task configuration information for executing the task sequence on the task execution device based on the task attribute information of the task sequence to be processed, where the task sequence includes multiple tasks with an execution order; An update module, configured to update the initial task configuration information to obtain task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information; and A sending module, configured to send the task configuration information to the power supply management device, so that the power supply management device outputs electrical energy matching the task configuration information to the task execution device.

12. A power supply device, comprising: A receiving module, configured to receive task configuration information from the task execution device, where the task configuration information is the initial task configuration information determined for executing the task sequence on the task execution device based on the task attribute information of the task sequence to be processed, where the task sequence includes multiple tasks with an execution order, and is obtained by updating the initial task configuration information when it is determined that there is a power consumption jump phenomenon in the task sequence based on the initial task configuration information; and A power supply module, configured to output electrical energy matching the task configuration information to the task execution device.

13. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1 to 9.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are for causing the computer to execute the method according to any one of claims 1 to 9.

15. A computer program product, comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 9.

Images