An adaptive checking method and system for an intermittent computing system

Abstract

The application belongs to the field of Internet of Things, and provides a self-adaptive checking method and system of an intermittent computing system, and the technical scheme is as follows: an intermittent computing system loads and runs a program; a switchable checkpoint is inserted in the running program, at the first checkpoint, according to the residual condition of energy and the scale of the WAR variable, the performance of different checking methods is compared, the checking method at the first checkpoint is determined, if a periodic checking method is adopted, all the WAR variables are backed up at each checkpoint, and at each checkpoint, the next checking method is determined according to the comparison result; if a JIT checking method is adopted, all the subsequent checkpoints are temporarily disabled, and the system voltage is collected by using an adaptive collection frequency, so that the checking mode is reselected at each checkpoint, and the overhead of the detection method in the system is reduced.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things (IoT), and particularly relates to an adaptive inspection method and system for the detection mode of its own backup nodes in an intermittent computing system in an IoT terminal. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] The rapid development of Internet of Things (IoT) devices has permeated our lives. These devices require batteries for normal operation. Using energy harvesting systems allows devices to obtain energy from their surroundings, eliminating battery dependence and enabling greater mobility and longer operating cycles. However, energy harvesting systems are unstable. If a power outage occurs, the system will cease operation, data will be lost, and the system may restart from a very early state after power is restored, leading to a non-progress problem. Therefore, intermittent computing systems have been proposed to address this issue. These systems use a checkpointing mechanism to save data to non-volatile memory (NVM) before a power outage. Once power is restored, the system retrieves the previously saved data and resumes operation from the point of failure.

[0004] Existing inspection methods can be broadly categorized into two types: periodic checkpointing methods and just-in-time (JIT) inspection methods.

[0005] Current inspection methods employ either a single periodic inspection approach or a just-in-time inspection approach. However, using only periodic inspection methods results in many redundant periodic checkpoints when energy supply is sufficient. Performing inspections and backups at these checkpoints incurs significant system overhead. Furthermore, when an energy outage occurs between two checkpoints, saved data variables may change, potentially leading to errors during recovery and re-reading. Figure 1 As shown. These variables that may cause errors are called write-attribute-read variables (WAR variables). Periodic checking methods need to be specially designed to avoid system errors caused by WAR variables.

[0006] Using only the Just-in-time checking method results in high energy costs for running the ADC and generating the standard voltage for reference, sometimes even exceeding the energy consumption of the system CPU itself. Summary of the Invention

[0007] To address at least one of the technical problems in the background art described above, the present invention provides an adaptive inspection method and system for intermittent computing systems, which combines the advantages of periodic detection methods and Just-in-time detection methods, and adaptively selects the optimal inspection point method and voltage sampling frequency from the periodic detection method and the Just-in-time detection method.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] The first aspect of the present invention provides an adaptive checking method for an intermittent computing system, comprising the following steps:

[0010] The program is loaded and run via an intermittent computing system;

[0011] Insert switchable checkpoints in the running program. At the first checkpoint, compare the performance of different check methods based on the remaining energy and the size of the WAR variable to determine the check method at the first checkpoint. If a periodic check method is used, back up all WAR variables at each checkpoint and continue to determine the next check method based on the comparison results at each checkpoint.

[0012] If the JIT (Just-In-Time) inspection method is used, all subsequent inspection points are temporarily disabled, and the system voltage is sampled using an adaptive sampling frequency, thereby reselecting the inspection method at each inspection point.

[0013] Furthermore, the comparison of the performance of different checking methods based on the remaining energy and the size of the WAR variable includes:

[0014] The system determines the energy required for the current backup by calculating the size of the current WAR variable: if the size of the WAR variable is greater than the set size threshold, the system is more inclined to choose the JIT check method.

[0015] The remaining energy in the current energy buffer is used to determine how long the system can sustain itself after an energy failure. If the energy in the energy buffer is greater than the set energy threshold, it means that there is a high probability that the system is in an external environment with sufficient energy, and the system is more likely to choose the JIT check method.

[0016] Furthermore, if a periodic check method is adopted, checkpoints are set periodically during system operation to save the current data to non-volatile storage;

[0017] If the JIT checking method is adopted, an analog-to-digital converter is added to the intermittent computing system to monitor the voltage of the intermittent computing system's energy buffer in real time. When the monitored voltage reaches the system's preset minimum threshold, the intermittent computing system automatically stops running and saves the current system state to non-volatile memory. When the monitored voltage reaches the preset maximum threshold, the intermittent computing system resumes running from the breakpoint.

[0018] Furthermore, under the JIT inspection method, the frequency of voltage detection is adaptively adjusted according to the rise and fall of voltage. When the voltage drops, the frequency of voltage detection is adaptively increased, and when the voltage rises, the frequency of voltage detection is adaptively decreased.

[0019] Furthermore, the adaptive increase or decrease of the voltage detection frequency includes:

[0020] When a certain voltage is detected, the relationship between the time interval for the next voltage sampling and the time it takes for the voltage value to drop to the lowest voltage value is set. The time interval is calculated using the RC discharge function of the MCU's maximum operating current.

[0021] Furthermore, the time interval can also be calculated by establishing a lookup table, based on the characteristics of the circuit where the MCU is located, to determine the time required for the circuit voltage to drop from an arbitrary value to a dangerous voltage, and then establishing a mapping relationship in the form of an array and storing it in non-volatile memory.

[0022] Furthermore, when switching from the JIT inspection method to the periodic inspection method, the JIT inspection mode needs to be maintained until the next inspection point is reached.

[0023] A second aspect of the present invention provides an adaptive checking system for an intermittent computing system, comprising:

[0024] The program loading module is used to load and run programs through the intermittent computing system.

[0025] The adaptive checking module is used to insert switchable checkpoints in the running program. At the first checkpoint, the performance of different checking methods is compared based on the remaining energy and the size of the WAR variable to determine the checking method at the first checkpoint. If the periodic checking method is adopted, all WAR variables are backed up at each checkpoint, and the next checking method is determined based on the comparison results at each checkpoint.

[0026] If the JIT (Just-In-Time) inspection method is used, all subsequent inspection points are temporarily disabled, and the system voltage is sampled using an adaptive sampling frequency, thereby reselecting the inspection method at each inspection point.

[0027] A third aspect of the present invention provides a computer-readable storage medium.

[0028] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the adaptive checking method for an intermittent computing system as described above.

[0029] A fourth aspect of the present invention provides a computer device.

[0030] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the adaptive checking method for an intermittent computing system as described above.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] This invention inserts switchable checkpoints into the running program. At the first checkpoint, the performance of different check methods is compared based on the remaining energy and the scale of the WAR variable to determine the check method for that first checkpoint. If a periodic check method is used, all WAR variables are backed up at each checkpoint, and the next check method is determined based on the comparison results at each checkpoint. If a JIT check method is used, all subsequent checkpoints are temporarily disabled, and the system voltage is sampled using an adaptive sampling frequency, thereby reselecting the check method at each checkpoint. This invention combines the advantages of both periodic and Just-in-time detection methods, adaptively selecting the optimal checkpoint method and voltage sampling frequency from the two approaches.

[0033] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0034] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0035] Figure 1 This is a schematic diagram illustrating the principle of WAR variables in an intermittent computing system;

[0036] Figure 2This is a flowchart of an adaptive checking method for an intermittent computing system provided in an embodiment of the present invention;

[0037] Figure 3 This is the simulated voltage for performance testing provided in the embodiments of the present invention;

[0038] Figure 4 This describes the performance of the adaptive inspection method of the present invention under different test procedures, as provided in the embodiments of the present invention. Detailed Implementation

[0039] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0040] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0041] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0042] The energy overhead of voltage detection in the system is linearly positively correlated with the frequency of voltage detection. The energy overhead of system backup is also linearly positively correlated with the size of the backed-up data. Regarding different data storage methods during backup: compared to CPU access, DMA (Direct Memory Access) incurs significantly less overhead during data backup.

[0043] In summary, the detection method for an intermittent system is affected by various factors, including backup methods, backup data size, detection methods, and surrounding environmental factors. Reducing the overhead of the detection method in the system is a key issue in the design of an intermittent computing system.

[0044] This invention proposes an adaptive checking method for intermittent computing systems, which combines the advantages of periodic detection methods and Just-in-time detection methods, and adaptively selects the optimal check point method and voltage sampling frequency from the periodic detection method and Just-in-time detection method.

[0045] The periodic check method works by periodically setting checkpoints during system operation and saving the current data to the NVM. This method performs fixed checks on the system and does not consider whether the current energy supply is sufficient. Its advantage lies in that when the check interval is set sufficiently small, it ensures that backups are completed before potential energy supply interruptions occur, thus guaranteeing the system's forward capability and correctness.

[0046] The just-in-time (JIT) check method works by adding an analog-to-digital converter (ADC) to the system to monitor the voltage of the system's energy buffer in real time. When the monitored voltage reaches a preset low threshold, the system automatically stops running and saves the current system state to the NVM. When the monitored voltage reaches a preset high threshold, the system resumes operation from the breakpoint. The advantage of this model is that it eliminates unnecessary periodic checks.

[0047] Example 1

[0048] This embodiment provides an adaptive checking method for an intermittent computing system, including the following steps:

[0049] Step 1: Load and run the program;

[0050] In this embodiment, the running program is loaded through an intermittent computing system, which includes several pre-statically inserted checkpoints (CPs).

[0051] Step 2: The system will adaptively select the inspection method based on the current state.

[0052] The program to be executed by the system will insert switchable checkpoints. At the first checkpoint, the system will compare the performance of different check methods based on the remaining energy and the size of the WAR variable, and select the better check method.

[0053] The comparison of the performance of different checking methods based on the remaining energy and the size of the WAR variable specifically includes:

[0054] The system calculates the size of the current WAR variable to determine the energy required for the current backup. If the size of the WAR variable is greater than the set size threshold, then periodic backups may result in significant redundancy, and the system will be more inclined to choose the JIT check method.

[0055] The remaining energy in the current energy buffer (such as capacitors) is used to determine how long the system can sustain itself after an energy failure: if the energy in the energy buffer is greater than the set energy threshold, it means that there is a high probability that the system is in an external environment with sufficient energy. In this case, periodic backups may bring a lot of redundancy, so the system will be more inclined to choose the JIT check method.

[0056] If the periodic check method is selected, the energy consumption for checking and backup under this method is the product of the average backup data size and the number of backups. Correspondingly, the energy consumption for checking and backup under the JIT check method depends on the backup size and the overhead of ADC detection at the set frequency. The system will comprehensively judge and select a check method at the initial check point by estimating the energy consumption of both, and will reselect or maintain the corresponding method at each subsequent check point.

[0057] If the system selects the periodic check method, the system will back up all WAR variables at each checkpoint and continue to determine the next check method based on performance at each checkpoint.

[0058] If the system selects the JIT checkpoint method, it will temporarily disable all subsequent checkpoints and use an adaptive sampling frequency to collect the system voltage. At each checkpoint, the system will reselect the check method. In this check method, the adaptive check system in this embodiment will adjust the voltage detection frequency. When the voltage drops, the voltage detection frequency will be adaptively increased, and when the voltage rises, the voltage detection frequency will be adaptively decreased.

[0059] For example, it can be set that when a certain voltage is detected, the time interval for the next voltage sampling is half the time it takes for that voltage value to "drop" to the lowest voltage value. This time interval can be calculated in advance using the RC discharge function of the MCU's maximum operating current. Furthermore, by creating a lookup table, based on the characteristics of the circuit containing the MCU, the time required for the circuit voltage to "drop" from an arbitrary value to a dangerous voltage can be calculated. The mapping relationship can then be established in the form of an array and stored in non-volatile memory for efficient and fast lookup.

[0060] When the system uses the JIT (Just-In-Time) checking method, the voltage sampling time is adaptively and dynamically changed, while the checkpoints in the program are statically inserted. Therefore, the checkpoints may not exactly match the voltage sampling points.

[0061] If the system switches directly from a JIT (Just-In-Time) check method to a periodic check method before reaching the next checkpoint, and an energy interruption occurs before reaching the checkpoint, there is no safe checkpoint to allow the system to resume operations once energy supply is restored. Therefore, when switching from a JIT check method to a periodic check method, it is necessary to maintain the JIT check mode until the next checkpoint is reached to ensure system correctness.

[0062] like Figure 2 As shown, in order to more clearly illustrate the technical solution of the present invention, the following description uses a specific switching process as an example.

[0063] Step 1: Load and run the program;

[0064] Step 2: When encountering the first checkpoint CP0, compare the performance of the two current check methods, and select the periodic check method PC after comparison;

[0065] Step 3: When the system reaches checkpoint CP2, if the method is adaptively selected as JIT check method, the system check method is switched from periodic check method to JIT check method, and the subsequent checkpoint CP3 is disabled;

[0066] Step 4: After switching to JIT inspection mode, the voltage sampling frequency is adaptively adjusted according to the current voltage sampling (energy remaining). In the process, from sampling point S1 to sampling point S3, the sampled voltage value continuously decreases, and the corresponding voltage sampling frequency continuously increases.

[0067] Step 5: At collection point S3, the system switches the inspection method to periodic inspection. However, the switch from JIT inspection method to periodic inspection method needs to maintain the JIT method until the next inspection point, i.e., CP4. At the same time, CP4 is also undisabled. After CP4, the system switches to the periodic inspection method.

[0068] In this embodiment, the running time of the PC method, the JIT method, and the adaptive method of the present invention were measured on a specific MCU under different running programs as benchmark programs, and the running time required by the present invention was set as the standard time.

[0069] To simulate unstable voltages in real-world applications, this invention simulates three different states of variable voltage, such as... Figure 3 As shown.

[0070] The optimal checking method varies depending on the program's execution conditions. For example, for the Cuckoo algorithm, the average data required to be saved at checkpoints reaches 383.1 bytes, the highest among the tested algorithms. Clearly, the JIT checking method performs better for the Cuckoo algorithm. Conversely, for the Cyclic Redundancy Check (CRC) algorithm, the backup data size at checkpoints is relatively small, so the PC method performs better. However, the overall performance of this invention is superior to using either checking method individually. Figure 4 As shown, the average speed of the adaptive checkpointing system is increased by 1.7 times compared to the static checkpointing method.

[0071] Meanwhile, the additional runtime overhead of the inspection method is reduced due to the establishment of a cheat sheet and a linear relationship, thus minimizing the overhead from calculating the update collection frequency and comparing performance. Experimental results show that the runtime overhead of adaptive checkpoint management is less than 5% of the total execution time.

[0072] Periodic checkpointing and Just-In-Time (JIT) checkpointing are the two most widely used checkpointing strategies in intermittent computing systems, but how to select the optimal checkpointing method remains a gap. This invention conducts an empirical study on the performance of the two systems. Experimental results show that software factors, hardware factors, and environmental factors affect the performance comparison of the two systems. Based on the experimental results, a runtime system that adaptively changes the checkpointing strategy is developed to achieve better performance under different running applications and environments.

[0073] Example 2

[0074] This embodiment provides an adaptive checking system for an intermittent computing system, including:

[0075] The program loading module is used to load and run programs through the intermittent computing system.

[0076] The adaptive checking module is used to insert switchable checkpoints in the running program. At the first checkpoint, the performance of different checking methods is compared based on the remaining energy and the size of the WAR variable to determine the checking method at the first checkpoint. If the periodic checking method is adopted, all WAR variables are backed up at each checkpoint, and the next checking method is determined based on the comparison results at each checkpoint.

[0077] If the JIT (Just-In-Time) inspection method is used, all subsequent inspection points are temporarily disabled, and the system voltage is sampled using an adaptive sampling frequency, thereby reselecting the inspection method at each inspection point.

[0078] Example 3

[0079] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the adaptive checking method for an intermittent computing system as described above.

[0080] Example 4

[0081] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the adaptive checking method for an intermittent computing system as described above.

[0082] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0083] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0087] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive checking method for an intermittent computing system, characterized in that, Includes the following steps: The program is loaded and run via an intermittent computing system; Insert switchable checkpoints in the running program. At the first checkpoint, compare the performance of different check methods based on the remaining energy and the size of the WAR variable to determine the check method at the first checkpoint. If a periodic check method is used, back up all WAR variables at each checkpoint and continue to determine the next check method based on the comparison results at each checkpoint. The comparison of the performance of different checking methods based on the remaining energy and the size of the WAR variable includes: The system determines the energy required for the current backup by calculating the size of the current WAR variable: if the size of the WAR variable is greater than the set size threshold, the system is more inclined to choose the JIT check method. The remaining energy in the current energy buffer is used to determine how long the system can sustain itself after an energy failure. If the energy in the energy buffer is greater than the set energy threshold, it means that the system is in an energy-rich external environment, and the system is more likely to choose the JIT check method. If the JIT (Just-In-Time) inspection method is used, all subsequent inspection points are temporarily disabled, and the system voltage is sampled using an adaptive sampling frequency, thereby reselecting the inspection method at each inspection point.

2. The adaptive checking method for an intermittent computing system as described in claim 1, characterized in that, If a periodic check method is adopted, checkpoints are set periodically during system operation, and the current data is saved to non-volatile storage; If the JIT checking method is adopted, an analog-to-digital converter is added to the intermittent computing system to monitor the voltage of the intermittent computing system's energy buffer in real time. When the monitored voltage reaches the system's preset minimum threshold, the intermittent computing system automatically stops running and saves the current system state to non-volatile memory. When the monitored voltage reaches the preset maximum threshold, the intermittent computing system resumes running from the breakpoint.

3. The adaptive checking method for an intermittent computing system as described in claim 1, characterized in that, Under the JIT inspection method, the voltage detection frequency is adaptively adjusted according to the rise and fall of the voltage. When the voltage drops, the voltage detection frequency is adaptively increased, and when the voltage rises, the voltage detection frequency is adaptively decreased.

4. The adaptive checking method for an intermittent computing system as described in claim 3, characterized in that, The adaptive increase or decrease of the voltage detection frequency includes: When a certain voltage is detected, the relationship between the time interval for the next voltage sampling and the time it takes for the voltage value to drop to the lowest voltage value is set. The time interval is calculated using the RC discharge function of the MCU's maximum operating current.

5. The adaptive checking method for an intermittent computing system as described in claim 4, characterized in that, The time interval is also calculated by establishing a lookup table based on the characteristics of the circuit where the MCU is located. The time required for the circuit voltage to drop from an arbitrary value to a dangerous voltage is calculated. The mapping relationship is established in the form of an array and stored in non-volatile memory.

6. The adaptive checking method for an intermittent computing system as described in claim 1, characterized in that, When switching from the JIT inspection method to the periodic inspection method, the JIT inspection mode needs to be maintained until the next inspection point is reached.

7. An adaptive checking system for an intermittent computing system employing the adaptive checking method for an intermittent computing system as described in any one of claims 1-6, characterized in that, include: The program loading module is used to load and run programs through the intermittent computing system. The adaptive checking module is used to insert switchable checkpoints in the running program. At the first checkpoint, the performance of different checking methods is compared based on the remaining energy and the size of the WAR variable to determine the checking method at the first checkpoint. If the periodic checking method is adopted, all WAR variables are backed up at each checkpoint, and the next checking method is determined based on the comparison results at each checkpoint. If the JIT (Just-In-Time) inspection method is used, all subsequent inspection points are temporarily disabled, and the system voltage is sampled using an adaptive sampling frequency, thereby reselecting the inspection method at each inspection point.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the adaptive checking method for an intermittent computing system as described in any one of claims 1-6.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the adaptive checking method for an intermittent computing system as described in any one of claims 1-6.

Images