Method for determining migration path, device, electronic equipment and program product

Abstract

The application discloses a kind of determination methods of migration path and its device, electronic equipment, program product, it is related to artificial intelligence technical field, wherein, the determination method includes: current edge node is collected at current measurement time point Multiple signal data, and based on all signal data, the index value of multiple indexes is calculated;In the case where there is index value not in the preset range, all index values are input to target model, and the service type handled by current edge node is obtained;Based on the index value of all indexes and service type, the migration generation value is calculated, wherein, migration generation value corresponds to preset cost range, each preset range corresponds to migration path;Determine the target preset cost range of migration generation value, determine the target migration path based on the target preset cost range.The present application solves the technical problems that the migration path cannot be accurately determined in the related art, resulting in a high service migration failure rate and affecting business continuity.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and more specifically, to a method, apparatus, electronic device, and program product for determining a migration path. Background Technology

[0002] Currently, service migration strategies in edge computing environments primarily employ fixed threshold triggering methods, which cannot dynamically adapt to network load fluctuations. For example, in virtual reality applications, when users move between different scenes, services need to quickly switch between edge nodes to maintain a smooth experience. However, due to the mismatch between dynamic changes in network conditions and fixed thresholds, service interruptions can occur, causing screen stuttering and dizziness, affecting the immersive experience. Furthermore, current methods ignore the susceptibility of millimeter-wave communication to building obstruction and multipath effects. In complex, high-speed moving scenarios, the failure to adequately predict or adapt to channel changes leads to a high service migration failure rate, resulting in frequent connection interruptions and data retransmissions, impacting business continuity.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and program product for determining migration paths, to at least solve the technical problem in related technologies where the inability to accurately determine migration paths leads to a high service migration failure rate and affects business continuity.

[0005] According to one aspect of the embodiments of this application, a method for determining a migration path is provided, comprising: collecting multiple signal data of a current edge node at a current measurement time point, and calculating the index values ​​of multiple indicators based on all signal data; if there are index values ​​that are not within a preset range, inputting all index values ​​into a target model to obtain the service type processed by the current edge node, wherein the target model is a time series model obtained through pre-training; calculating a migration cost value based on the service type and the index values ​​of all indicators, wherein the migration cost value corresponds to a preset cost range, and each preset cost range corresponds to a migration path; determining a target preset cost range for the migration cost value, and determining a target migration path based on the target preset cost range.

[0006] Furthermore, the signal data includes at least: signal-to-noise ratio (SNR), current delay, and current power consumption. The indicators include at least: channel quality, delay deviation, and energy efficiency. The steps for calculating the indicator values ​​of multiple indicators based on all signal data include: performing a numerical conversion on the SNR to obtain a converted value, and performing a level mapping on the converted value to obtain a level value, which is used to characterize the channel quality; determining the target delay, calculating the deviation value based on the current delay and the target delay, and calculating the current delay deviation value based on the deviation value and the previous delay deviation value, wherein the previous delay deviation value is calculated based on the previous delay and the target delay, and the measurement time point of the previous delay is earlier than the measurement time point of the current delay; obtaining the power consumption of the current edge node in the idle state, and calculating the energy efficiency value based on the power consumption and the current power consumption.

[0007] Furthermore, the step of calculating the energy efficiency value based on the power consumption and the current power consumption includes: calculating the energy efficiency value based on the preset power consumption threshold and the current power consumption when the power consumption is less than the preset power consumption threshold; or, calculating the energy efficiency value based on the power consumption and the current power consumption when the power consumption is greater than or equal to the preset power consumption threshold.

[0008] Furthermore, the steps for calculating the migration cost based on the business type and the values ​​of all indicators include: constructing multiple preset function factors based on all indicators, and performing a weighted summation of all preset function factors to construct a migration cost function; and calculating the migration cost based on the values ​​of all indicators and the migration cost function.

[0009] Furthermore, after determining the target migration path based on the target preset cost range, the process also includes: generating multiple checkpoint data for the current edge node based on preset time intervals, wherein the checkpoint data includes at least: a data file and a checksum; determining the checkpoint data closest to the data at the current time point, and determining the difference between the data and the data file in the closest checkpoint data to obtain incremental data; compressing the incremental data to obtain a compressed package, and sending the compressed package to the target node in the target migration path, wherein the target node receives the compressed package and decompresses it; performing double verification on the decompressed data to obtain a verification result, and collecting multiple signal data of the target node at multiple preset time points when the verification result indicates that the verification is successful; and calculating the index values ​​of multiple indicators based on all signal data at each preset time point.

[0010] Furthermore, the indicator values ​​include at least: the level value at a preset time point, the current delay deviation value at a preset time point, and the energy efficiency value at a preset time point. After calculating the indicator values ​​of multiple indicators, the following steps are also included: obtaining the historical successful migration count of the target node; and using a Kalman filter to determine the level value at the next preset time point based on the level value at the current preset time point. The state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point, and the control matrix in the Kalman filter is determined based on the historical successful migration count. The control input quantity in the Kalman filter is used to characterize the historical successful migration count, and the current preset time point is earlier than the next preset time point.

[0011] Furthermore, after calculating the numerical values ​​of multiple indicators, the process also includes: constructing multiple evaluation functions, wherein the evaluation functions include at least: a latency evaluation function, an energy consumption evaluation function, and a channel quality evaluation function; for each preset time point, calculating multiple evaluation values ​​based on all indicator values ​​and all evaluation functions; calculating a comprehensive evaluation value based on all evaluation values ​​at all preset time points, and generating an evaluation report based on the comprehensive evaluation value and all evaluation values.

[0012] According to another aspect of the embodiments of this application, a migration path determination device is also provided, comprising: a first calculation unit, configured to collect multiple signal data of the current edge node at the current measurement time point, and calculate the index values ​​of multiple indicators based on all signal data; a first input unit, configured to input all index values ​​into a target model to obtain the service type processed by the current edge node when there are index values ​​outside a preset range, wherein the target model is a time series model obtained through pre-training; a second calculation unit, configured to calculate the migration cost value based on the service type and the index values ​​of all indicators, wherein the migration cost value corresponds to a preset cost range, and each preset cost range corresponds to a migration path; and a first determination unit, configured to determine the target preset cost range of the migration cost value, and determine the target migration path based on the target preset cost range.

[0013] Furthermore, the signal data includes at least: signal-to-noise ratio (SNR), current delay, and current power consumption. The indicators include at least: channel quality, delay deviation, and energy efficiency. The first calculation unit includes: a first conversion module, used to perform numerical conversion on the SNR to obtain the converted value, and to perform level mapping on the converted value to obtain a level value, which is used to characterize the channel quality; a first calculation module, used to determine the target delay, calculate the deviation value based on the current delay and the target delay, and calculate the current delay deviation value based on the deviation value and the previous delay deviation value, wherein the previous delay deviation value is calculated based on the previous delay and the target delay, and the measurement time point of the previous delay is earlier than the measurement time point of the current delay; and a second calculation module, used to obtain the power consumption of the current edge node in the idle state, and to calculate the energy efficiency value based on the power consumption and the current power consumption.

[0014] Furthermore, the second calculation module includes: a first calculation submodule, used to calculate an energy efficiency value based on the preset power consumption threshold and the current power consumption when the power consumption is less than the preset power consumption threshold; and a second calculation submodule, used to calculate an energy efficiency value based on the power consumption and the current power consumption when the power consumption is greater than or equal to the preset power consumption threshold.

[0015] Furthermore, the second calculation unit includes: a first construction module, used to construct multiple preset function factors based on all indicators, and to perform a weighted summation of all preset function factors to construct a migration cost function; and a third calculation module, used to calculate the migration cost based on the values ​​of all indicators and the migration cost function.

[0016] Furthermore, the migration path determination device further includes: a first generation module, used to generate multiple checkpoint data of the current edge node based on a preset time interval after determining the target migration path based on a preset cost range, wherein the checkpoint data includes at least: a data file and a checksum; a first determination module, used to determine the checkpoint data closest to the data at the current time point, and determine the difference between the data and the data file in the closest checkpoint data to obtain incremental data; a first compression module, used to compress the incremental data to obtain a compressed package, and send the compressed package to the target node in the target migration path, wherein the target node receives the compressed package and decompresses the compressed package; a first verification module, used to perform double verification on the decompressed data to obtain a verification result, and, if the verification result indicates that the verification is successful, collect multiple signal data of the target node at multiple preset time points; and a fourth calculation module, used to calculate the index values ​​of multiple indicators based on all signal data for each preset time point.

[0017] Furthermore, the indicator values ​​include at least: the level value at a preset time point, the current delay deviation value at the preset time point, and the energy efficiency value at the preset time point. The migration path determination device further includes: a first acquisition module, used to acquire the historical successful migration count of the target node after calculating the indicator values ​​of multiple indicators; and a second determination module, used to determine the level value at the next preset time point based on the level value at the current preset time point using a Kalman filter. The state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point, the control matrix in the Kalman filter is determined based on the historical successful migration count, and the control input quantity in the Kalman filter is used to characterize the historical successful migration count. The current preset time point is earlier than the next preset time point.

[0018] Furthermore, the migration path determination device also includes: a second construction module, used to construct multiple evaluation functions after calculating the index values ​​of multiple indicators, wherein the evaluation functions include at least: a latency evaluation function, an energy consumption evaluation function, and a channel quality evaluation function; a fifth calculation module, used to calculate multiple evaluation values ​​for each preset time point based on all index values ​​and all evaluation functions; and a sixth calculation module, used to calculate a comprehensive evaluation value based on all evaluation values ​​at all preset time points, and generate an evaluation report based on the comprehensive evaluation value and all evaluation values.

[0019] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method for determining any of the above-mentioned migration paths.

[0020] According to another aspect of the embodiments of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors cause the one or more processors to implement any of the above-described migration path determination methods.

[0021] In this invention, multiple signal data of the current edge node at the current measurement time point are collected, and the indicator values ​​of multiple indicators are calculated based on all signal data. If the indicator values ​​are not within the preset range, all indicator values ​​are input into the target model to obtain the service type processed by the current edge node. Based on the service type and the indicator values ​​of all indicators, the migration cost is calculated, and the target preset cost range of the migration cost is determined. Based on the target preset cost range, the target migration path is determined. This solves the technical problem in related technologies where the migration path cannot be accurately determined, resulting in a high service migration failure rate and affecting business continuity.

[0022] In this invention, lightweight probes are used to collect multi-dimensional network signal data of edge nodes in real time. Based on all signal data, comprehensive index values ​​reflecting network status and service requirements can be calculated. If an index value deviates from the preset normal range, an intelligent decision-making mechanism can be immediately activated. All calculated index values ​​are used as inputs and fed into a time series model to accurately predict the service type being processed by the current edge node. Then, based on the predicted service type and all index values, the migration cost is further calculated. The migration cost is compared with multiple preset cost ranges to obtain the target preset cost range to which the migration cost belongs. The preset cost range corresponds one-to-one with the migration path. Based on the target preset cost range, the target migration path is determined, achieving seamless and rapid migration between edge nodes, reducing service migration failure rate, lowering unnecessary migration overhead, and improving business continuity and service quality. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0024] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for determining migration paths is shown.

[0025] Figure 2 This is a flowchart of the method for determining the migration path according to Embodiment 1 of this application;

[0026] Figure 3 This is a schematic diagram of an optional migration path determination device according to an embodiment of this application;

[0027] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] It should be noted that all related information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, and displayed data) collected and involved in this invention are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data comply with the relevant laws, regulations, and standards of the relevant regions, necessary confidentiality measures have been taken, and it does not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse. For example, this system has an interface with relevant users or organizations. Before obtaining relevant information, a request to obtain the information needs to be sent to the aforementioned user or organization through the interface, and the relevant information is obtained only after receiving consent from the aforementioned user or organization.

[0031] In this invention, the current node migration strategy lacks accurate prediction of future network conditions and intelligent optimization of resource allocation in complex scenarios, which may lead to an increased service migration failure rate, affecting business continuity and user experience. Furthermore, the hardware resource scheduling algorithm is too conservative and fails to dynamically adjust according to real-time load, resulting in energy waste and increased equipment heat generation, impacting system stability. This invention, however, can adjust resource allocation strategies in advance based on the business type predicted by the model. For example, when a high-load scenario such as autonomous driving or industrial control is predicted, GPU resources can be preheated; conversely, idle resources can be released in a timely manner to avoid unnecessary energy consumption.

[0032] The present invention will now be described in detail with reference to various embodiments.

[0033] Example 1

[0034] According to an embodiment of this application, an embodiment of a method for determining a migration path is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0035] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal (or mobile device) for implementing a method for determining migration paths is shown. Figure 1 As shown, computer terminal 10 (or mobile device) may include one or more ( Figure 1 The processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions may also be included. In addition, it may include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera, wherein the network interface can be connected to wired and / or wireless networks. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0036] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0037] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the migration path determination method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned migration path determination method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0038] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0039] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0040] Under the aforementioned operating environment, this application provides the following: Figure 2 The method for determining the migration path is shown. Figure 2 This is a flowchart of the migration path determination method according to Embodiment 1 of this application, as follows: Figure 2 As shown, the method includes the following steps:

[0041] Step S201: Collect multiple signal data of the current edge node at the current measurement time point, and calculate the index values ​​of multiple indicators based on all signal data.

[0042] In this embodiment of the invention, an edge node is an entity in an edge computing architecture responsible for performing computing tasks and storing data. Located near the data source, it reduces network latency and bandwidth pressure. Edge nodes can collect wireless signals from terminal devices via radio frequency hardware. Lightweight probes can sample the signal data of the edge node at a preset frequency (e.g., 1 kHz) to obtain the signal data of the current edge node at the current measurement time. The sampled signal data can be stored in a circular buffer, enabling fast data access and minimizing latency. For example, if the sampling frequency is 1 kHz (i.e., 1000 samples per second), the circular buffer can store the signal data from the most recent 200 ms (i.e., 200 samples). Based on the signal data stored in the circular buffer, multiple metrics (such as channel quality metrics, latency deviation, migration power ratio, etc.) can be calculated.

[0043] Step S202: If there are indicator values ​​that are not within the preset range, input all indicator values ​​into the target model to obtain the business type processed by the current edge node. The target model is a time series model obtained through pre-training.

[0044] In this embodiment of the invention, the preset range is a threshold range of pre-set normal operation indicators. Exceeding this range may indicate poor network status or changes in business requirements. If there are indicator values ​​that are not within the preset range, all indicator values ​​can be input into the target model. The target model is a time series model obtained through pre-training, which can identify and predict the types of business processed by edge nodes, such as autonomous driving, industrial control, or ordinary data services.

[0045] Step S203: Calculate the migration cost based on the business type and the value of all indicators. The migration cost corresponds to a preset cost range, and each preset cost range corresponds to a migration path.

[0046] Optionally, after determining the business type, a random number generator can be used to generate three sets of perturbation parameters based on a preset perturbation range and random noise. Each set of perturbation parameters corresponds to a different migration path, and each migration path corresponds to a preset cost range. The preset cost range can be (0, 1], (1, 2], and (2, 3], which correspond to migration path 1, migration path 2, and migration path 3, respectively.

[0047] In this embodiment of the invention, the migration cost is calculated using a pre-constructed cost migration function based on the values ​​of all indicators. The migration cost can be used to assess migration costs; the smaller the migration cost, the lower the migration cost.

[0048] Step S204: Determine the target preset cost range for migration cost, and determine the target migration path based on the target preset cost range.

[0049] In this embodiment of the invention, the migration cost is compared with each preset cost range to determine the target preset cost range to which the migration cost belongs, and a target migration path can be determined based on the target preset cost range. For example, if the migration cost is in the range (0, 1], migration path 1 can be selected; if the migration cost is in the range (1, 2], migration path 2 can be selected; and if the migration cost is in the range (2, 3], migration path 3 can be selected.

[0050] In summary, by collecting multi-dimensional signal data from edge nodes, including channel quality, latency deviation, and energy efficiency ratio, and then calculating indicator values, real-time data support is provided for dynamic decision-making. When indicator values ​​deviate from preset ranges, all indicator values ​​are input into a pre-trained time series model to accurately identify the specific service type being processed by the current node, providing contextual awareness for subsequent decisions. Based on the identified service type and current indicator values, the migration cost is calculated, and the target preset cost range to which the migration cost belongs is determined. This determines the target migration path, achieving precise control over edge node migration. This reduces service interruption time, improves business continuity, and solves the technical problem in related technologies where the inability to accurately determine migration paths leads to high service migration failure rates and impacts business continuity.

[0051] The signal data includes at least: signal-to-noise ratio (SNR), current delay, and current power consumption. The indicators include at least: channel quality, delay deviation, and energy efficiency. To accurately calculate the energy efficiency value, in the migration path determination method provided in Embodiment 1 of this application, the SNR is numerically converted to obtain a converted value, and the converted value is mapped to a level to obtain a level value, which is used to characterize the channel quality. The target delay is determined, and the deviation value is calculated based on the current delay and the target delay. The current delay deviation value is calculated based on the deviation value and the previous delay deviation value, wherein the previous delay deviation value is calculated based on the previous delay and the target delay, and the measurement time point of the previous delay is earlier than the measurement time point of the current delay. The power consumption of the current edge node in the idle state is obtained, and the energy efficiency value is calculated based on the power consumption and the current power consumption.

[0052] In this embodiment of the invention, the signal data includes at least the signal-to-noise ratio (SINR), the current delay, and the current power consumption. The linear SINR can be converted into a dB value, that is, the signal-to-noise ratio is numerically converted to obtain the converted value (e.g., sinr_db = 10 × log10 (SINR)). The converted value is then mapped to a level, and the mapping rule can be: fmin(15, fmax(0, (sinr_db + 6) / 3)). The fmin() function indicates that the level will not exceed 15, and the fmax() function indicates that the level will not be less than 0. The obtained level value can be used to characterize the channel quality.

[0053] Determine the target latency (which can be set according to specific circumstances, such as 5ms). Based on the current latency and the target latency, calculate the deviation value (i.e., delta = current latency - target latency). Based on the deviation value and the previous latency deviation value (i.e., last_delta), calculate the current latency deviation value (i.e., delta_d = 0.8 × last_delta + 0.2 × delta). The previous latency deviation value is calculated based on the previous latency and the target latency, and the measurement time of the previous latency is earlier than the measurement time of the current latency. For example, the initial value of the previous latency deviation value can be set to 0, i.e., last_delta = 0, then the current latency deviation value (i.e., delta_d = 0.2 × delta) can be used as the previous latency deviation value for the next calculation (i.e., last_delta = delta_d) to calculate the next latency deviation value.

[0054] The power consumption of the current edge node in the idle state (i.e., when no service is running) is obtained, and the power consumption is compared with the preset power consumption threshold. The comparison result indicates that different comparison results indicate different power consumption values. Based on the power consumption value and the current power consumption, the energy efficiency value is calculated.

[0055] By calculating the values ​​of multiple indicators and determining whether they are within a preset range, the system can intelligently determine when to perform a migration operation, avoiding unnecessary waste of resources and achieving efficient resource utilization and a smooth task transition.

[0056] In order to accurately calculate the energy efficiency value, in the migration path determination method provided in Embodiment 1 of this application, when the power consumption is less than a preset power consumption threshold, the energy efficiency value is calculated based on the preset power consumption threshold and the current power consumption; or, when the power consumption is greater than or equal to the preset power consumption threshold, the energy efficiency value is calculated based on the power consumption and the current power consumption.

[0057] In this embodiment of the invention, a preset power consumption threshold (e.g., 0.1) needs to be set first to prevent the power consumption from being too low or zero. The fmax() function can be used to compare the power consumption with the preset power consumption threshold, i.e., fmax(power_idle, 0.1). When the power consumption is less than the preset power consumption threshold, the power consumption in the idle state is 0.1 (i.e., the preset power consumption threshold). The energy efficiency value can be calculated based on the preset power consumption threshold and the current power consumption (i.e., energy efficiency value = current power consumption / 0.1). Alternatively, when the power consumption is greater than or equal to the preset power consumption threshold, the power consumption is power_idle. The energy efficiency value can be calculated based on the power consumption and the current power consumption (i.e., energy efficiency value = current power consumption / power_idle).

[0058] In order to accurately calculate the migration cost, in the migration path determination method provided in Embodiment 1 of this application, multiple preset function factors are constructed based on all indicators, and all preset function factors are weighted and summed to construct a migration cost function; the migration cost is calculated based on the values ​​of all indicators and the migration cost function.

[0059] In this embodiment of the invention, the preset function factors may include, but are not limited to: delay factor, energy consumption factor, and reliability factor (used to represent channel quality). A weighted sum is performed on all preset function factors to construct the migration cost function (i.e., ,in, Indicates the delay factor. Indicates the energy consumption factor. Here, T represents the reliability factor, E represents the current delay deviation parameter, and E represents the energy efficiency parameter. (where BER represents the maximum allowable energy consumption parameter and BER represents the level parameter). By substituting all the indicator values ​​and the maximum allowable energy consumption into the migration cost function, the migration cost can be calculated.

[0060] It should be noted that the weights of all preset function factors, such as 0.6, 0.3, and 0.1, can be dynamically adjusted.

[0061] To improve the integrity and reliability of migration data transmission, the migration path determination method provided in Embodiment 1 of this application generates multiple checkpoint data for the current edge node based on a preset time interval. The checkpoint data includes at least a data file and a checksum. The method then determines the checkpoint data closest to the current time point and identifies the differences between the data file and the data file in the closest checkpoint data to obtain incremental data. The incremental data is compressed to obtain a compressed package, which is then sent to the target node in the target migration path. The target node receives and decompresses the compressed package. The decompressed data undergoes dual verification to obtain a verification result. If the verification result indicates successful verification, multiple signal data from the target node are collected at multiple preset time points. For each preset time point, multiple indicator values ​​are calculated based on all signal data.

[0062] In this embodiment of the invention, edge nodes periodically (i.e., based on a preset time interval, for example, every 50ms) generate checkpoint data. This checkpoint data includes at least a data file and a checksum, such as CRC32. When edge node A needs to migrate to target node B, it can find the checkpoint closest to the current time point from the checkpoint data, such as CPn, and then calculate the difference between the current node state and CPn to obtain incremental data. An efficient compression algorithm (such as Zstd (a compression and decompression algorithm)) can be used to compress the incremental data, generating a compressed package, which is then sent to the target node in the target migration path. The target node receives the compressed package and decompresses it using the same compression algorithm to recover the incremental data. Through incremental data transmission and compression technology, the amount of data transmitted is reduced, the migration time is shortened, and the network load is reduced. Afterwards, the decompressed data undergoes dual verification (e.g., checksum verification and hash value verification) to obtain the verification result. This dual verification mechanism improves the integrity and security of the data migration process, avoiding the risk of data corruption or tampering. If the verification result indicates that the verification is successful, multiple signal data of the target node are collected at multiple preset time points (i.e., multiple time points when the target node is used for service), including but not limited to signal interference signal-to-noise ratio, current latency and current power consumption. For each preset time point, based on all signal data, the index values ​​of multiple indicators are calculated, such as channel quality, latency deviation and energy efficiency.

[0063] Optionally, if the target node does not respond for more than 500ms, the transmission path can be switched (i.e., a path can be randomly selected from the remaining two paths for data migration). Alternatively, if the target node fails to verify three times in a row, the transmission path can be switched, and then the same verification steps as the target node can be performed.

[0064] The indicator values ​​include at least: the level value at a preset time point, the current delay deviation value at a preset time point, and the energy efficiency value at a preset time point. In order to accurately predict the channel quality at the next time point, in the migration path determination method provided in Embodiment 1 of this application, the historical successful migration count of the target node is obtained; based on the level value at the current preset time point, a Kalman filter is used to determine the level value at the next preset time point. The state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point, the control matrix in the Kalman filter is determined based on the historical successful migration count, and the control input quantity in the Kalman filter is used to characterize the historical successful migration count. The current preset time point is earlier than the next preset time point.

[0065] In this embodiment of the invention, the grade value is used to represent channel quality; the higher the grade value, the better the channel quality. The index values ​​at each preset time point include at least the grade value, the current delay deviation value, and the energy efficiency value. Based on the grade value at the current preset time point, a Kalman filter is used to determine the grade value at the next preset time point. The state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point, and the control matrix in the Kalman filter is determined based on the number of historical successful migrations. The control input quantity in the Kalman filter is used to characterize the number of historical successful migrations. The current preset time point is earlier than the next preset time point. By dynamically predicting the channel quality grade value and combining historical migration experience with current environmental conditions, the trend of channel quality changes can be predicted in advance, allowing for the selection of the optimal timing during migration, thereby better assisting in planning the migration decisions of edge nodes.

[0066] To accurately generate an evaluation report, the migration path determination method provided in Embodiment 1 of this application constructs multiple evaluation functions, which include at least: a latency evaluation function, an energy consumption evaluation function, and a channel quality evaluation function; for each preset time point, multiple evaluation values ​​are calculated based on all indicator values ​​and all evaluation functions; a comprehensive evaluation value is calculated based on all evaluation values ​​at all preset time points, and an evaluation report is generated based on the comprehensive evaluation value and all evaluation values.

[0067] In this embodiment of the invention, multiple evaluation functions are constructed, including but not limited to: a latency cost evaluation function, an energy consumption cost evaluation function, and a reliability cost evaluation function.

[0068] Optionally, for the latency cost evaluation function, if the current latency deviation value (x) is less than or equal to a preset threshold (e.g., 5), the cost is the current latency deviation value itself (i.e., cost = x). If it exceeds the preset threshold, the cost is a punitive increase based on the latency threshold (i.e., latency cost = x + 10 × (x - preset threshold)).

[0069] Alternatively, for the energy consumption evaluation function, as the energy efficiency value (y) increases, the rate of increase of the function value gradually slows down. By using the logarithmic function, i.e., energy cost = log(1+y), where log is the natural logarithm function with base e, it is possible to avoid the high energy consumption value having too much impact on the total cost.

[0070] Alternatively, the reliability cost evaluation function can be calculated based on the Sigmoid function (an activation function) formula, where reliability cost = The target value can be set to 0.99. The Sigmoid function has an S-shaped curve, with a gentler slope when it approaches 0 and 1, and a steeper slope in the middle section.

[0071] For each preset time point, multiple evaluation values ​​are calculated based on all indicator values ​​and all evaluation functions. Then, the average of all evaluation values ​​is calculated to obtain a comprehensive evaluation value (i.e., total cost). A smaller value indicates a better migration path. An evaluation report can be generated based on the comprehensive evaluation value and all evaluation values ​​to help decision-makers understand the performance trend of the target node. For example, {"Total Cost": 32.17, "Cost Composition": {"Latency Cost": 4.2, "Energy Consumption Cost": 5.30, "Reliability Cost": 22.67}}.

[0072] The migration path determination method provided in this application, through the construction of a dynamic migration decision mechanism and optimization algorithm, realizes an efficient edge node migration solution for latency-sensitive services. Specifically, by using lightweight probe design and Zstd incremental data compression transmission, the bandwidth requirements for data migration are significantly reduced. Simultaneously, the integrity and security of the data migration process are ensured through a dual verification mechanism using checksums and hash values. Employing an improved Kalman filter prediction algorithm, changes in the network environment can be predicted in advance. Through a multi-objective decision model, the migration cost is accurately assessed, resulting in more economical and low-latency migration decisions. This not only optimizes resource allocation and improves the energy efficiency and service quality of edge computing but also enhances adaptability to complex environments and intelligent management. Through an intelligent dynamic migration algorithm for edge nodes, integrating deep learning and efficient data transmission technologies, microsecond-level accurate decision-making and lossless migration are achieved, effectively reducing service interruption time and improving business continuity and resource utilization efficiency.

[0073] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0074] Example 2

[0075] This application also provides a migration path determination apparatus. It should be noted that the migration path determination apparatus of this application can be used to execute the migration path determination method provided in this application. The migration path determination apparatus provided in this application will be described below.

[0076] According to an embodiment of this application, an apparatus for implementing the above-described method for determining migration paths is also provided. Figure 3 This is a schematic diagram of an optional migration path determination device according to an embodiment of this application, such as... Figure 3 As shown, the migration path determination device may include: a first calculation unit 30, a first input unit 31, a second calculation unit 32, and a first determination unit 33.

[0077] The first calculation unit 30 is used to collect multiple signal data of the current edge node at the current measurement time point, and calculate the index values ​​of multiple indicators based on all signal data.

[0078] The first input unit 31 is used to input all indicator values ​​into the target model when there are indicator values ​​that are not within the preset range, so as to obtain the business type processed by the current edge node. The target model is a time series model obtained through pre-training.

[0079] The second calculation unit 32 is used to calculate the migration cost based on the business type and the indicator values ​​of all indicators. The migration cost corresponds to a preset cost range, and each preset cost range corresponds to a migration path.

[0080] The first determining unit 33 is used to determine the target preset cost range of migration cost and, based on the target preset cost range, determine the target migration path.

[0081] The migration path determination device provided in this application embodiment can collect multiple signal data of the current edge node at the current measurement time point through the first calculation unit 30, and calculate the index values ​​of multiple indicators based on all signal data. If there are index values ​​outside the preset range, the first input unit 31 can input all index values ​​into the target model to obtain the service type processed by the current edge node. The second calculation unit 32 can calculate the migration cost based on the service type and the index values ​​of all indicators. The first determination unit 33 can determine the target preset cost range of the migration cost, and determine the target migration path based on the target preset cost range.

[0082] Optionally, the signal data includes at least: signal-to-noise ratio (SNR), current delay, and current power consumption, and the indicators include at least: channel quality, delay deviation, and energy efficiency. The first calculation unit 30 includes: a first conversion module, used to perform numerical conversion on the signal-to-noise ratio (SNR) to obtain the converted value, and to perform level mapping on the converted value to obtain a level value, which is used to characterize the channel quality; a first calculation module, used to determine the target delay, calculate the deviation value based on the current delay and the target delay, and calculate the current delay deviation value based on the deviation value and the previous delay deviation value, wherein the previous delay deviation value is calculated based on the previous delay and the target delay, and the measurement time point of the previous delay is earlier than the measurement time point of the current delay; and a second calculation module, used to obtain the power consumption of the current edge node in the idle state, and calculate the energy efficiency value based on the power consumption and the current power consumption.

[0083] Optionally, the second calculation module includes: a first calculation submodule, used to calculate an energy efficiency value based on the preset power consumption threshold and the current power consumption when the power consumption is less than the preset power consumption threshold; and a second calculation submodule, used to calculate an energy efficiency value based on the power consumption and the current power consumption when the power consumption is greater than or equal to the preset power consumption threshold.

[0084] Optionally, the second calculation unit 32 includes: a first construction module, used to construct multiple preset function factors based on all indicators, and to perform a weighted summation of all preset function factors to construct a migration cost function; and a third calculation module, used to calculate the migration cost based on the values ​​of all indicators and the migration cost function.

[0085] Optionally, the migration path determination device further includes: a first generation module, used to generate multiple checkpoint data of the current edge node based on a preset time interval after determining the target migration path based on a preset cost range, wherein the checkpoint data includes at least: a data file and a checksum; a first determination module, used to determine the checkpoint data closest to the data at the current time point, and determine the difference between the data and the data file in the closest checkpoint data to obtain incremental data; a first compression module, used to compress the incremental data to obtain a compressed package, and send the compressed package to the target node in the target migration path, wherein the target node receives the compressed package and decompresses the compressed package; a first verification module, used to perform double verification on the decompressed data to obtain a verification result, and, if the verification result indicates that the verification is successful, collect multiple signal data of the target node at multiple preset time points; and a fourth calculation module, used to calculate the index values ​​of multiple indicators based on all signal data for each preset time point.

[0086] Optionally, the indicator values ​​include at least: the level value at a preset time point, the current delay deviation value at the preset time point, and the energy efficiency value at the preset time point. The migration path determination device further includes: a first acquisition module, used to acquire the historical successful migration count of the target node after calculating the indicator values ​​of multiple indicators; and a second determination module, used to determine the level value at the next preset time point based on the level value at the current preset time point using a Kalman filter. The state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point, the control matrix in the Kalman filter is determined based on the historical successful migration count, and the control input quantity in the Kalman filter is used to characterize the historical successful migration count. The current preset time point is earlier than the next preset time point.

[0087] Optionally, the migration path determination device further includes: a second construction module, used to construct multiple evaluation functions after calculating the index values ​​of multiple indicators, wherein the evaluation functions include at least: a latency evaluation function, an energy consumption evaluation function, and a channel quality evaluation function; a fifth calculation module, used to calculate multiple evaluation values ​​for each preset time point based on all index values ​​and all evaluation functions; and a sixth calculation module, used to calculate a comprehensive evaluation value based on all evaluation values ​​at all preset time points, and generate an evaluation report based on the comprehensive evaluation value and all evaluation values.

[0088] The aforementioned migration path determination device may further include a processor and a memory. The first calculation unit 30, the first input unit 31, the second calculation unit 32, the first determination unit 33, etc., are all stored in the memory as program units, and the processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0089] The aforementioned processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and by adjusting kernel parameters, a target preset cost range for migration costs is determined. Based on this target preset cost range, the target migration path is determined.

[0090] The aforementioned memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0091] It should be noted that the first calculation unit 30, the first input unit 31, the second calculation unit 32, and the first determination unit 33 mentioned above correspond to steps S201 to S204 in Embodiment 1. The instances and application scenarios implemented by the above units and the corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above units can also be part of a device and run in the computer terminal 10 provided in Embodiment 1.

[0092] Example 3

[0093] Embodiments of this application may provide a computer terminal, which may be any computer terminal device in a group of computer terminals. Optionally, in this embodiment, the aforementioned computer terminal may also be replaced with a mobile terminal or an electronic device, etc.

[0094] Optionally, in this embodiment, the computer terminal may be located in at least one of a plurality of network devices in a computer network.

[0095] In this embodiment, the computer terminal described above can execute the program code for the following steps in the method for determining the migration path: collecting multiple signal data of the current edge node at the current measurement time point, and calculating the index values ​​of multiple indicators based on all signal data; if there are index values ​​that are not within a preset range, inputting all index values ​​into the target model to obtain the service type processed by the current edge node, wherein the target model is a time series model obtained through pre-training; calculating the migration cost value based on the service type and the index values ​​of all indicators, wherein the migration cost value corresponds to a preset cost range, and each preset cost range corresponds to a migration path; determining the target preset cost range of the migration cost value, and determining the target migration path based on the target preset cost range.

[0096] Optionally, the aforementioned computer terminal may execute the following steps in the migration path determination method: perform numerical conversion on the signal-to-noise ratio (SNR) to obtain the converted value, and perform level mapping on the converted value to obtain a level value, which is used to characterize the channel quality; determine the target delay, calculate the deviation value based on the current delay and the target delay, and calculate the current delay deviation value based on the deviation value and the previous delay deviation value, wherein the previous delay deviation value is calculated based on the previous delay and the target delay, and the measurement time point of the previous delay is earlier than the measurement time point of the current delay; obtain the power consumption of the current edge node in the idle state, and calculate the energy efficiency value based on the power consumption and the current power consumption.

[0097] Optionally, the computer terminal may execute the program code for the following steps in the method for determining the migration path: if the power consumption is less than a preset power consumption threshold, calculate the energy efficiency value based on the preset power consumption threshold and the current power consumption; or, if the power consumption is greater than or equal to the preset power consumption threshold, calculate the energy efficiency value based on the power consumption and the current power consumption.

[0098] Optionally, the computer terminal described above can execute the program code for the following steps in the method for determining the migration path: constructing multiple preset function factors based on all indicators, and performing a weighted summation of all preset function factors to construct a migration cost function; calculating the migration cost value based on the values ​​of all indicators and the migration cost function.

[0099] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the migration path determination method: generating multiple checkpoint data for the current edge node based on a preset time interval, wherein the checkpoint data includes at least: a data file and a checksum; determining the checkpoint data closest to the data at the current time point, and determining the difference between the data and the data file in the closest checkpoint data to obtain incremental data; compressing the incremental data to obtain a compressed package, and sending the compressed package to the target node in the target migration path, wherein the target node receives the compressed package and decompresses it; performing double verification on the decompressed data to obtain a verification result, and if the verification result indicates that the verification is successful, collecting multiple signal data of the target node at multiple preset time points; for each preset time point, calculating the index values ​​of multiple indicators based on all signal data.

[0100] Optionally, the computer terminal described above can execute the program code for the following steps in the method for determining the migration path: obtaining the historical number of successful migrations of the target node; based on the level value at the current preset time point, using a Kalman filter to determine the level value at the next preset time point, wherein the state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point, the control matrix in the Kalman filter is determined based on the historical number of successful migrations, the control input quantity in the Kalman filter is used to characterize the historical number of successful migrations, and the current preset time point is earlier than the next preset time point.

[0101] Optionally, the aforementioned computer terminal may execute the program code for the following steps in the method for determining the migration path: constructing multiple evaluation functions, wherein the evaluation functions include at least: a latency evaluation function, an energy consumption evaluation function, and a channel quality evaluation function; for each preset time point, calculating multiple evaluation values ​​based on all indicator values ​​and all evaluation functions; calculating a comprehensive evaluation value based on all evaluation values ​​at all preset time points, and generating an evaluation report based on the comprehensive evaluation value and all evaluation values.

[0102] Optionally, Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown) Processor 402, memory 404, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0103] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the migration path determination method and apparatus in this application embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned migration path determination method. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0104] The processor can invoke the information and application programs stored in the memory through the transmission device to execute the steps described above in the method for determining the migration path.

[0105] The embodiments of this application provide a scheme for determining migration paths. By calculating indicator values ​​in real time and dynamically generating multiple candidate migration paths, and comparing the migration cost of the indicator values ​​with a preset cost range, the migration path can be accurately determined, realizing intelligent and efficient path migration decision-making. Through a three-step method (i.e., checkpoint generation, incremental data compression, and verification), data migration is executed efficiently and accurately, ensuring business continuity. This solves the technical problem in related technologies where the migration path cannot be accurately determined, resulting in a high service migration failure rate and affecting business continuity.

[0106] Those skilled in the art will understand that Figure 4 The structure shown is for illustrative purposes only. Electronic devices can also be terminal devices such as smartphones, tablets, PDAs, and mobile internet devices (MIDs). Figure 4 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 4 The different configurations shown.

[0107] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0108] Example 4

[0109] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the migration path determination method provided in Embodiment 1.

[0110] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0111] This application also provides a computer program product, which, when executed on a data processing device, is adapted to perform the steps of a method for determining a migration path.

[0112] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0113] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0114] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0115] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0118] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining a migration path, characterized in that, include: Multiple signal data of the current edge node at the current measurement time point are collected, and the index values ​​of multiple indicators are calculated based on all the signal data. In this process, multiple signal data are obtained by sampling the signal data of the current edge node at the current measurement time point at a preset frequency. The signal data within a preset time range from the current measurement time point are stored in a circular buffer. The index values ​​are calculated based on all the signal data in the circular buffer. If the value of the indicator is outside the preset range, all the value of the indicator are input into the target model to obtain the business type processed by the current edge node. The target model is a time series model obtained through pre-training. Based on the business type and the indicator values ​​of all the indicators, a migration cost is calculated, wherein the migration cost corresponds to a preset cost range, and each preset cost range corresponds to a migration path; wherein, according to the business type, a random number generator is used to generate multiple sets of perturbation parameters based on a preset perturbation range and random noise, and each set of perturbation parameters corresponds to a different migration path; based on all the indicators, multiple preset function factors are constructed, and all the preset function factors are weighted and summed to construct a migration cost function; based on all the indicator values ​​and the migration cost function, the migration cost is calculated; Determine the target preset cost range for the migration cost, and determine the target migration path based on the target preset cost range.

2. The method for determining a migration path according to claim 1, characterized in that, The signal data includes at least: signal-to-noise ratio (SNR), current delay, and current power consumption. The indicators include at least: channel quality, delay deviation, and energy efficiency. The step of calculating the values ​​of multiple indicators based on all the signal data includes: The signal-to-noise ratio (SNR) of the signal interference is converted to obtain a converted value, and the converted value is mapped to a level to obtain a level value, which is used to characterize the channel quality. Determine the target delay, calculate the deviation value based on the current delay and the target delay, calculate the current delay deviation value based on the deviation value and the previous delay deviation value, wherein the previous delay deviation value is calculated based on the previous delay and the target delay, and the measurement time point of the previous delay is earlier than the measurement time point of the current delay; Obtain the power consumption of the current edge node in the idle state, and calculate the energy efficiency value based on the power consumption and the current power consumption.

3. The method for determining a migration path according to claim 2, characterized in that, The step of calculating the energy efficiency value based on the power consumption and the current power consumption includes: If the power consumption is less than a preset power consumption threshold, calculate the energy efficiency value based on the preset power consumption threshold and the current power consumption; or, If the power consumption is greater than or equal to the preset power consumption threshold, the energy efficiency value is calculated based on the power consumption and the current power consumption.

4. The method for determining a migration path according to claim 1, characterized in that, After determining the target migration path based on the preset cost range, the process also includes: Based on a preset time interval, multiple checkpoint data of the current edge node are generated, wherein the checkpoint data includes at least: a data file and a checksum; Determine the checkpoint data closest to the current time point, and determine the difference between the data file and the data file in the closest checkpoint data to obtain incremental data; The incremental data is compressed to obtain a compressed package, and the compressed package is sent to the target node in the target migration path, wherein the target node receives the compressed package and decompresses it; The decompressed data is double-verified to obtain a verification result. If the verification result indicates that the verification is successful, multiple signal data of the target node are collected at multiple preset time points. For each preset time point, the index values ​​of multiple indicators are calculated based on all the signal data.

5. The method for determining a migration path according to claim 4, characterized in that, The indicator values ​​include at least: a level value at a preset time point, a current delay deviation value at a preset time point, and an energy efficiency value at a preset time point. After calculating the indicator values ​​of multiple indicators, the method further includes: Obtain the historical number of successful migrations of the target node; Based on the level value at the current preset time point, a Kalman filter is used to determine the level value at the next preset time point. The state transition matrix in the Kalman filter is determined based on the energy efficiency value and the current delay deviation value at the current preset time point. The control matrix in the Kalman filter is determined based on the number of historical successful migrations. The control input quantity in the Kalman filter is used to characterize the number of historical successful migrations. The current preset time point is earlier than the next preset time point.

6. The method for determining a migration path according to claim 4, characterized in that, After calculating the value of each of the multiple metrics, the method further includes: Construct multiple evaluation functions, wherein the evaluation functions include at least: a delay evaluation function, an energy consumption evaluation function, and a channel quality evaluation function; For each of the preset time points, multiple evaluation values ​​are calculated based on all the indicator values ​​and all the evaluation functions; Based on all the evaluation values ​​at all the preset time points, a comprehensive evaluation value is calculated, and an evaluation report is generated based on the comprehensive evaluation value and all the evaluation values.

7. A device for determining a migration path, characterized in that, include: The first calculation unit is used to collect multiple signal data of the current edge node at the current measurement time point, and calculate the index values ​​of multiple indicators based on all the signal data. The unit obtains multiple signal data by sampling the signal data of the current edge node at the current measurement time point at a preset frequency, stores the signal data within a preset time range from the current measurement time point in a circular buffer, and calculates the index values ​​based on all the signal data in the circular buffer. The first input unit is used to input all the indicator values ​​into the target model when there are indicator values ​​that are not within the preset range, so as to obtain the business type processed by the current edge node, wherein the target model is a time series model obtained by pre-training; The second calculation unit is used to calculate the migration cost value based on the business type and the indicator values ​​of all the indicators, wherein the migration cost value corresponds to a preset cost range, and each preset cost range corresponds to a migration path; wherein, based on the business type, a random number generator is used to generate multiple sets of perturbation parameters based on a preset perturbation range and random noise, and each set of perturbation parameters corresponds to a different migration path; based on all the indicators, multiple preset function factors are constructed, and all the preset function factors are weighted and summed to construct a migration cost function; the migration cost value is calculated based on all the indicator values ​​and the migration cost function; The first determining unit is used to determine the target preset cost range of the migration cost, and to determine the target migration path based on the target preset cost range.

8. A computer program product, characterized in that, The method includes a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for determining a migration path as described in any one of claims 1 to 6.

9. An electronic device, characterized in that, It includes one or more processors and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the migration path determination method according to any one of claims 1 to 6.

Images