Display device and storage unit risk prediction method

Abstract

The application provides a display device and a storage unit risk prediction method. The controller monitors the erase and read access operations of the storage unit, collects two types of operation parameters, i.e., the erase operation time length and the number of read error correction bits, constructs an operation feature vector, inputs a local machine learning model, combines the time length threshold and the error correction bit threshold to calculate the risk probability of the storage unit in a future preset period, and performs data migration on the storage unit exceeding the preset risk threshold. The embodiment of the application can identify high-risk storage units that tend to be bad blocks in advance, avoid important data from being written into bad blocks to cause program damage and data loss problems through active data migration, and effectively improve the safety and stability of data storage of the display device.

Description

Technical Field

[0001] This application relates to the field of display device technology, and in particular to a risk prediction method for display devices and storage units. Background Technology

[0002] Display devices are intelligent devices capable of presenting a user interface and supporting user interaction. Taking smart TVs as an example, smart TVs are television products based on Internet application technology, possessing an open operating system and chip, an open application platform, and enabling two-way human-computer interaction. They integrate multiple functions such as audio-visual entertainment and data to meet diverse and personalized user needs. Display devices can present different media asset recommendation interfaces, which may include multiple media asset options for users to select and play.

[0003] In display devices, eMMC (Embedded Multi Media Card) is widely used as the primary storage chip for storing operating systems, applications, and user data. eMMC consists of multiple storage cells. With continued use of the display device, the storage cells in the eMMC will wear down due to repeated programming and erasing, leading to the appearance of bad blocks. A bad block refers to a storage cell that cannot reliably store data due to physical damage or degradation of its electronic properties.

[0004] However, display devices cannot identify in advance which storage cells in eMMC are trending towards becoming bad blocks, lacking early warning capabilities. This can lead to display devices writing important data to bad blocks or high-risk storage cells that are about to become bad blocks, resulting in program corruption or data loss. Summary of the Invention

[0005] This application provides a method for predicting risks in display devices and storage units, in order to solve the problem that existing display devices cannot predict the risk of bad blocks in eMMC storage units in advance, resulting in important data being written to bad blocks, causing program corruption or data loss.

[0006] In a first aspect, some embodiments of this application provide a display device, including: a display; a memory including a plurality of storage units for storing various types of data; and a controller configured to: Monitor access operations to the storage unit, wherein the access operations include erase operations and read operations; In response to the access operation to the storage unit, the operating parameters of the storage unit are collected, including the operation duration of the erase operation and the number of error correction bits of the error correction code generated when the read operation is performed; Convert the operating parameters into operating feature vectors; The running feature vector is input into a locally deployed machine learning model so that the machine learning model performs risk probability prediction on the running feature vector based on preset conditions to obtain the risk probability of the storage unit in the future preset period. The preset conditions include whether the operation duration is greater than or equal to a duration threshold and whether the number of error correction bits is greater than or equal to an error correction bit number threshold. If the risk probability of the storage unit is greater than a preset risk threshold, data migration is performed on the storage unit.

[0007] The above technical solutions have the following beneficial effects or advantages: This application monitors the erase and read access operations of storage units through a controller, collects two types of operational parameters—erasing operation duration and the number of error correction bits—and constructs an operational feature vector. This vector is then input into a local machine learning model, combined with duration and error correction bit thresholds, to calculate the risk probability of the storage unit over a preset period. Data migration is then performed on storage units exceeding the preset risk threshold. This application quantifies storage unit wear based on the erase duration and error correction bit count parameters, and outputs the risk probability through a local machine learning model. This proactively identifies storage units trending towards bad blocks and actively transfers data, preventing important data from being written to high-risk storage units and causing program corruption and data loss, thus improving data storage security.

[0008] In some embodiments of this application, the controller performs monitoring of access operations to the storage unit, specifically configured as follows: The erase and read operations on the storage unit are monitored in real time by a background resident process. Alternatively, the erase and read operations on the storage unit can be monitored periodically according to a preset time period.

[0009] The above technical solutions have the following beneficial effects or advantages: This embodiment employs a background resident process for real-time monitoring or a preset periodic monitoring of storage unit erase and read operations. Real-time monitoring by the resident process can capture sudden high-frequency erase and read operations without omission, and promptly collect instantaneous deterioration operating parameters to avoid the storage unit's rapid wear and tear going undetected in a short period of time. Periodic monitoring can reduce the controller's background computing power usage and adapt to low-load operation scenarios of display devices.

[0010] In some embodiments of this application, the controller is configured to convert the operating parameters into operating feature vectors, specifically as follows: The operation duration within a preset time window is statistically analyzed to obtain a first statistical parameter corresponding to the operation duration. The first statistical parameter includes the average value of the operation duration and the standard deviation of the operation duration. The number of error correction bits within a preset time window is statistically analyzed to obtain a second statistical parameter corresponding to the number of error correction bits. The second statistical parameter includes the average number of error correction bits and the maximum number of error correction bits. According to the preset time window, the timing change rate of the operation duration and the number of error correction bits is calculated to obtain the change rate characteristics; The first statistical parameter, the second statistical parameter, and the rate of change feature are combined to form the running feature vector.

[0011] The above technical solutions have the following beneficial effects or advantages: This embodiment uses two types of statistical parameters—mean and standard deviation of statistical operation duration and mean and maximum number of error correction bits—based on a preset time window. It combines these with time-series change rate characteristics to generate an operational feature vector. This avoids the large deviation between the predicted risk probability and the actual risk probability caused by using only single-run parameters to construct the feature vector, thus improving the accuracy of the predicted risk probability.

[0012] In some embodiments of this application, the memory is further configured to store a risk database, which is configured to store the risk probability corresponding to the storage unit; the controller performs data migration on the storage unit when the risk probability of the storage unit is greater than a preset risk threshold, specifically configured as follows: The risk probability corresponding to the storage unit is stored in the risk database; Based on the preset risk threshold, the risk probability stored in the risk database is periodically verified, thereby performing data migration on storage units with risk probabilities greater than the preset risk threshold.

[0013] The above technical solutions have the following beneficial effects or advantages: This embodiment records the risk probability of each storage unit in a risk database stored in the memory. The controller periodically verifies the probability values ​​in the risk database and performs data migration for storage units that exceed the threshold. This solves the problem of no long-term retention after a single judgment and the inability to continuously track the degradation process of storage units, thereby improving the durability of risk monitoring of storage units.

[0014] In some embodiments of this application, the controller performing data migration on the storage unit is further configured to: Obtain the load value of the display device; If the load value is less than or equal to a preset load threshold, the data migration is performed on the storage unit; If the load value is greater than a preset load threshold, the data migration will be postponed until the load value is less than or equal to the preset load threshold.

[0015] The above technical solutions have the following beneficial effects or advantages: This embodiment obtains the load value of the display device through the controller and performs data migration only when the load value is less than or equal to the load threshold. In high-load scenarios, the migration task is postponed, thereby performing the migration task in a concentrated manner during the idle and low-load period of the display device. The data transfer of the storage unit is completed without affecting the user's normal use of the display device, reducing the load pressure on the display device to perform data migration.

[0016] In some embodiments of this application, the controller performs data migration on the storage unit, specifically configured as follows: The risk database is used to query storage units whose risk probability is less than or equal to the risk threshold as low-risk storage units. Read valid data stored in storage units where the risk probability is greater than a preset risk threshold; The read valid data is written to the low-risk storage unit.

[0017] The above technical solutions have the following beneficial effects or advantages: The system filters low-risk storage units with a risk probability less than or equal to the risk threshold from the risk database, reads the valid data from the storage unit and writes it to the low-risk storage unit, thus avoiding writing the valid data of the storage unit to another storage unit with a high risk probability, thereby improving the effectiveness of data migration.

[0018] In some embodiments of this application, the controller performs data migration on the storage unit when the risk probability of the storage unit is greater than a preset risk threshold, specifically configured as follows: During the data migration process, a migration persistence flag is written into the memory. The migration persistence flag includes the physical address of the storage unit of the valid data to be migrated, the physical address of the low-risk storage unit, and migration status information. If the data migration is interrupted and the display device is powered on again, the migration persistence flag is detected; If the migration persistence flag is detected, the incomplete data migration is resumed based on the physical address of the storage unit, the physical address of the low-risk storage unit, and the migration status information.

[0019] The above technical solutions have the following beneficial effects or advantages: In this embodiment, a persistent migration flag containing the physical address and migration status is written during the migration process. After the device restarts, the interrupted migration process can be resumed based on the flag, which solves the problems of data migration interruption, data incompleteness, or logical address mapping disorder caused by power outages or shutdowns during the data migration. The persistent migration flag is permanently stored in the memory and is not affected by device power outages or restarts. After power is restored, the controller automatically identifies the incomplete migration task, improving the effectiveness of the data migration process.

[0020] In some embodiments of this application, after the controller executes the action of writing the read valid data into the low-risk storage unit, it is further configured to: Update the file system metadata or the mapping table maintained by the controller so that the logical address of the valid data is mapped to the physical address of the low-risk storage unit.

[0021] The above technical solutions have the following beneficial effects or advantages: In this embodiment, after writing valid data to the low-risk storage unit, the file system metadata or controller mapping table is updated to associate the data logical address with the new physical address, thereby improving the timeliness of updating the data logical address and the new physical address after data migration.

[0022] In some embodiments of this application, the controller performs updates to the file system metadata or the mapping table maintained by the controller, specifically configured as follows: Call the file system interface to update the mapping relationship between the logical address of the valid data recorded in the file system metadata and the physical address of the low-risk storage unit; Alternatively, update the mapping relationship between the logical address of the valid data and the physical address of the low-risk storage unit recorded in the mapping table maintained by the controller.

[0023] The above technical solutions have the following beneficial effects or advantages: By calling the file system interface to update the metadata mapping, or by directly updating the mapping table maintained by the controller itself, the storage management architecture of different display devices can be adapted. Both methods can complete the mapping update from logical address to new physical address, ensuring that data access can correctly jump to the low-risk storage unit that has been migrated, and avoiding address access errors.

[0024] Secondly, some embodiments of this application provide a method for predicting storage cell risks, including: Monitor access operations to the storage unit, wherein the access operations include erase operations and read operations; In response to the access operation to the storage unit, the operating parameters of the storage unit are collected, including the operation duration of the erase operation and the number of error correction bits of the error correction code generated when the read operation is performed; Convert the operating parameters into operating feature vectors; The running feature vector is input into a locally deployed machine learning model so that the machine learning model performs a risk probability threshold on the running feature vector based on preset conditions to obtain the risk probability of the storage unit in a future preset period. The preset conditions include whether the operation duration is greater than or equal to a duration threshold and whether the number of error correction bits is greater than or equal to an error correction bit number threshold. If the risk probability of the storage unit is greater than a preset risk threshold, data migration is performed on the storage unit.

[0025] As can be seen from the above technical solutions, this application provides a risk prediction method for display devices and storage units. It monitors erase and read access operations of storage units through a controller, collects two types of operating parameters—erasing operation duration and the number of error correction bits—and constructs an operating feature vector. This vector is then input into a local machine learning model, combined with duration and error correction bit thresholds, to calculate the risk probability of the storage unit over a preset period. Data migration is performed on storage units exceeding the preset risk threshold. This application's embodiments can identify high-risk storage units trending towards bad blocks in advance, and proactively migrate data to avoid important data being written to bad blocks, thus preventing program corruption and data loss, effectively improving the security and stability of data storage in display devices. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram illustrating an operational scenario between a display device and a control device provided in some embodiments of this application; Figure 2 This is a schematic diagram of the hardware configuration of a display device provided in some embodiments of this application; Figure 3 This is a schematic diagram of the software configuration of a display device provided in some embodiments of this application; Figure 4 A flowchart illustrating the process of performing storage unit risk prediction and data migration in a display device, provided in some embodiments of this application; Figure 5Timing diagrams for performing storage unit risk prediction and data migration in display devices provided in some embodiments of this application; Figure 6 A flowchart illustrating the monitoring of storage unit access operations in a display device, provided in some embodiments of this application; Figure 7 A flowchart illustrating how a display device converts operating parameters into operating feature vectors, provided in some embodiments of this application; Figure 8 A flowchart illustrating how a display device manages risk probabilities based on a risk database, provided in some embodiments of this application; Figure 9 A flowchart illustrating the data migration process performed by a display device based on system load, provided in some embodiments of this application; Figure 10 Flowcharts illustrating the specific process of data migration performed by a display device according to some embodiments of this application; Figure 11 A flowchart illustrating the data migration recovery process based on migration persistence flags for some embodiments of this application. Detailed Implementation

[0028] The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with this application. They are merely examples of systems and methods consistent with some aspects of this application as detailed in the claims.

[0029] It should be noted that the brief descriptions of terms in this application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.

[0030] The terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar or related objects or entities, and do not necessarily imply a specific order or sequence, unless otherwise specified. It should be understood that such terms are interchangeable where appropriate.

[0031] The terms “comprising” and “having”, and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device that includes a range of components is not necessarily limited to all of the components that are clearly listed, but may include other components that are not clearly listed or that are inherent to such product or device.

[0032] The term "module" refers to any known or subsequently developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functions associated with that element.

[0033] In this embodiment, the display device 200 generally refers to a device with screen display and data processing capabilities. For example, the display device 200 includes, but is not limited to, smart TVs, mobile terminals, computers, monitors, advertising screens, wearable devices, virtual reality devices, augmented reality devices, etc.

[0034] Figure 1 This is a schematic diagram illustrating an operational scenario between a display device and a control device provided in some embodiments of this application. For example... Figure 1 As shown, a user can operate the display device 200 via touch operation, a mobile terminal 300, and a control device 100. The control device 100 receives user input commands and converts them into control commands that the display device 200 can recognize and respond to. For example, the control device 100 can be a remote control, a stylus, a gamepad, etc.

[0035] The mobile terminal 300 can function as a control device for human-computer interaction between the user and the display device 200. It can also function as a communication device for establishing a communication connection with the display device 200 and exchanging data. In some embodiments, the mobile terminal 300 can have software applications installed on it and communicate with the display device 200 via network communication protocols to achieve one-to-one control and data communication. Furthermore, it can transmit audio and video content displayed on the mobile terminal 300 to the display device 200 for synchronized display.

[0036] In some embodiments, the mobile terminal 300 or other electronic devices may also simulate the functions of the control device 100 by running an application that controls the display device 200.

[0037] like Figure 1 The diagram also shows that the display device 200 communicates with the server 400 via various communication methods. This allows the display device 200 to communicate via a local area network (LAN), a wireless local area network (WLAN), and other networks.

[0038] Display device 200 can provide broadcast television reception function, and can also be equipped with intelligent network television function that provides computer support function, including but not limited to network television, smart television, Internet Protocol television (IPTV), etc.

[0039] Figure 2 Provided for some embodiments of this application Figure 1 Hardware configuration block diagram of display device 200.

[0040] In some embodiments, the display device 200 may include at least one of a tuner 210, a communication device 220, a detector 230, a device interface 240, a controller 250, a display 260, an audio output device 270, a memory, a power supply, and a user input interface 280.

[0041] In some embodiments, detector 230 is used to acquire signals from the external environment or to interact with the outside world. For example, detector 230 includes a light receiver, a sensor for acquiring ambient light intensity; or, detector 230 includes an image acquisition device, such as a camera, which can be used to acquire external environmental scenes, user attributes, or user interaction gestures; or, detector 230 includes a sound acquisition device, such as a microphone, for receiving external sounds.

[0042] In some embodiments, the display 260 includes display function components for presenting images and driving components for driving image display. The display 260 is used to receive and display image signals output from the controller 250. For example, the display 260 can be used to display video content, image content, menu control interface components, and user control UI interfaces, etc.

[0043] In some embodiments, the communication device 220 is a component used to communicate with external devices or the server 400 according to various communication protocol types. The display device 200 may have multiple communication devices 220 depending on the supported communication methods. For example, when the display device 200 supports wireless network communication, it may have a communication device 220 with WiFi functionality. When the display device 200 supports Bluetooth connectivity, it needs to have a communication device 220 with Bluetooth functionality.

[0044] The communication device 220 enables the display device 200 to communicate with external devices or the server 400 via wireless or wired connections. Wired connections utilize data cables, interfaces, or other components to connect the display device 200 to external devices. Wireless connections utilize wireless signals or wireless networks. The display device 200 can directly establish a connection with external devices or indirectly through gateways, routers, or other connection devices.

[0045] In some embodiments, the controller 250 may include at least one of a central processing unit, a video processor, an audio processor, a graphics processor, and a power processor, and a first to an nth interface for input / output. The controller 250 controls the operation of the display device and responds to user operations through various software control programs stored in memory. The controller 250 controls the overall operation of the display device 200.

[0046] In some embodiments, the controller 250 and the tuner 210 may be located in different separate devices, that is, the tuner 210 may also be located in an external device of the main device where the controller 250 is located, such as an external set-top box.

[0047] In some embodiments, a user can input user commands through a graphical user interface (GUI) displayed on a display 260, and the user input interface receives user input commands through the graphical user interface (GUI).

[0048] In some embodiments, the audio output device 270 can be a built-in speaker of the display device 200 or an external audio output device connected to the display device 200. For the external audio output device connected to the display device 200, the display device 200 may also be provided with an external audio output terminal, through which the audio output device can be connected to the display device 200 to output sound from the display device 200.

[0049] In some embodiments, the user input interface 280 can be used to receive instructions from user input.

[0050] To enable user interaction, in some embodiments, the display device 200 may run an operating system. The operating system is a computer program used to manage and control the hardware and software resources of the display device 200. The operating system can control the display device to provide a user interface; for example, the operating system can directly control the display device to provide a user interface, or it can provide a user interface by running an application. The operating system also allows users to interact with the display device 200.

[0051] It should be noted that the operating system can be a native operating system based on a specific operating platform, a third-party operating system that is deeply customized based on a specific operating platform, or an independent operating system specifically developed for display devices.

[0052] An operating system can be divided into different modules or levels based on the functions it implements, for example... Figure 3 As shown, in some embodiments, the system is divided into four layers, from top to bottom: the Applications layer (referred to as the "Application Layer"), the Application Framework layer (referred to as the "Framework Layer"), the System Library layer, and the Kernel layer.

[0053] In some embodiments, the application layer provides services and interfaces for applications, enabling the display device 200 to run applications and interact with the user based on the applications. The application layer may contain at least one application, which may be a built-in Windows program, system settings program, or clock program of the operating system; or it may be an application developed by a third-party developer. In specific implementations, the application packages in the application layer are not limited to the examples above.

[0054] The framework layer provides application programming interfaces (APIs) and a programming framework for applications. The application framework layer includes predefined functions. It acts as a central processing unit, determining the actions taken by applications within the application layer. Through the API, applications can access system resources and obtain system services during execution.

[0055] like Figure 3 As shown, the application framework layer in this embodiment includes a view system, managers, and content providers. The view system designs and implements the application's interface and interactions, and includes lists, grids, text boxes, and buttons. The managers include at least one of the following modules: an activity manager for interacting with all running activities in the system; a location manager for providing system services or applications with access to system location services; a package manager for retrieving various information related to application packages currently installed on the device; a notification manager for controlling the display and clearing of notification messages; and a window manager for managing icons, windows, toolbars, wallpapers, and desktop widgets on the user interface.

[0056] In some embodiments, the Activity Manager manages the lifecycle of individual applications and common navigation and back functions, such as controlling application exit, opening, and back actions. The Window Manager manages all window programs, such as obtaining the screen size, determining if a status bar is present, locking the screen, capturing the screen, and controlling changes to the display window, such as shrinking the display window, shaking the display, or distorting the display.

[0057] In some embodiments, the system runtime library layer can provide support for the framework layer. When the framework layer is used, the operating system runs the instruction library contained in the system runtime library layer, such as the C / C++ instruction library, to implement the functions to be performed by the framework layer.

[0058] In some embodiments, the kernel layer is a functional layer situated between the hardware and software of the display device 200. The kernel layer can implement functions such as hardware abstraction, multitasking, and memory management. For example, ... Figure 3 As shown, hardware drivers can be configured in the kernel layer. The kernel layer can contain at least one of the following drivers: audio driver, display driver, Bluetooth driver, camera driver, WIFI driver, USB driver, HDMI driver, sensor driver (such as fingerprint sensor, temperature sensor, pressure sensor, etc.), and power driver, etc.

[0059] It should be noted that the above examples are merely a simple division of operating system functions and do not limit the specific form of the operating system of the display device 200 in this application embodiment. Depending on the function of the display device, the type of operating system, and other factors, the number of layers and the specific type of the operating system may take other forms.

[0060] Display device 200 needs to store a large amount of data during operation, including operating system files, system configuration files, applications and their running data, user personal data, and various media content. As the functions of display devices such as smart TVs become increasingly sophisticated, their data storage needs continue to grow, placing higher demands on the capacity, read / write speed, and reliability of storage media.

[0061] To meet the above requirements, display device 200 typically uses eMMC (Embedded Multi Media Card) as the primary non-volatile storage medium. The NAND flash memory chips inside eMMC use memory cells as the basic storage unit, also known as memory blocks. Memory cells have inherent physical limitations: the number of programming / erase cycles that can be performed on each memory cell is finite. As erase cycles accumulate, the write speed and data retention time of the memory cell gradually deteriorate. When the degradation of a memory cell reaches a certain level, the cell will no longer be able to reliably store data, i.e., it becomes a bad block.

[0062] The presence of bad blocks poses a serious threat to the data security and system stability of the display device 200. When data is written to bad blocks, subsequent readings of that data may result in uncorrectable bit errors, leading to data corruption or loss. When bad blocks appear in critical storage areas of the eMMC, they can cause system-level failures. For example, if bad blocks are located in the partition storing the bootloader, the display device 200 will be unable to complete the system startup; if bad blocks are located in the partition storing critical system configuration files or database files, it may cause abnormal system operation, application crashes, or loss of user data.

[0063] To address the above technical problems, some embodiments of this application provide a display device 200, including a display 260, a memory, and a controller 250, wherein the memory can be an eMMC and includes multiple storage units, each of which is used to store various types of data.

[0064] In some embodiments, the operating system of this application further includes an eMMC hardware layer and an eMMC driver layer. The eMMC hardware layer includes an eMMC controller and a NAND Flash array. The eMMC controller is a built-in hardware management unit of the eMMC, natively integrating bad block management, wear leveling, ECC error correction frequency statistics, physical block address mapping, PE (Program-Erase) write / erase cycle count statistics, and block erase time statistics. The eMMC driver layer can collect hardware operating parameters such as the write / erase cycle count, ECC error correction count, and single erase operation time of each physical storage block in real time, identify and isolate faulty blocks that have failed, and balance the write / erase load of each physical block through a wear leveling mechanism. The NAND Flash array is composed of several physical storage blocks arranged in an array, used for persistent storage of the embedded device's system firmware, applications, and user business data.

[0065] Figure 4 This document provides a flowchart illustrating the process of performing storage unit risk prediction and data migration in a display device according to some embodiments of this application. Figure 5 Timing diagrams for performing storage unit risk prediction and data migration in display devices provided in some embodiments of this application.

[0066] See Figure 4 and Figure 5 The controller 250 is configured to perform the following steps: S410: Monitors access operations to storage units.

[0067] In some embodiments, the controller 250 sends monitoring requests to the eMMC driver layer by running a system-level background service, so as to continuously monitor access operations to each storage unit in the eMMC through the eMMC driver layer. These access operations include erase operations and read operations. An erase operation refers to the operation of erasing data from a storage unit to clear the data in the storage unit and restore the storage unit to a writable state; a read operation refers to the operation of reading stored data from a storage unit.

[0068] In some embodiments, the controller 250 can capture event information for each erase and read operation through the eMMC driver layer. This event information includes the operation type, the physical address of the target storage unit, the time the operation was initiated, and the operation result status code. By continuously monitoring access operations to storage units, the controller 250 can obtain the real-time usage status of each storage unit, providing a data source for subsequent operational parameter acquisition and risk prediction.

[0069] S420: In response to the access operation to the storage unit, the operating parameters of the storage unit are acquired.

[0070] In some embodiments, after detecting an access operation to a storage unit, the controller 250 collects the operating parameters of the storage unit in response to the access operation. The operating parameters are key indicator data reflecting the current health status of the storage unit, including the operation time of performing an erase operation and the number of error correction bits of the Error Correction Code (ECC) generated when performing a read operation.

[0071] Operation time refers to the time elapsed from when the controller 250 sends an erase command to the eMMC controller until the eMMC controller returns an erase completion signal. As the physical wear of the storage cell intensifies, the erase operation time of the storage cell usually gradually increases. Therefore, operation time can be used as an important indicator reflecting the degree of wear of the storage cell.

[0072] Error correction bits refer to the number of error bits detected and successfully corrected by the eMMC controller's ECC engine during a read operation. When writing data, the eMMC calculates and stores ECC checksums, and uses these checksums to detect and correct bit errors in the data during read operations. As storage cells age, data retention capabilities decline, and the number of bit errors during reads increases, leading to a corresponding increase in the number of ECC error correction bits. Therefore, error correction bits serve as an important indicator of a storage cell's data retention capability and reliability.

[0073] After each erase and read operation is completed, the controller 250 associates the collected operation duration and number of error correction bits with the corresponding physical address of the storage unit and stores them to form a record of the historical operating parameters of the storage unit.

[0074] S430: Convert the running parameters into running feature vectors.

[0075] In some embodiments, after acquiring the operating parameters, the controller 250 converts the operating parameters into operating feature vectors that can be recognized by a machine learning model. The operating feature vector is a multi-dimensional numerical vector obtained by statistically transforming the original operating parameters, and is used to describe the operating state characteristics of the storage unit at different time dimensions.

[0076] Specifically, the controller 250 performs statistical analysis on the operation duration and error correction bits within a preset time window, calculating statistical parameters such as the average and standard deviation of the operation duration and the average and maximum values ​​of the error correction bits, and calculates the temporal change rate of the operation duration and error correction bits. These statistical parameters and change rate features are then combined into a runtime feature vector. The generation process of the runtime feature vector will be described in detail in subsequent embodiments. By converting the original operating parameters into a standardized runtime feature vector, deep features reflecting changes in the health status of the storage units can be extracted, providing high-quality input data for machine learning models.

[0077] S440: Input the running feature vector into the locally deployed machine learning model so that the machine learning model performs a risk probability threshold on the running feature vector based on preset conditions, and obtains the risk probability of the storage unit in the future preset period.

[0078] In some embodiments, the controller 250 inputs the runtime feature vector obtained in step S430 into a locally deployed machine learning model. The machine learning model is a pre-trained prediction model deployed locally on the display device 200, used to predict the risk probability of a storage unit within a future preset period based on the input runtime feature vector. The risk probability ranges from 0% to 100%, representing the likelihood that the storage unit will develop into a bad block within the future preset period. The machine learning model performs risk probability prediction on the runtime feature vector based on preset conditions, including whether the operation duration is greater than or equal to a duration threshold and whether the number of error correction bits is greater than or equal to an error correction bit threshold. The duration threshold is a critical value for the erase operation duration determined based on the specifications provided by the eMMC manufacturer and statistical analysis of a large amount of actual operating data; the error correction bit threshold is a critical value for the number of error correction bits determined based on the eMMC's ECC error correction capability and data reliability requirements.

[0079] It is understood that the preset cycle can be set to the next erase / write cycle, a preset number of days, or a preset number of erase / write cycles, etc., and this application does not limit this. By deploying the machine learning model locally on the display device 200, risk prediction can be completed independently in an environment without a network connection, ensuring the real-time nature of the prediction and the privacy and security of user data.

[0080] In some embodiments, after the machine learning model outputs the risk probability of a storage unit within a preset future period, the controller 250 obtains the risk probability of the storage unit. The risk probability is a quantitative expression of the risk to the health status of the storage unit, enabling the controller 250 to make decisions based on explicit numerical thresholds, rather than relying on empirical fuzzy judgments.

[0081] S450: Perform data migration on the storage unit if the risk probability of the storage unit is greater than the preset risk threshold.

[0082] In some embodiments, the controller 250 compares the risk probability output by the machine learning model with a preset risk threshold. The preset risk threshold is a critical probability value used to determine whether data migration of a storage unit is necessary. It can be set according to the actual application scenario, for example, the preset risk threshold can be set to 80%. When the risk probability of a storage unit is greater than the preset risk threshold, the controller 250 determines that the storage unit has a high risk of bad blocks and triggers the data migration process.

[0083] Data migration refers to reading valid data stored in high-risk storage units and writing it to pre-selected low-risk storage units. After data migration is completed, the controller 250 can mark the original storage unit as high-risk or migrated, ensuring that the display device 200 will not assign new write tasks to that storage unit. When the risk probability of a storage unit is less than or equal to a preset risk threshold, the controller 250 determines that the storage unit is currently in a safe state, maintains normal use of the storage unit, and returns to step S410 to continue cyclic monitoring.

[0084] Based on the above technical solutions, this application embodiment monitors the erase and read access operations of storage units, collects the operation duration and error correction bits as operating parameters, converts the operating parameters into operating feature vectors, and inputs them into a locally deployed machine learning model for risk probability prediction. When the risk probability exceeds a preset risk threshold, data migration is proactively triggered. The display device 200 can identify high-risk storage units in advance before the eMMC storage units actually become bad blocks, and proactively migrate data from high-risk storage units. This achieves a shift from passive bad block detection to proactive risk prevention, effectively preventing serious problems such as operating system failure to boot, application crashes, and user data loss caused by sudden storage unit damage, thereby improving the system reliability and data security of the display device.

[0085] Figure 6 This is a flowchart illustrating the monitoring of storage unit access operations for display devices according to some embodiments of this application.

[0086] See Figure 6In some embodiments, the controller 250 performs the monitoring of access operations to the storage unit, which may include the following steps: S610: Real-time monitoring of erase and read operations on storage units via a background resident process.

[0087] In some embodiments, the controller 250 implements real-time monitoring of storage unit access operations by starting a background resident process. The background resident process automatically loads and runs after the operating system of the display device 200 starts, and remains continuously resident in the system background during the operation of the display device 200. When the eMMC controller performs an erase or read operation on a storage unit, the controller 250 sends an operation notification to the background resident process. The notification includes the operation type, the physical address of the target storage unit, the operation initiation time, and the operation result status code. Upon receiving the operation notification, the background resident process immediately triggers the eMMC driver layer runtime parameter acquisition process, recording the operation duration of the erase operation or the number of error correction bits generated by the read operation.

[0088] By using a background resident process for real-time monitoring, the controller 250 can capture relevant operating parameters the instant each storage unit access operation occurs through the eMMC driver layer, ensuring that no critical operation event is missed, thereby obtaining the most complete operating parameter data, providing sufficient and timely input data for machine learning models, and improving the accuracy and timeliness of risk prediction.

[0089] S620: Monitors erase and read operations on storage units at preset time intervals.

[0090] In other embodiments, the controller 250 can periodically monitor erase and read operations on the storage unit according to a preset time period, i.e., it adopts an idle monitoring mode. The preset time period can be configured according to actual needs, for example, the preset time period can be set to once per hour, once per day, or the monitoring can be performed when the display device 200 enters standby mode.

[0091] In timed monitoring mode, when a preset time period arrives, the controller 250 reads the erase and read operation records of each storage unit within that time period through the query interface of the eMMC driver layer, and extracts operating parameters such as operation duration and error correction bits from the records. Compared with real-time monitoring mode, timed monitoring mode consumes fewer system resources and is suitable for application scenarios that are more sensitive to system resources, or scenarios where the storage unit access frequency is low and the real-time requirements for risk prediction are not high.

[0092] In some embodiments, the display device 200 can support both of the above monitoring modes simultaneously, and the controller 250 can dynamically switch the monitoring mode according to the current operating state of the display device 200. For example, when the display device 200 is in active use, a real-time monitoring mode with a background resident process is adopted to ensure that no critical operation events are missed; when the display device 200 is in standby or low-load state, it is switched to a timed monitoring mode to further reduce system resource consumption.

[0093] Based on the above technical solutions, this application provides two access operation monitoring methods: real-time monitoring of background resident processes and timed monitoring with preset time periods. Through flexible monitoring strategies, the display device 200 can select the most suitable monitoring method according to its own operating status and actual needs, ensuring the accuracy of risk prediction while optimizing system resource usage and improving the practicality and adaptability of the monitoring mechanism.

[0094] Figure 7 This is a flowchart illustrating how a display device, according to some embodiments of this application, converts operating parameters into operating feature vectors.

[0095] See Figure 7 In some embodiments, the controller 250 performs the conversion of operating parameters into operating feature vectors, which may include the following steps: S710: Calculate the operation duration within the preset time window and obtain the first statistical parameter corresponding to the operation duration.

[0096] In some embodiments, the controller 250 first determines a preset time window, which is a time range for statistical analysis. For example, the preset time window can be set to the time range corresponding to the most recent ten erase operations. After determining the preset time window, the controller 250 extracts all operation duration data within the preset time window from the historical operating parameter records of the storage unit, and performs statistical analysis on the operation duration data to obtain a first statistical parameter. The first statistical parameter includes the average operation duration and the standard deviation of the operation duration. The average operation duration reflects the overall level of erase operation time consumed by the storage unit within the preset time window.

[0097] S720: Count the number of error correction bits within a preset time window to obtain the second statistical parameter corresponding to the number of error correction bits. The second statistical parameter includes the average number of error correction bits and the maximum number of error correction bits.

[0098] In some embodiments, the controller 250 extracts a preset number of error correction bits (e.g., the number of error correction bits from nearly one hundred read operations) from the historical operating parameter records of the storage unit within a preset time window, and performs statistical analysis on the error correction bit data to obtain a second statistical parameter. The second statistical parameter includes the average number of error correction bits and the maximum number of error correction bits. The average number of error correction bits reflects the average error level of the storage unit during data reading within the preset time window. The maximum number of error correction bits reflects the most severe error situation that occurs during data reading by the storage unit within the preset time window.

[0099] S730: According to the preset time window, calculate the timing change rate of the operation duration and the number of error correction bits to obtain the change rate characteristics.

[0100] In some embodiments, based on the first and second statistical parameters, the controller 250 calculates the timing change rate of the operation duration and the number of error correction bits according to a preset time window to obtain the change rate characteristic. The timing change rate is used to describe the evolution trend of operating parameters over time and can reflect whether the health status of the storage unit is stable, improving, or deteriorating.

[0101] S740: Combine the first statistical parameter, the second statistical parameter, and the rate of change feature into a running feature vector.

[0102] In some embodiments, the controller 250 combines the obtained first statistical parameter, second statistical parameter, and rate of change feature into a multidimensional running feature vector. After generating the running feature vector, the controller 250 can also perform normalization processing on the feature values ​​of each dimension in the running feature vector, mapping the feature values ​​of each dimension to a uniform numerical range, such as between 0 and 1, eliminating the dimensional differences between different feature dimensions, and improving the training stability and prediction accuracy of the machine learning model.

[0103] Based on the above technical solutions, this application embodiment performs statistical analysis on the operation duration and error correction bits within a preset time window, extracting the average and standard deviation of the operation duration and the average and maximum values ​​of the error correction bits as statistical parameters. It also calculates the temporal change rate characteristics of the operation duration and error correction bits, combining the statistical parameters and change rate characteristics into a runtime feature vector. Through this multi-dimensional feature extraction method, the runtime feature vector can comprehensively reflect the health status and changing trends of the storage unit from multiple dimensions such as erase operation time and data read errors, providing high-quality input data with rich information for machine learning models, thereby improving the accuracy and reliability of bad block risk prediction.

[0104] Figure 8 A flowchart illustrating risk probability management based on a risk database for some embodiments of this application.

[0105] See Figure 8 In some embodiments, the memory is also configured to store a risk database, which stores the risk probability corresponding to each storage unit. When the controller 250 performs data migration on a storage unit if the risk probability of that storage unit is greater than a preset risk threshold, this may include the following steps: S810: Store the risk probability corresponding to the storage unit into the risk database.

[0106] In some embodiments, after the machine learning model outputs the risk probability of a certain storage unit, the controller 250 associates the risk probability with the physical address of the storage unit and stores it in a risk database in the memory. The risk database is a structured database used to centrally manage the risk probabilities of all storage units in the eMMC. Each record in the risk database may include the physical address of the storage unit, the most recently predicted risk probability, the risk probability prediction timestamp, historical risk probability trends, and the current status identifier of the storage unit, such as normal, under observation, high risk, or migrated.

[0107] After each risk prediction by the machine learning model, the controller 250 updates the risk probability record of the corresponding storage unit in the risk database, ensuring that the risk database always stores the latest risk assessment information for each storage unit. By establishing a risk database, the controller 250 can uniformly manage and continuously track the risk status of all storage units in the eMMC, providing a data foundation for subsequent risk verification and data migration decisions.

[0108] S820: Based on a preset risk threshold, periodically verify the risk probability stored in the risk database, and then perform data migration on storage units with risk probabilities greater than the preset risk threshold.

[0109] In some embodiments, the controller 250 periodically verifies the risk probability of each storage unit stored in the risk database based on a preset verification period. The verification period can be configured according to actual needs, for example, setting the verification period to once per hour or once per day.

[0110] During each verification process, the controller 250 iterates through all records in the risk database, comparing the risk probability of each storage unit with the preset risk threshold. For storage units with a risk probability greater than the preset risk threshold, the controller 250 adds the storage unit to the migration queue and performs data migration on the storage units in the queue sequentially. For storage units with a risk probability less than or equal to the preset risk threshold, the controller 250 maintains normal use of the storage unit without performing any special processing.

[0111] The controller 250 can also sort the storage units in the migration queue according to their risk probability, prioritizing the migration of the storage units with the highest risk probability to ensure that the most dangerous storage units receive the highest priority. By periodically verifying the risk database, the controller 250 can perform a comprehensive check on the health status of all storage units in the eMMC at fixed time points, ensuring that no storage unit whose risk probability has exceeded the preset risk threshold but has not yet triggered data migration is missed, thus improving the timeliness and comprehensiveness of data migration.

[0112] Based on the above technical solutions, this application embodiment establishes a risk database to centrally manage the risk probability of each storage unit, and periodically verifies the risk database based on a preset risk threshold. Data migration is performed on storage units with a risk probability greater than the preset risk threshold. Through this mechanism, the display device 200 can continuously track and periodically verify the risk status of all storage units, ensuring that high-risk storage units are handled promptly and improving the timeliness of data migration.

[0113] Figure 9 This is a flowchart illustrating the data migration process performed by a display device based on system load, as provided in some embodiments of this application.

[0114] See Figure 9 In some embodiments, the controller 250 performs data migration on the storage unit, which may further include the following steps: S910: Obtain the load value of the display device.

[0115] In some embodiments, after determining that data migration needs to be performed, the controller 250 first obtains the current load value of the display device 200. The load value is a quantitative indicator used to measure the current system activity level of the display device 200, and may include a combination of one or more dimensions such as Central Processing Unit (CPU) utilization, memory usage, Input / Output (I / O) bandwidth utilization, and the number and type of currently running foreground applications. By obtaining the real-time load value of the display device 200, the controller 250 can understand the current system resource usage and provide a basis for deciding when to perform data migration.

[0116] S920: Perform data migration on storage units when the load value is less than or equal to a preset load threshold.

[0117] In some embodiments, the controller 250 compares the acquired load value with a preset load threshold. The preset load threshold is a critical value used to distinguish whether the display device 200 is in a low-load or idle state, and can be set according to the actual application scenario. When the load value is less than or equal to the preset load threshold, the controller 250 determines that the display device 200 is currently in a low-load or idle state. At this time, performing data migration operation will not significantly affect the smoothness and response speed of normal user use. The controller 250 then initiates the data migration process to transfer the valid data in the high-risk storage unit. The data migration process may involve a large number of data read and write operations, which consumes a certain amount of eMMC I / O bandwidth. Therefore, performing data migration under low system load can minimize the impact on user experience.

[0118] S930: If the load value is greater than the preset load threshold, postpone the data migration until the load value is less than or equal to the preset load threshold.

[0119] In some embodiments, when the load value exceeds a preset load threshold, the controller 250 determines that the display device 200 is currently in a high-load state, such as when the user is playing a high-definition video, running a large application, or performing a system update. If data migration is performed immediately at this time, the I / O operation generated by the data migration may compete with the user's current task for system resources, resulting in problems such as video playback stuttering and application response delay, which will affect the user experience.

[0120] Therefore, the controller 250 postpones the data migration operation, continuously monitors changes in the load value, and only initiates the data migration when the load value drops below a preset load threshold. It is understandable that during the delayed data migration period, the data in high-risk storage units still faces a certain risk of corruption. To reduce the risk of data loss during the delay, the controller 250 can perform additional verification and protection measures on the data in high-risk storage units during the delay period, such as increasing the frequency of ECC checks for data reads or performing temporary backups of critical data.

[0121] In some embodiments, the controller 250 may also set a maximum delay time. When the data migration delay time exceeds the maximum delay time, the data migration will be forcibly executed even if the load value is still greater than the preset load threshold, so as to avoid the data migration being delayed indefinitely and thus causing a decrease in data security.

[0122] Based on the above technical solutions, this embodiment of the application obtains the load value of the display device 200 before performing data migration. Data migration is performed when the load value is less than or equal to a preset load threshold, and postponed when the load value is greater than the preset load threshold. Therefore, the data migration operation can be performed during system idle periods, avoiding competition for system resources with normal user tasks, ensuring that the data migration process is imperceptible to the user experience, and maintaining the smooth operation of the display device while ensuring data security.

[0123] Figure 10 This is a flowchart illustrating the specific process of data migration performed by a display device according to some embodiments of this application.

[0124] See Figure 10 In some embodiments, the controller 250 performs data migration on the storage unit, which may include the following steps: S1010: Storage units where the probability of risk in the risk database query is less than or equal to the risk threshold are designated as low-risk storage units.

[0125] In some embodiments, when performing data migration, the controller 250 first needs to determine the target location for the data migration, i.e., a low-risk storage unit. The controller 250 queries the risk probability of all current storage units from the risk database and filters out storage units with a risk probability less than or equal to a preset risk threshold as low-risk storage units. Among the low-risk storage units, the controller 250 can further select the storage unit with the lowest wear level and the longest remaining life as the target storage unit based on the storage unit's wear level indicators, such as cumulative write / erase cycles and remaining life assessment information, thereby receiving valid data migrated from high-risk storage units.

[0126] In some embodiments, the controller 250 may preferentially select low-risk storage units that are in an idle state (i.e., currently not storing valid data) as target storage units, avoiding the need to perform additional data erasure operations on the target storage units during data migration. By querying low-risk storage units from the risk database, it can be ensured that the target location for migration has sufficient data storage reliability, avoiding the migration of data from one high-risk storage unit to another storage unit with similar risks.

[0127] S1020: Read valid data stored in a storage unit where the risk probability is greater than a preset risk threshold.

[0128] In some embodiments, after determining the target storage unit, the controller 250 performs a read operation on the valid data stored in the high-risk storage unit (i.e., the storage unit with a risk probability greater than a preset risk threshold). Valid data refers to user data or system data currently stored in the high-risk storage unit that has a valid logical address mapping relationship.

[0129] S1030: Write the read valid data to the low-risk storage unit.

[0130] In some embodiments, the controller 250 writes valid data read from a high-risk storage unit to a low-risk storage unit (target storage unit). The write operation is performed block by block, and after each data block is written, the eMMC controller performs a verification of the write result to ensure that the data has been correctly written to the target location.

[0131] After the data is successfully migrated to a low-risk storage unit, the controller 250 can mark the high-risk storage unit as migrated in the risk database. The system will then no longer assign new write tasks to the high-risk storage unit, thereby avoiding the reuse of the unreliable storage unit.

[0132] Based on the above technical solutions, in the data migration process, this application embodiment queries a low-risk storage unit from the risk database as the migration target, reads valid data from the high-risk storage unit, and writes it to the low-risk storage unit. Through the above complete data migration process, the data in the high-risk storage unit is safely transferred to a reliable storage location, effectively avoiding data loss due to sudden storage unit failure and ensuring data security and integrity.

[0133] Figure 11 A flowchart illustrating the data migration recovery process based on migration persistence flags for some embodiments of this application.

[0134] See Figure 11 In some embodiments, the controller 250 performs data migration on the storage unit when the risk probability of the storage unit is greater than a preset risk threshold, which may include the following steps: S1110: During the data migration process, write the migration persistence flag into memory.

[0135] In some embodiments, when initiating the data migration process, the controller 250 writes a persistent migration flag to the non-volatile storage area of ​​the memory. The persistent migration flag is structured data used to record the current execution status of the data migration task and will not be lost after the display device 200 is powered off. The persistent migration flag may include the physical address of the storage unit of the valid data to be migrated, the physical address of the low-risk storage unit, and migration status information. The migration status information indicates the current execution progress of the data migration task and may include a pending migration status, a migration in progress status, and a migration completed status.

[0136] When the data migration begins, the controller 250 sets the migration status information to the pending migration state; during the process of reading and writing data block by block, it continuously updates the migration status information in the migration persistence flag, and records the sequence number or offset of the data block that has been successfully migrated; after all valid data migration is completed, the migration status information is updated to the migration completed state, and the migration persistence flag is cleared from the memory.

[0137] S1120: If the data migration is interrupted and the display device is powered on again, detect the migration persistence flag.

[0138] In some embodiments, when data migration is interrupted, for example, when the display device 200 experiences an unexpected power outage during data migration, causing the data migration to be interrupted, the controller 250 automatically detects whether there is a valid migration persistence flag in the memory during the initialization phase of the system startup when the display device 200 is powered on again.

[0139] The controller 250 can check whether the migration persistent flag data structure exists by reading a specified non-volatile storage area of ​​the memory, and verify the validity of the migration persistent flag by checking the integrity of the migration persistent flag.

[0140] If a valid migration persistence flag is detected, the controller 250 determines that there was an incomplete data migration task that was interrupted due to a power outage during the last operation, and then enters the data migration recovery process; if no migration persistence flag is detected or the migration persistence flag has expired, the controller 250 determines that there is no incomplete data migration task, and normally enters the regular eMMC storage unit monitoring and risk prediction process.

[0141] S1130: If a migration persistence flag is detected, resume the execution of the incomplete data migration based on the physical address of the storage unit, the physical address of the low-risk storage unit, and the migration status information.

[0142] In some embodiments, after detecting a valid migration persistence flag, the controller 250 parses the physical addresses of high-risk storage units (i.e., storage units with a risk probability greater than a preset risk threshold), the physical addresses of low-risk storage units, and migration status information recorded in the migration persistence flag. Based on the migration status information, the controller 250 determines the range of data blocks that have been migrated, and then continues to perform read and write operations on the remaining valid data, starting from the next unmigrated data block.

[0143] In some embodiments, during the resumption of data migration, the controller 250 continuously updates the migration status information in the migration persistence flag until all valid data migration is completed, at which point the migration persistence flag is cleared. It should be noted that during the resumption of data migration, the controller 250 still needs to determine the current load state of the display device 200, and only performs the data migration resumption operation when the display device 200 is in a low-load or idle state, avoiding the impact on user experience from performing data migration during the high-load initialization phase after power-on.

[0144] Based on the above technical solutions, this application embodiment writes a persistent migration flag containing the physical address of a high-risk storage unit, the physical address of a low-risk storage unit, and migration status information during the data migration process. When the display device 200 is powered on again after the data migration is interrupted, the persistent migration flag is detected, and the incomplete data migration is resumed based on the information in the persistent migration flag.

[0145] In some embodiments, after the controller 250 executes the writing of the read valid data to the low-risk storage unit, it may also update the metadata of the file system or the mapping table maintained by the controller so that the logical address of the valid data is mapped to the physical address of the low-risk storage unit.

[0146] In some embodiments, after valid data is successfully written from a high-risk storage unit to a low-risk storage unit, the controller 250 needs to update the address mapping relationship of the data access to ensure that subsequent access to the valid data can be correctly routed to the low-risk storage unit, rather than continuing to access the marked high-risk storage unit.

[0147] The mapping relationship between logical addresses and physical addresses can be recorded through file system metadata or a mapping table maintained by controller 250. When valid data is migrated from a high-risk storage unit to a low-risk storage unit, the physical address of the data changes, but the logical address of the data remains unchanged. Therefore, it is necessary to update the mapping relationship between logical addresses and physical addresses, remapping the logical address of the valid data to the physical address of the target low-risk storage unit.

[0148] Based on the above technical solutions, this embodiment of the application updates the file system metadata or the mapping table maintained by the controller in a timely manner after the data migration is completed, mapping the logical address of valid data to the physical address of low-risk storage units, and performing marking and notification processing on high-risk storage units. Through the above address mapping update mechanism, it is ensured that the upper-layer application's access to data is not affected by the underlying data migration, realizing complete transparency of the data migration process to the upper-layer application, ensuring the correctness and consistency of data access, and achieving complete isolation of faulty storage units.

[0149] In some embodiments, the controller 250 may call the file system interface to update the mapping relationship between the logical address of valid data recorded in the file system metadata and the physical address of low-risk storage unit when updating the file system metadata.

[0150] In some embodiments, when the display device 200 uses a file system to manage the eMMC storage space, the mapping relationship between logical addresses and physical addresses is maintained by the file system's metadata. The file system's metadata is a set of structured information used by the file system to manage storage space and file data, recording the correspondence between the logical address and physical address of each file data block. When valid data is migrated from a high-risk storage unit to a low-risk storage unit, the controller 250 updates the file system's metadata by calling the application programming interface (API) provided by the file system, modifying the metadata record that originally mapped the valid data to the physical address of the high-risk storage unit to map it to the physical address of the low-risk storage unit.

[0151] In other embodiments, the controller 250 updates a controller-maintained mapping table, which may update the mapping relationship between the logical addresses of valid data and the physical addresses of low-risk storage units recorded in the controller-maintained mapping table. The mapping relationship between logical addresses and physical addresses is recorded by the mapping table maintained by the controller 250. The mapping table maintained by the controller 250 is a lookup table stored in memory or non-volatile memory, indexed by logical addresses or logical address ranges, recording the eMMC physical address corresponding to each logical address or logical address range. When valid data is migrated from a high-risk storage unit to a low-risk storage unit, the controller 250 directly updates the corresponding entry in the mapping table, replacing the source storage unit's physical address with the target low-risk storage unit's physical address.

[0152] In some embodiments, the controller 250 can also employ both of the above-mentioned address mapping update methods simultaneously. For example, at the file system level, the mapping relationship in the metadata is updated by calling the file system interface, while at the controller 250 level, the mapping table is updated synchronously, forming a dual address mapping guarantee and improving the reliability and fault tolerance of address mapping updates.

[0153] Based on the above technical solutions, this application provides two address mapping update methods: calling the file system interface to update metadata and updating the mapping table maintained by the controller, and supports a dual mapping update mechanism. By flexibly selecting the address mapping update method that adapts to the current storage management architecture, it is possible to ensure that correct and efficient address mapping updates after data migration can be achieved under different storage management schemes, guaranteeing the correctness of data access and the reliability of migration, and improving the versatility and applicability of the solution.

[0154] Some embodiments of this application also provide a method for predicting storage cell risk, including: S410: Monitors access operations to storage units.

[0155] The access operations include erase operations and read operations.

[0156] S420: In response to access operations on storage units, collect the operating parameters of the storage units.

[0157] The operating parameters include the operation time for the erase operation and the number of error correction bits generated during the read operation.

[0158] S430: Convert the running parameters into running feature vectors.

[0159] S440: Input the running feature vector into the locally deployed machine learning model so that the machine learning model performs risk probability prediction on the running feature vector based on preset conditions, and obtains the risk probability of the storage unit in the future preset period.

[0160] The preset conditions include whether the operation time is greater than or equal to the time threshold, and whether the number of error correction bits is greater than or equal to the number of error correction bits threshold.

[0161] S450: Perform data migration on the storage unit if the risk probability of the storage unit is greater than the preset risk threshold.

[0162] As can be seen from the above technical solutions, this application provides a risk prediction method for display devices and storage units. It monitors erase and read access operations of storage units through a controller, collects two types of operating parameters—erasing operation duration and the number of error correction bits—and constructs an operating feature vector. This vector is then input into a local machine learning model, combined with duration and error correction bit thresholds, to calculate the risk probability of the storage unit over a preset period. Data migration is performed on storage units exceeding the preset risk threshold. This application's embodiments can identify high-risk storage units trending towards bad blocks in advance, and proactively migrate data to avoid important data being written to bad blocks, thus preventing program corruption and data loss, effectively improving the security and stability of data storage in display devices.

[0163] The same or similar parts among the various embodiments in this specification can be referred to mutually, and will not be repeated here.

[0164] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or certain parts of the embodiments of the present invention.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0166] For ease of explanation, the above description has been provided in conjunction with specific embodiments. However, the above exemplary discussion is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed above. Various modifications and variations can be obtained based on the above teachings. The selection and description of the above embodiments are for the purpose of better explaining the principles and practical applications, thereby enabling those skilled in the art to better utilize the described embodiments and various different variations of embodiments suitable for specific use considerations.

Claims

1. A display device, characterized in that, include: monitor; A memory includes multiple storage units, which are used to store various types of data; The controller is configured as follows: Monitor access operations to the storage unit, wherein the access operations include erase operations and read operations; In response to the access operation to the storage unit, the operating parameters of the storage unit are collected, including the operation duration of the erase operation and the number of error correction bits of the error correction code generated when the read operation is performed; Convert the operating parameters into operating feature vectors; The running feature vector is input into a locally deployed machine learning model so that the machine learning model performs risk probability prediction on the running feature vector based on preset conditions to obtain the risk probability of the storage unit in the future preset period. The preset conditions include whether the operation duration is greater than or equal to a duration threshold and whether the number of error correction bits is greater than or equal to an error correction bit number threshold. If the risk probability of the storage unit is greater than a preset risk threshold, data migration is performed on the storage unit.

2. The display device according to claim 1, characterized in that, The controller performs monitoring of access operations to the storage unit, specifically configured as follows: The erase and read operations on the storage unit are monitored in real time by a background resident process. Alternatively, the erase and read operations on the storage unit can be monitored periodically according to a preset time period.

3. The display device according to claim 1, characterized in that, The controller is configured to convert the operating parameters into an operating feature vector. The operation duration within a preset time window is statistically analyzed to obtain a first statistical parameter corresponding to the operation duration. The first statistical parameter includes the average value of the operation duration and the standard deviation of the operation duration. The number of error correction bits within a preset time window is statistically analyzed to obtain a second statistical parameter corresponding to the number of error correction bits. The second statistical parameter includes the average number of error correction bits and the maximum number of error correction bits. According to the preset time window, the timing change rate of the operation duration and the number of error correction bits is calculated to obtain the change rate characteristics; The first statistical parameter, the second statistical parameter, and the rate of change feature are combined to form the running feature vector.

4. The display device according to claim 1, characterized in that, The memory is also configured to store a risk database, which stores the risk probabilities corresponding to the storage units; the controller performs data migration on the storage units when the risk probability of a storage unit is greater than a preset risk threshold, specifically configured as follows: The risk probability corresponding to the storage unit is stored in the risk database; Based on the preset risk threshold, the risk probability stored in the risk database is periodically verified, thereby performing data migration on storage units with risk probabilities greater than the preset risk threshold.

5. The display device according to claim 4, characterized in that, The controller performs data migration on the storage unit and is further configured to: Obtain the load value of the display device; If the load value is less than or equal to a preset load threshold, the data migration is performed on the storage unit; If the load value is greater than a preset load threshold, the data migration will be postponed until the load value is less than or equal to the preset load threshold.

6. The display device according to claim 5, characterized in that, The controller performs data migration on the storage unit, specifically configured as follows: The risk database is used to query storage units whose risk probability is less than or equal to the risk threshold as low-risk storage units. Read valid data stored in storage units where the risk probability is greater than a preset risk threshold; The read valid data is written to the low-risk storage unit.

7. The display device according to claim 6, characterized in that, The controller performs data migration on the storage unit when the risk probability of the storage unit is greater than a preset risk threshold, specifically configured as follows: During the data migration process, a migration persistence flag is written into the memory. The migration persistence flag includes the physical address of the storage unit of the valid data to be migrated, the physical address of the low-risk storage unit, and migration status information. If the data migration is interrupted and the display device is powered on again, the migration persistence flag is detected; If the migration persistence flag is detected, the incomplete data migration is resumed based on the physical address of the storage unit, the physical address of the low-risk storage unit, and the migration status information.

8. The display device according to claim 6, characterized in that, After the controller executes the action of writing the read valid data into the low-risk storage unit, it is further configured to: Update the file system metadata or the mapping table maintained by the controller so that the logical address of the valid data is mapped to the physical address of the low-risk storage unit.

9. The display device according to claim 8, characterized in that, The controller performs updates to the file system metadata or the mapping table maintained by the controller, specifically configured as follows: Call the file system interface to update the mapping relationship between the logical address of the valid data recorded in the file system metadata and the physical address of the low-risk storage unit; Alternatively, update the mapping relationship between the logical address of the valid data and the physical address of the low-risk storage unit recorded in the mapping table maintained by the controller.

10. A method for predicting the risk of a storage cell, characterized in that, include: Monitor access operations to the storage unit, wherein the access operations include erase operations and read operations; In response to the access operation to the storage unit, the operating parameters of the storage unit are collected, including the operation duration of the erase operation and the number of error correction bits of the error correction code generated when the read operation is performed; Convert the operating parameters into operating feature vectors; The running feature vector is input into a locally deployed machine learning model so that the machine learning model performs a risk probability threshold on the running feature vector based on preset conditions to obtain the risk probability of the storage unit in a future preset period. The preset conditions include whether the operation duration is greater than or equal to a duration threshold and whether the number of error correction bits is greater than or equal to an error correction bit number threshold. If the risk probability of the storage unit is greater than a preset risk threshold, data migration is performed on the storage unit.

Images