Write management method of NAND type solid state storage device

Abstract

A write management method of an NAND type solid state storage device dynamically generates a write prohibition condition and controls a write operation by establishing a write context record, monitoring a write consequence and performing a responsibility attribution analysis. Before each time of data writing, a context record containing a writing strategy, an NAND block state and a data attribute is established; after data writing is completed, ECC correction events, retention errors, and unintended rewrite or read failure events of the NAND block are monitored during background observation. When the monitoring result exceeds a preset allowable range, the context record is traced back, the failure possibility of the combination of the specific writing strategy and the block state is judged, and a corresponding writing forbidding condition is established. According to the method, the accumulated failure frequency of a single writing strategy or a single NAND block is calculated, strategy or block level limitation is established, and the strategy and the block limitation are dynamically updated according to an observation period, so that the SSD can still effectively reduce the high-risk event occurrence rate at different loads and aging stages, the data reliability is improved, and the service life of the NAND block is prolonged.

Description

Technical Field

[0001] This invention relates to control technology for solid-state storage devices (SSDs), and more particularly to a write governance method for NAND-type solid-state storage devices. This method combines write policy management, block health monitoring, and dynamic prohibition condition generation to improve data reliability, reduce the occurrence rate of high-risk write combinations, and extend the lifespan of NAND blocks. Background Technology

[0002] Existing NAND solid-state storage devices mostly employ write optimization mechanisms such as data reclamation, block balancing, and caching management to maintain storage performance and extend NAND block lifespan. These methods typically prevent overuse of some blocks by periodically moving data or evenly distributing write operations, thereby delaying block aging and improving average lifespan. However, these technologies are mostly static or periodic adjustments, lacking the ability to perceive block health status in real time, and also lacking continuous monitoring and dynamic adjustment mechanisms for the consequences of data writes. This means that some high-risk blocks may still experience frequent failures under long-term use or high-frequency write conditions. In practical applications, enterprise-level or high-performance storage systems often face large-scale, high-frequency data writes, and some NAND blocks are prone to cumulative aging, leading to decreased data retention, increased error correction frequency, and even unexpected write failures or data loss events. Existing technologies mostly only improve overall performance or average lifespan, lacking fine-grained management methods for individual blocks or single write operations. Especially under long-term operation or high load environment, the block health status of SSD may change rapidly, but existing methods cannot update write policies in real time or automatically limit high-risk operations, which can easily lead to a decline in data reliability.

[0003] Furthermore, existing technologies largely rely on static rules or empirical parameters for write scheduling, such as preset block balancing or periodic data reclamation. The adjustment frequency of these parameters is unrelated to block health status, making it difficult to respond quickly to sudden or cumulative problems. For high-frequency hot data writes, some blocks may experience a large number of write operations in a short period, causing error accumulation to exceed the controller's expectations. Such short-term health degradation is difficult to detect and handle in real time with existing technologies, increasing the risk of data loss. Existing technologies also lack a unified failure event assessment mechanism. Most only perform average management on a single block or a single write strategy, failing to integrate historical write data, error events, and block health status for a comprehensive assessment. They also lack means to dynamically generate write limits or adjust write operations. Therefore, even if SSDs are equipped with health monitoring or error correction mechanisms, the lack of refined and dynamic governance management can still lead to high-risk write combinations under different operating conditions and aging stages, resulting in decreased data reliability or premature block failure. Therefore, it is necessary to provide a method that integrates write strategy determination, block health analysis, and dynamic management control to effectively reduce the incidence of high-risk events, improve data reliability, and extend the lifespan of NAND blocks. Summary of the Invention

[0004] One objective of this invention is to provide a write management method for NAND solid-state storage devices. This method dynamically generates write prohibition conditions and manages write operations by establishing write context records, monitoring write consequences, and performing responsibility attribution analysis. This improves data reliability, reduces the incidence of high-risk write combinations, and extends the lifespan of each NAND block. The method can make refined judgments based on data write strategies, block health status, and data attributes. Combined with background observation and cumulative failure analysis, it dynamically adjusts the write control strategy, enabling the storage device to maintain stable operation and data integrity under different operating loads and block aging stages.

[0005] To achieve the above objectives, the present invention provides a write management method for a NAND solid-state storage device, comprising: (A) a write context establishment step: before each data write to multiple NAND blocks, a write context record is established, wherein the write context record includes a write strategy, multiple NAND block states, and a data attribute set; wherein the write strategy refers to the write mode and parameters selected by the controller, the multiple NAND block states refer to the historical write / erase counts and error conditions of each corresponding NAND block, the data attribute set refers to the data's reliability or lifespan requirements, and the write context record corresponds to the multiple NAND blocks; (B) a write consequence observation step: during a background observation period after the data write is completed, the multiple NAND blocks associated with the write context record are monitored; wherein the background observation period refers to a settable time used to collect ECC correction status, holding force errors, and unexpected rewrite or read failure events of the multiple NAND blocks; (C) a responsibility attribution analysis step: when any monitoring item exceeds the pre-defined value, the write context record is used to monitor the write context record. When setting the allowable range, the system backtracks the write context record to determine whether the combination of the specific write strategy and the multiple NAND block states is related to the failure event, and then assesses the possibility of the write strategy causing failure based on past write and observation results; (D) Write prohibition condition generation step: If the cause analysis determines that the specific write strategy and the multiple NAND block states have caused two consecutive adverse consequences in the past; or the specific write strategy and the multiple NAND block states have caused more than three intermittent adverse consequences in the past, then a corresponding write prohibition condition is established; wherein, the write prohibition condition refers to restricting or prohibiting the write strategy from being executed again under the specific NAND block state; and (E) Write governance execution step: The write operation is managed and controlled according to the established write prohibition condition. When a write request is received, the system first checks whether the identification information in the write strategy of the write request meets any existing write prohibition condition; if it does, the write is rejected and other write strategies are replaced or different multiple NAND blocks are written.

[0006] Preferably, the system further calculates the cumulative number of failures for each write strategy across all NAND blocks. If the cumulative number of failures exceeds a preset threshold, a corresponding policy prohibition condition is established to restrict or prohibit the write operation of the write strategy across all blocks. The cumulative number of failures refers to the total number of ECC corrections, holding power errors, or write failure events caused by the write combination of the strategy and any of the NAND blocks during the background observation period.

[0007] Preferably, the system further calculates the cumulative number of failures for each of the plurality of NAND blocks under all write strategies. If the cumulative number of failures exceeds a preset threshold, a corresponding block restriction condition is established to limit or reduce the write operations of the NAND block. The cumulative number of failures refers to the total number of ECC corrections, holding force errors, or write failure events caused by any combination of write strategy and the plurality of NAND blocks during the background observation period.

[0008] Preferably, the system automatically recalculates the number of combined failures of the write policy and the multiple NAND blocks, as well as the number of failure events accumulated by a single policy or a single block during the observation period in each preset observation period, and dynamically updates the policy prohibition conditions and block restriction conditions based on the calculation results, so as to ensure that the write governance operation can continuously reduce the probability of high-risk combinations and high-risk elements occurring under different usage loads and block aging stages.

[0009] In summary, the NAND solid-state storage device write governance method provided by this invention can effectively identify and control high-risk write strategies and block combinations by establishing write context records, background observation, and responsibility attribution analysis. It dynamically generates write limits to reduce ECC corrections, retention degradation, and unexpected write failures, thereby improving data reliability. Simultaneously, by finely managing data write allocation and block health status, it can slow down NAND block aging, extend the storage device's lifespan, and ensure data integrity and system stability under high-frequency writing, long-term operation, and different aging stages. Therefore, compared to existing technologies, this invention provides a finely crafted, dynamic, and self-adaptive write governance method suitable for enterprise-level and high-load storage environments, effectively addressing the shortcomings of existing technologies in high-risk operation control, block aging management, and data reliability assurance. Attached Figure Description

[0010] Figure 1 This is a flowchart of a preferred embodiment of the present invention.

[0011] Explanation of reference numerals in the attached figures: (A) to (E) - steps. Detailed Implementation

[0012] To enable those skilled in the art to clearly understand the technical content of this invention, the following embodiments are described in conjunction with the accompanying drawings to further illustrate the operation of this invention. Please refer to... Figure 1 This is a flowchart of a preferred embodiment of the present invention, wherein each step will be described sequentially in the following embodiments to illustrate the technical features and operation flow of the present invention.

[0013] This embodiment describes a write management method designed for solid-state drives (SSDs). First, a write context establishment step (A) is performed to create a complete context record before each data write, serving as the basis for subsequent operations and monitoring. That is, before each data write to multiple NAND blocks, the controller establishes a write context record. This write context record includes a write strategy, the status of multiple NAND blocks, and a set of data attributes. The write strategy refers to the write mode and related parameters selected by the controller. For example, selecting "sequential write mode" allocates data to NAND blocks sequentially according to logical addresses, or "random write mode" distributes data across different blocks to balance wear. Strategy parameters may also include write priority, data batch size, and write rate control. The status of the multiple NAND blocks refers to the historical write / erase counts and error conditions of each corresponding NAND block. For example, block A has completed 500 erase cycles and experienced 3 ECC correction events in the past week; block B has completed 200 erase cycles and experienced no ECC correction events. The controller can use this information to determine the health status of each block for subsequent write judgment or management.

[0014] Additionally, this set of data attributes refers to the data's reliability or lifespan requirements. For example, important transaction records require high reliability, while temporary cache files can tolerate low durability; or different write priorities are set according to data type to ensure that healthy blocks are used when critical data is written. Once established, this write context record is temporarily stored in the controller's cache so that it can be used for subsequent monitoring, analysis, or judgment after the data is written. This step ensures that each write operation has complete background data and can establish individual reference data for each NAND block according to different write strategies and data requirements, providing accurate basis for subsequent steps.

[0015] For example, suppose a system wants to write a 4 KB transaction to NAND blocks A, B, and C using a sequential write strategy, and the data attributes require high reliability. The controller will create a write context record, recording the sequential write strategy, the number of erases and ECC records for blocks A, B, and C, and the reliability requirements of this data. If, during background observation, this data experiences ECC correction or read failure, the system can perform a cause-of-fact analysis based on this context record.

[0016] Next, the (B) write consequence observation step is executed to monitor the multiple NAND blocks associated with the write context record during the background observation period after the data write is completed, in order to collect the potential subsequent effects of the write operation. After each data write is completed, the controller initiates a background observation program to continuously monitor all NAND blocks involved in the write. This background observation period is a settable time period used to collect information such as ECC correction status, hold power errors, and unexpected rewrite or read failure events for each NAND block. The observation period can be set to several seconds, several minutes, or extended to several hours depending on the importance of the data, to obtain sufficient monitoring data. The controller associates each observation event with the corresponding write context record. For example, if an ECC correction event occurs in block A during the observation period, the event will be recorded in the block A item under the write context record; similarly, if a read failure occurs in block B, it will also be marked in the context record. In this way, the system can fully understand the impact of each write on each NAND block and retain historical observation data for subsequent analysis.

[0017] For example, suppose the system writes 4 KB of transaction data to NAND blocks A, B, and C, with a 30-minute background observation period after the write. During this period, block A experiences one ECC correction, block C fails to read, and block B shows no anomalies. The system will mark these events in the respective block fields of the write context record for subsequent attribution analysis. This step ensures that after each write, the system can collect complete and traceable consequences information, providing reliable data support for attribution and write governance.

[0018] Following steps (A) Write Context Establishment and (B) Write Consequence Observation, step (C) Responsibility Attribution Analysis is executed to determine whether the combination of a specific write strategy and the states of the multiple NAND blocks is related to a failure event, and to assess the likelihood of the write strategy causing a failure. Specifically, when any monitored item exceeds a preset tolerance range, the controller will backtrack the corresponding write context record to analyze the write strategy used for that write and the states of the multiple NAND blocks involved. The states of the multiple NAND blocks refer to the historical write / erase counts and past error events of each block, such as ECC corrections, hold-force errors, or read failures. By comparing the historical events of the write strategy and the block states, the system can determine whether a specific strategy is significantly correlated with block anomaly events. The system will also statistically analyze the frequency of failure events caused by the same strategy on different blocks and determine whether there are cumulatively high-risk strategies or blocks, serving as the basis for subsequently establishing write prohibition conditions or block restriction conditions.

[0019] For example, suppose transaction data was written to blocks A, B, and C using a sequential write strategy. During background observation, block A was found to have undergone two ECC corrections, block B showed no anomalies, and block C experienced one read failure. The system backtracks the write context records, analyzes the combination of the write strategy and the states of each block, and determines that the sequential write strategy may be a high-risk combination on block A, while the sequential write strategy is low-risk on block B. This provides a reliable basis for establishing subsequent write prohibition conditions. Through this step, the system can clarify the possible sources of failure events, accurately attribute the cause of write failure to a specific strategy and block combination, and provide complete data to support dynamic write governance decisions.

[0020] Following steps (A) Write Context Establishment, (B) Write Consequence Observation, and (C) Responsibility Attribution Analysis, step (D) Write Prohibition Condition Generation is executed. This step establishes a corresponding write prohibition condition based on the responsibility attribution analysis results, restricting or prohibiting the re-execution of a specific write strategy under a specific NAND block state. When the system determines that a specific write strategy combined with multiple NAND block states has caused two consecutive adverse consequences in the past, or has caused more than three intermittent adverse consequences, the controller establishes a corresponding write prohibition condition. This write prohibition condition restricts or prohibits the re-execution of the write strategy under a specific NAND block state, or only allows it to operate on specific blocks or data types.

[0021] Specifically, the significance of the aforementioned design logic lies in the following: continuous adverse consequences can quickly identify short-term high-risk combinations, indicating that the strategy and recent operations on that block have continuously caused ECC corrections, hold-force errors, or read failures; intermittent cumulative adverse consequences target long-term potential risks. Even if failure events do not occur consecutively, if the cumulative total exceeds a threshold, it indicates that the combination may have long-term reliability issues. In this way, the system takes into account both short-term real-time risks and long-term cumulative risks, accurately establishing prohibition conditions, avoiding excessive restrictions on other healthy combinations, and maintaining a balance between overall write performance and SSD reliability. Furthermore, the controller stores each write prohibition condition in an internal management table and compares it when receiving a write request to execute prohibition or replacement operations in real time. For example, if the sequential write strategy has caused two consecutive ECC corrections in block A, the system will establish a condition of "prohibiting writes to block A by the sequential write strategy"; if the sequential write strategy has intermittently caused three ECC corrections in block B, the system will establish a prohibition condition for block B. In this way, the controller can prevent high-risk policies from being applied to problematic blocks again, reduce the incidence of future failure events, and provide historical data to support subsequent write governance and policy optimization.

[0022] In short, when the system determines that a specific write strategy and the combination of multiple NAND block states have caused two consecutive adverse consequences in the past, or have caused more than three consecutive adverse consequences, it indicates that the strategy and the block have continuously caused ECC correction, hold power errors, or read failures in recent operations, indicating that this combination may have a risk of real-time failure. The determination of intermittent cumulative adverse consequences manages historical cumulative risks. Even if the failure events do not occur consecutively, if the total number exceeds a preset threshold, it may indicate that the combination has potential problems in the long term, such as uneven block aging or the strategy not being adapted to the block characteristics.

[0023] Therefore, this design balances short-term real-time risks with long-term cumulative risks, avoiding judgments based solely on single or consecutive events while ignoring potentially high-risk combinations, thus ensuring more comprehensive write governance. Simultaneously, this logic allows the system to precisely establish prohibition conditions for specific policies and blocks without excessively restricting other healthy combinations, achieving a balance between reducing the incidence of high-risk events and not affecting overall write performance. For example: if a sequential write policy causes two consecutive ECC corrections in block A, the system considers this combination to be high-risk in the short term; if the sequential write policy causes three ECC corrections in block B at different times, the system determines that this combination has cumulative risk. Through this logic, the controller can establish separate write prohibition conditions for blocks A and B to prevent similar failure events from recurring in the future.

[0024] Following steps (A) Write Context Establishment, (B) Write Consequence Observation, (C) Attribution Analysis, and (D) Write Prohibition Condition Generation, step (E) Write Governance Execution is executed to manage and control write operations based on the established write prohibition conditions. This ensures that high-risk policies are prevented from affecting problematic blocks in real time when write requests are received. Upon receiving each write request, the controller first compares the identification information of the write policy with any existing write prohibition conditions. If they match, the system rejects the write operation and, based on the settings, chooses to replace the write policy or rewrite to other healthy NAND blocks. This prevents ECC correction, hold-up errors, or read failures from occurring again due to a combination of specific policies and high-risk blocks, improving SSD reliability and data security.

[0025] For example, suppose a write request is received with a "sequential write" strategy and the target block is A. If the previous step (D) established the condition that "the sequential write strategy prohibits writing to block A," the controller will reject the write and automatically redirect the data to block B or execute it using a "distributed write strategy." Conversely, if the strategy and the target block do not violate any prohibition conditions, the system will execute the write normally. Furthermore, the controller continuously tracks the results of all write operations and updates new observation events to the write context record, providing complete data support for subsequent attribution analysis and the generation of write prohibition conditions. This step ensures that write governance operations are executed dynamically, continuously, and automatically, reducing the likelihood of high-risk write combinations and block failure events, while maintaining overall write performance and data security.

[0026] Furthermore, in the aforementioned write governance method, the system further executes a policy-level cumulative failure assessment mechanism to determine whether a single write strategy itself constitutes an overall high-risk source. Specifically, the controller calculates the cumulative number of failures for each write strategy across all the plurality of NAND blocks and compares this cumulative number of failures with a preset threshold. The cumulative number of failures refers to the total number of ECC corrections, hold-power errors, or write failure events caused by the write strategy in each of the plurality of NAND blocks during their respective background observation periods. That is, the assessment is not limited to a single block or a specific combination, but rather statistically analyzes the overall failure situation caused by the write strategy under different NAND block states.

[0027] When the system determines that the cumulative number of failures exceeds a preset threshold, it indicates that the write strategy continues to cause failure events even when applied to different blocks, suggesting that the strategy itself may not be suitable for the current operating environment or aging state of the storage device. In this case, the controller establishes a corresponding policy prohibition condition to restrict or prohibit write operations of the write strategy on all NAND blocks. For example, if a high-speed write strategy causes ECC correction, hold-force errors, or write failures on multiple different NAND blocks, and the cumulative number exceeds a preset threshold, even if the health states of the multiple blocks are different, the system will still determine that the strategy has a systemic risk and establish a policy prohibition condition, preventing subsequent write requests from using this write strategy and instead using other lower-risk write strategies. Through this auxiliary design, the system can further elevate the risk assessment from "strategy and block combination" to the risk governance level of "single write strategy itself," ensuring that write governance is not limited to specific blocks but comprehensively avoids high-risk write strategies repeatedly causing failure events throughout the entire storage device.

[0028] Furthermore, in the aforementioned write governance method, the system further executes a block-level cumulative failure assessment mechanism to determine whether a single NAND block becomes a high-risk write target due to aging or physical degradation. Specifically, the controller calculates the cumulative failure count for each of the multiple NAND blocks under all write strategies and compares this cumulative failure count with a preset threshold. The cumulative failure count refers to the total number of ECC corrections, hold-up errors, or write failure events caused by any write strategy combined with the multiple NAND blocks during their respective background observation periods. That is, this assessment focuses on a single NAND block, statistically analyzing the overall failure situation of that block under different write strategies, rather than judging only a specific strategy or a single event. When the system determines that the cumulative failure count exceeds the preset threshold, it indicates that the NAND block repeatedly experiences failure events even with different write strategies, suggesting that the block may have entered a high-risk or late-life state. At this point, the controller establishes corresponding block constraints to limit or reduce write operations on the NAND block and prevent it from continuing to handle high-reliability or long-lifetime data writes.

[0029] For example, if a NAND block experiences ECC correction, hold-up errors, or write failures under various write strategies, and the cumulative number of these events exceeds a preset threshold, the system determines that the block has a block-level failure risk and establishes block restrictions. This prevents subsequent write requests from prioritizing or allowing writes to that NAND block, instead assigning write operations to other healthy blocks. Through this auxiliary design, the system elevates risk assessment from a "strategy and block combination" level to a "single NAND block itself" level. This allows write governance to reflect block aging status in real time, preventing high-risk blocks from continuously bearing write loads, thereby improving the overall reliability and lifespan of the storage device.

[0030] In the aforementioned write governance method, the system further automatically recalculates the combined failure count of the write strategy and the multiple NAND blocks, as well as the cumulative failure event count of a single write strategy or a single NAND block during each preset observation period, and dynamically updates the policy prohibition conditions and block restriction conditions based on the calculation results. The preset observation period refers to a configurable time segment or write volume segment, such as a fixed time length or cumulative write data volume, to reflect the actual operating status of the storage device at different operating stages. At the end of each observation period, the controller re-counts the failure events of each write strategy and each NAND block, including ECC correction, hold-force errors, or write failure events, and updates their corresponding cumulative failure counts. Through this recalculation mechanism, the system can determine whether a write strategy or NAND block previously considered high-risk continues to generate failure events, or whether the risk has decreased due to changes in usage load, data characteristics, or aging states; it can also identify in real time whether previously unrestricted strategies or blocks are gradually becoming high-risk elements. The controller dynamically adjusts or updates existing policy prohibition conditions and block restriction conditions based on the above judgment results, rather than maintaining a fixed restriction state.

[0031] For example, if a write policy was prohibited due to frequent failure events in a previous observation period, but the failure events significantly decreased or disappeared in subsequent observation periods, the system can reduce the restriction level or remove the prohibition condition. Conversely, if a NAND block rapidly accumulates failure events in a recent observation period, even if it was not previously restricted, the system can establish new block restriction conditions in real time. Through the above mechanism, the write governance operation of this invention can be continuously adjusted according to different usage loads, data behaviors, and NAND block aging stages, avoiding the long-term presence of high-risk combinations and high-risk elements in the write process, thereby effectively reducing the probability of failure events and improving the overall reliability and lifespan of the storage device.

[0032] In summary, the NAND solid-state storage device write governance method provided by this invention can effectively identify and control high-risk write strategies and block combinations by establishing write context records, background observation, and responsibility attribution analysis. It dynamically generates write limits to reduce ECC corrections, retention degradation, and unexpected write failures, thereby improving data reliability. Simultaneously, by finely managing data write allocation and block health status, it can slow down NAND block aging, extend the storage device's lifespan, and ensure data integrity and system stability under high-frequency writing, long-term operation, and different aging stages. Therefore, compared to existing technologies, this invention provides a finely crafted, dynamic, and self-adaptive write governance method suitable for enterprise-level and high-load storage environments, effectively addressing the shortcomings of existing technologies in high-risk operation control, block aging management, and data reliability assurance.

[0033] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Therefore, any equivalent changes and modifications made without departing from the scope of the present invention should be covered within the patent scope of the present invention.

Claims

1. A write management method for a NAND-type solid-state storage device, characterized in that, Include: (A) Write the context creation steps: Before each data write to multiple NAND blocks, a write context record is established. This write context record includes a write strategy, multiple NAND block states, and a set of data attributes. The write strategy refers to the write mode and parameters selected by the controller. The multiple NAND block states refer to the historical write / erase counts and error conditions of each corresponding NAND block. The set of data attributes refers to the data's reliability or lifespan requirements. The write context record corresponds to the multiple NAND blocks. (B) Write the consequences observation steps: During a background observation period after data writing is completed, the multiple NAND blocks associated with the write context record are monitored; wherein, the background observation period refers to a settable time period used to collect ECC correction status, hold force errors and unexpected rewrite or read failure events of the multiple NAND blocks; (C) Steps for responsibility attribution analysis: When any monitored item exceeds the preset allowable range, the system backtracks the write context record to determine whether the combination of the specific write strategy and the multiple NAND block states is related to the failure event, and then assesses the possibility of the write strategy causing failure based on past write and observation results. (D) Write the prohibition condition generation step: If the cause-of-fact analysis determines that a specific write strategy combined with multiple NAND block states has caused two consecutive adverse consequences in the past; or that a specific write strategy combined with multiple NAND block states has caused more than three intermittent adverse consequences in the past, then a corresponding write prohibition condition is established; wherein, the write prohibition condition refers to restricting or prohibiting the write strategy from being executed again under a specific NAND block state; and (E) Write governance execution steps: Based on the established write prohibition conditions, the system manages and controls write operations. When a write request is received, the system first checks whether the identification information in the write policy of the write request meets any existing write prohibition conditions. If it does, the write is rejected and other write policies are replaced or different NAND blocks are written.

2. The write governance method as described in claim 1, characterized in that, The system also calculates the cumulative number of failures for each write strategy across all NAND blocks. If the cumulative number of failures exceeds a preset threshold, a corresponding policy prohibition condition is established to restrict or prohibit the write operation of the write strategy across all blocks. The cumulative number of failures refers to the total number of ECC corrections, holding power errors, or write failure events caused by the write combination of the strategy and any of the NAND blocks during the background observation period.

3. The write governance method as described in claim 2, characterized in that, The system also calculates the cumulative number of failures for each of the multiple NAND blocks under all write strategies. If the cumulative number of failures exceeds a preset threshold, a corresponding block restriction condition is established to limit or reduce the write operations of the NAND block. The cumulative number of failures refers to the total number of ECC corrections, holding force errors, or write failure events caused by any combination of write strategy and the multiple NAND blocks during the background observation period.

4. The write governance method as described in claim 3, characterized in that, The system automatically recalculates the number of combined failures of the write policy and the multiple NAND blocks, as well as the cumulative number of failure events of a single policy or a single block during the observation period in each preset observation period. Based on the calculation results, the system dynamically updates the prohibition conditions of the policy and the restriction conditions of the block to ensure that the write governance operation can continuously reduce the probability of high-risk combinations and high-risk elements under different usage loads and block aging stages.

Images