A method for dynamically adjusting reserved space of solid-state hard disk

Abstract

The present invention discloses a method for dynamically adjusting the reserved space of a solid-state drive (SSD), belonging to the field of dynamically adjusting the reserved space of a SSD. The method comprises the following steps: detecting the NAND capacity of the SSD and calculating the actual capacity, while obtaining the health indicator data of the SSD; obtaining the original OP space threshold, and adjusting the original OP threshold according to the health status of the SSD; evaluating the space requirements of the current operation, combining the load type with the health status, formulating different OP thresholds, and formulating a strategy for increasing OP space; dynamically adjusting the OP space according to the current load type and the set OP space threshold; after a single operation is completed, using a sliding window algorithm to calculate the load trend, and obtaining a recovery method for the original OP based on the load trend. By combining the load and health status to adjust the OP space threshold and using the sliding window algorithm to formulate a recovery strategy, the present invention can effectively improve the comprehensive performance and efficiency of the SSD.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamically adjusting solid-state hard disk reserved space, and in particular to a method for dynamically adjusting solid-state hard disk reserved space. Background Art

[0002] During the use of solid-state drives (SSDs), the proper configuration of over-provisioning (OP) plays a crucial role in their performance, lifespan, and data security. Traditional SSDs typically use a fixed threshold for OP space, which lacks flexibility based on actual load types and drive health. Fixed thresholds struggle to balance performance and space utilization in diverse load scenarios. Under high loads, insufficient OP space can lead to read and write performance bottlenecks, while under low loads, storage space is wasted. Furthermore, as SSDs age, flash memory chips gradually age, causing changes in health indicators such as spare block percentage, TBW consumption, and write amplification. However, fixed OP space settings cannot promptly respond to these changes in drive health, effectively reducing flash memory chip wear and the risk of data loss. Furthermore, existing OP space management models based on fixed thresholds lack dynamic analysis and response to load data, making it difficult to strike a balance between performance assurance and capacity utilization. Therefore, a method that dynamically adjusts OP space based on load types and drive health is urgently needed to improve the overall performance and lifespan of SSDs.

[0003] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0004] In response to the problems in the related art, the present invention proposes a method for dynamically adjusting the reserved space of a solid-state hard disk to overcome the above-mentioned technical problems existing in the existing method for adjusting the reserved space of a solid-state hard disk.

[0005] To this end, the specific technical solutions adopted in the present invention are as follows:

[0006] A method for dynamically adjusting reserved space of a solid-state drive comprises the following steps:

[0007] S1. Detect the NAND capacity of the solid-state drive, calculate the actual capacity, and obtain the health indicator data of the hard drive;

[0008] S2. Obtain the original OP space threshold and adjust the original OP threshold according to the hard disk health status;

[0009] S3. Evaluate the space requirements of current operations, combine load types with health status, develop different OP thresholds, and develop strategies to increase OP space;

[0010] S4. Dynamically adjust the OP space according to the current load type and the set OP space threshold;

[0011] S5. After a single operation is completed, the load trend is calculated using a sliding window algorithm, and the recovery method of the original OP is obtained based on the load trend.

[0012] As a preferred embodiment, the method of detecting the NAND capacity of the solid-state drive, calculating the actual capacity, and obtaining the health indicator data of the hard drive includes the following steps:

[0013] S11. Use the detection tool to check the physical device capacity of the SSD and calculate the actual capacity of the SSD;

[0014] When calculating the actual capacity of a solid-state drive, you need to convert the physical device capacity obtained and deduct the system usage to obtain the actual capacity of the solid-state drive.

[0015] S12. Use the detection tool to view SMART parameters, including remaining life, spare blocks, and total write volume.

[0016] As a preferred embodiment, obtaining the original OP space threshold and adjusting the original OP threshold according to the hard disk health status includes the following steps:

[0017] S21, obtaining the original OP space threshold of the solid state drive;

[0018] S22. Calculate TBW consumption and write amplification ratio based on the obtained SMART parameters of the solid-state drive;

[0019] S23. Health indicators include spare blocks, TBW consumption, and write amplification. The normal range of health indicators is preset. For every 5% decrease in spare block percentage, the OP space increases by 3%. For every 10% increase in TBW consumption, the OP space increases by 2 percentage points. For every 0.5 increase in write amplification, the OP space increases by 1%.

[0020] The spare block indicator is the ratio of remaining spare blocks to total spare blocks in the SSD. For every 5% decrease in the spare block percentage, the OP space increases by 3%. This means that for every 5% decrease in the ratio of remaining spare blocks to total spare blocks in the SSD, the OP space increases by 3%.

[0021] As a preferred embodiment, the process of evaluating the space requirements of current operations, combining load types with health status, formulating different OP thresholds, and formulating strategies for increasing OP space includes the following steps:

[0022] S31. Evaluate the current load type and classify it into ultra-low load, sequential write-intensive, mixed load, and random write-intensive. For ultra-low load, there is no need to adjust the OP space threshold. For sequential write-intensive, adjust the original OP space threshold to 10%-15%. For mixed load, adjust the original OP space threshold to 15%-20%. For random write-intensive, adjust the original OP space threshold to 20%-25%.

[0023] S32. Combining the load status with the health status, further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive.

[0024] As a preferred embodiment, the method of combining the load status with the health status to further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive traffic includes the following steps:

[0025] S321. When the indicators of the health status, namely, spare blocks, TBW consumption, and write amplification, are within the preset normal range, no adjustment is required. When the indicators of the health status exceed the preset normal range, the load status is combined with the health status to further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive workloads. The calculation formula is:

[0026] ;

[0027] in, is the adjusted over-provisioning threshold. They are the percentage of spare block decrease, TBW increase, and write amplification increase. are the weights of the percentage of spare block decrease, the percentage of TBW increase, and the write amplification rate increase, respectively. When the load type is sequential write-intensive, The value of is 30%, 40%, 30%. When the load type is mixed load, The values ​​of are 35%, 30%, and 35%. When the load type is random write intensive, The values ​​of are 40%, 30%, and 30%. is the total weight, is the load sensitivity coefficient. When the load type is sequential write intensive, The value is 1.0. When the load type is mixed load, The value is 1.1. When the load type is random write intensive, The value is 1.2.

[0028] As a preferred embodiment, the dynamically adjusting the OP space according to the current load type and the set OP space threshold comprises the following steps:

[0029] S41. Use system monitoring tools to determine the current load type;

[0030] S42. Input the final OP space threshold into the storage management interface so that the system automatically reallocates space and controls the upper limit of the OP space threshold for sequential write-intensive workloads to 25%, the upper limit of the OP space threshold for mixed loads to 30%, and the upper limit of the OP space threshold for random write-intensive workloads to 35%.

[0031] As a preferred embodiment, after the single operation is completed, a sliding window algorithm is used to calculate the load trend, and a recovery method for obtaining the original OP according to the load trend includes the following steps:

[0032] S51. Obtain load data for the past seven days, including the average daily write volume and peak IOPS, and use a sliding window algorithm to calculate the load trend. If the recent load continues to decrease, the OP space is quickly reduced. If the load is stable, the OP space is slowly reduced. If the load fluctuates frequently, the current OP threshold is maintained.

[0033] As a preferred embodiment, the load data of the past 7 days is obtained, including the average daily write volume and peak IOPS, and the load trend is calculated using a sliding window algorithm. If the recent load continues to decrease, the OP space is quickly reduced; if the load is stable, the OP space is slowly reduced; if the load fluctuates frequently, the current OP threshold is maintained, including the following steps:

[0034] S511. The window is fixed at 7 days and updated daily. The moving average of the average daily write volume and the standard deviation of the peak IOPS in the window are calculated. The standard deviation of the peak IOPS in the window is used as the IOPS volatility. If the moving average of the write volume decreases by more than 15% for three consecutive days and the peak IOPS volatility is within 10%, it is judged that the recent load has been continuously decreasing. The OP space is quickly reduced by 3% to 5% each time, with a single interval of 2 hours. If the moving average fluctuation of the average daily write volume is within 5% and the peak IOPS volatility is within 15%, it is judged that the recent load is stable. The OP space is slowly reduced by 1% to 2% each time, with a single interval of 1 to 2 days. If the moving average of the average daily write volume exceeds 10% or the peak IOPS volatility exceeds 20%, it is judged that the recent load is fluctuating frequently, and the current OP threshold is maintained.

[0035] The beneficial effects of the present invention are:

[0036] 1. This invention dynamically adjusts the OP space threshold by combining load type and health status. This allows for precise adaptation and intelligent optimization of OP space: differentiated OP threshold ranges are pre-set for different load types, prioritizing reserved space in high-load scenarios to avoid read and write performance bottlenecks caused by insufficient space. When health indicators such as spare block percentage, TBW consumption, and write amplification exceed normal ranges, the OP threshold is dynamically adjusted through a weighted formula that integrates the load sensitivity coefficient and indicator weight to expand space in advance to cope with hardware aging, reduce flash memory particle wear, and mitigate the risk of data loss. This mechanism takes into account both load characteristics and health status, ensuring performance under high load and proactively protecting against health degradation, thereby improving the read and write efficiency, stability, and service life of the solid-state drive.

[0037] 2. This invention collects load data from the past seven days and formulates an OP space recovery strategy based on a sliding window algorithm. This allows for intelligent dynamic management of reserved space. By analyzing the load data from the past seven days, OP space can be released on demand, avoiding space waste or the risk of blind adjustment under fixed threshold modes. This allows for a dynamic balance between performance assurance and capacity utilization, enhancing the SSD's adaptability in dynamic load environments and improving storage resource utilization efficiency.

[0038] At the same time, this strategy of recovering OP space in batches can not only avoid performance fluctuations caused by one-time large-scale adjustments, but also accurately reclaim idle space while maintaining threshold stability during load fluctuations to reduce interference, thereby achieving a dynamic balance between performance stability, space utilization, and hardware protection, and improving the adaptability and long-term operating efficiency of solid-state drives in different business scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 The present invention is a flowchart of a method for dynamically adjusting reserved space of a solid-state drive. DETAILED DESCRIPTION

[0041] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0042] According to an embodiment of the present invention, a method for dynamically adjusting reserved space of a solid-state drive is provided.

[0043] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, a method for dynamically adjusting the reserved space of a solid-state drive according to an embodiment of the present invention includes the following steps:

[0044] S1. Detect the NAND capacity of the solid-state drive, calculate the actual capacity, and obtain the health indicator data of the hard drive;

[0045] Furthermore, detecting the NAND capacity of the solid-state drive, calculating the actual capacity, and obtaining the health indicator data of the hard drive include the following steps:

[0046] S11. Use the detection tool to check the physical device capacity of the SSD and calculate the actual capacity of the SSD;

[0047] S12. Use the detection tool to view SMART parameters, including remaining life, spare blocks, and total write volume.

[0048] S2. Obtain the original OP space threshold and adjust the original OP threshold according to the hard disk health status;

[0049] Furthermore, obtaining the original OP space threshold and adjusting the original OP threshold according to the hard disk health status include the following steps:

[0050] S21, obtaining the original OP space threshold of the solid state drive;

[0051] S22. Calculate TBW consumption and write amplification ratio based on the obtained SMART parameters of the solid-state drive;

[0052] S23. Health indicators include spare blocks, TBW consumption, and write amplification. The normal range of health indicators is preset. For every 5% decrease in spare block percentage, the OP space increases by 3%. For every 10% increase in TBW consumption, the OP space increases by 2 percentage points. For every 0.5 increase in write amplification, the OP space increases by 1%.

[0053] It should be noted that adjusting the original OP space threshold according to the health status of the hard disk can significantly improve the performance and life of the solid-state drive. By calculating the TBW consumption and write amplification ratio through SMART parameters, and dynamically adjusting the OP space based on key indicators such as the percentage of spare blocks, the reserved space can be expanded in advance when the spare blocks are reduced, effectively avoiding the degradation of read and write performance due to insufficient available blocks; monitoring and adjusting TBW consumption and write amplification ratio can reasonably allocate OP space, reduce data writing pressure, and reduce flash memory particle wear, thereby optimizing the overall operational stability and reliability of the solid-state drive, delaying hardware aging, and ensuring the long-term and efficient operation of storage devices.

[0054] S3. Evaluate the space requirements of current operations, combine load types with health status, develop different OP thresholds, and develop strategies to increase OP space;

[0055] Furthermore, evaluating the space requirements of current operations, combining load types with health status, setting different OP thresholds, and formulating strategies for increasing OP space include the following steps:

[0056] S31. Evaluate the current load type and classify it into ultra-low load, sequential write-intensive, mixed load, and random write-intensive. For ultra-low load, there is no need to adjust the OP space threshold. For sequential write-intensive, adjust the original OP space threshold to 10%-15%. For mixed load, adjust the original OP space threshold to 15%-20%. For random write-intensive, adjust the original OP space threshold to 20%-25%.

[0057] It should be noted that ultra-low load refers to long-term idleness or extremely low writes. At this time, there is no need to adjust the OP space threshold. Sequential write-intensive refers to continuous large file writing, random write-intensive refers to high-frequency small file writing, and mixed load refers to alternating read and write.

[0058] S32: Combine the load status with the health status to further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive workloads.

[0059] Furthermore, combining the load status with the health status to further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive workloads includes the following steps:

[0060] S321. When the indicators of the health status, namely, spare blocks, TBW consumption, and write amplification, are within the preset normal range, no adjustment is required. When the indicators of the health status exceed the preset normal range, the load status is combined with the health status to further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive workloads. The calculation formula is:

[0061] ;

[0062] in, is the adjusted over-provisioning threshold. They are the percentage of spare block decrease, TBW increase, and write amplification increase. are the weights of the percentage of spare block decrease, the percentage of TBW increase, and the write amplification rate increase, respectively. When the load type is sequential write-intensive, The value of is 30%, 40%, 30%. When the load type is mixed load, The values ​​of are 35%, 30%, and 35%. When the load type is random write intensive, The values ​​of are 40%, 30%, and 30%. is the total weight, is the load sensitivity coefficient. When the load type is sequential write intensive, The value is 1.0. When the load type is mixed load, The value is 1.1. When the load type is random write intensive, The value is 1.2.

[0063] It should be noted that combining the load status with the health status and adjusting the OP space thresholds of different types can accurately adapt to the operating needs of the solid-state drive. According to the characteristics of sequential write-intensive, mixed load, and random write-intensive, combined with the current health status of the hard drive, dynamic adjustment can reasonably allocate reserved space under high load to avoid performance bottlenecks; when the health status is poor, the OP space can be expanded in time to reduce wear and tear, prevent data loss, and effectively improve the read and write efficiency, stability and service life of the solid-state drive, and give full play to its storage performance.

[0064] S4. Dynamically adjust the OP space according to the current load type and the set OP space threshold;

[0065] Furthermore, dynamically adjusting the OP space according to the current load type and the set OP space threshold includes the following steps:

[0066] S41. Use system monitoring tools to determine the current load type;

[0067] S42. Input the final OP space threshold into the storage management interface so that the system automatically reallocates space and controls the upper limit of the OP space threshold for sequential write-intensive workloads to 25%, the upper limit of the OP space threshold for mixed loads to 30%, and the upper limit of the OP space threshold for random write-intensive workloads to 35%.

[0068] S5. After a single operation is completed, the load trend is calculated using a sliding window algorithm, and the recovery method of the original OP is obtained based on the load trend.

[0069] Furthermore, after a single operation is completed, a sliding window algorithm is used to calculate the load trend, and a recovery method for obtaining the original OP based on the load trend includes the following steps:

[0070] S51. Obtain the load data for the past 7 days, including the average daily write volume and peak IOPS, and use the sliding window algorithm to calculate the load trend. If the recent load continues to decline, the OP space will be quickly reduced. If the load is stable, the OP space will be slowly reduced. If the load fluctuates frequently, the current OP threshold will be maintained.

[0071] Furthermore, load data for the past seven days is obtained, including the average daily write volume and peak IOPS. A sliding window algorithm is used to calculate the load trend. If the recent load continues to decrease, the OP space is quickly reduced. If the load is stable, the OP space is slowly reduced. If the load fluctuates frequently, the current OP threshold is maintained. The following steps are included:

[0072] S511. The window is fixed at 7 days and updated daily. The moving average of the average daily write volume and the standard deviation of the peak IOPS in the window are calculated. The standard deviation of the peak IOPS in the window is used as the IOPS volatility. If the moving average of the write volume decreases by more than 15% for three consecutive days and the peak IOPS volatility is within 10%, it is judged that the recent load has been continuously decreasing. The OP space is quickly reduced by 3% to 5% each time, with a single interval of 2 hours. If the moving average fluctuation of the average daily write volume is within 5% and the peak IOPS volatility is within 15%, it is judged that the recent load is stable. The OP space is slowly reduced by 1% to 2% each time, with a single interval of 1 to 2 days. If the moving average of the average daily write volume exceeds 10% or the peak IOPS volatility exceeds 20%, it is judged that the recent load is fluctuating frequently, and the current OP threshold is maintained.

[0073] In summary, the present invention dynamically adjusts the OP space threshold by combining the load type and the health status, and dynamically adjusts the OP space threshold by combining the load type and the health status of the hard disk, so as to achieve precise adaptation and intelligent optimization of the OP space: for different load types, differentiated OP threshold ranges are pre-set, and reserved space is prioritized in high-load scenarios to avoid read and write performance bottlenecks caused by insufficient space; when health indicators such as spare block percentage, TBW consumption, and write amplification rate exceed the normal range, the OP threshold is dynamically adjusted by a weighted formula that integrates the load sensitivity coefficient and the indicator weight to expand the space in advance to cope with hardware aging, reduce flash memory particle wear, and reduce the risk of data loss. This mechanism takes into account both load characteristics and health status, and can not only guarantee performance under high load, but also take active protection when the health status declines, thereby improving the read and write efficiency and stability of the solid-state hard disk. and service life; the present invention collects load data from the past 7 days and formulates an OP space recovery strategy based on a sliding window algorithm to achieve intelligent dynamic management of reserved space. By analyzing the load data from the past 7 days, the OP space can be released on demand, avoiding space waste or the risk of blind adjustment under a fixed threshold mode, achieving a dynamic balance between performance guarantee and capacity utilization, enhancing the adaptability of the solid-state drive in a dynamic load environment, and improving the efficiency of storage resource utilization; at the same time, this strategy of recovering OP space in batches can not only avoid performance fluctuations caused by a one-time large-scale adjustment, but also accurately reclaim idle space while maintaining a stable threshold during load fluctuations to reduce interference, thereby achieving a dynamic balance between performance stability, space utilization, and hardware protection, and improving the adaptability and long-term operation efficiency of the solid-state drive in different business scenarios.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for dynamically adjusting reserved space of a solid-state drive, characterized in that: The method comprises the following steps: S1. Detect the NAND capacity of the solid-state drive, calculate the actual capacity, and obtain the health indicator data of the hard drive; S2. Obtain the original OP space threshold and adjust the original OP threshold according to the hard disk health status; S3. Evaluate the space requirements of current operations, combine load types with health status, develop different OP thresholds, and develop strategies to increase OP space; S31. Evaluate the current load type and classify it into ultra-low load, sequential write-intensive, mixed load, and random write-intensive. For ultra-low load, there is no need to adjust the OP space threshold. For sequential write-intensive, adjust the original OP space threshold to 10%-15%. For mixed load, adjust the original OP space threshold to 15%-20%. For random write-intensive, adjust the original OP space threshold to 20%-25%. S32. Combining the load status with the health status, further adjusting the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive workloads; S321. When the indicators of the health status, namely, spare blocks, TBW consumption, and write amplification, are within the preset normal range, no adjustment is required. When the indicators of the health status exceed the preset normal range, the load status is combined with the health status to further adjust the OP space thresholds for sequential write-intensive, mixed load, and random write-intensive workloads. The calculation formula is: ； in, is the adjusted over-provisioning threshold. They are the percentage of spare block decrease, TBW increase, and write amplification increase. are the weights of the percentage of spare block decrease, the percentage of TBW increase, and the write amplification rate increase, respectively. When the load type is sequential write-intensive, The value of is 30%, 40%, 30%. When the load type is mixed load, The values ​​of are 35%, 30%, and 35%. When the load type is random write intensive, The values ​​of are 40%, 30%, and 30%. is the total weight, is the load sensitivity coefficient. When the load type is sequential write intensive, The value is 1.

0. When the load type is mixed load, The value is 1.

1. When the load type is random write intensive, The value is 1.2; S4. Dynamically adjust the OP space according to the current load type and the set OP space threshold; S5. After a single operation is completed, the load trend is calculated using a sliding window algorithm, and the recovery method of the original OP is obtained based on the load trend.

2. The method for dynamically adjusting the reserved space of a solid-state drive according to claim 1, wherein: The method of detecting the NAND capacity of the solid-state drive, calculating the actual capacity, and obtaining the health indicator data of the hard drive includes the following steps: S11. Use the detection tool to check the physical device capacity of the SSD and calculate the actual capacity of the SSD; S12. Use the detection tool to view SMART parameters, including remaining life, spare blocks, and total write volume.

3. The method for dynamically adjusting the reserved space of a solid-state drive according to claim 1, wherein: The obtaining of the original OP space threshold and adjusting the original OP threshold according to the hard disk health status includes the following steps: S21, obtaining the original OP space threshold of the solid state drive; S22. Calculate TBW consumption and write amplification ratio based on the obtained SMART parameters of the solid-state drive; S23. Health indicators include spare blocks, TBW consumption, and write amplification. The normal range of health indicators is preset. For every 5% decrease in spare block percentage, the OP space increases by 3%. For every 10% increase in TBW consumption, the OP space increases by 2 percentage points. For every 0.5 increase in write amplification, the OP space increases by 1%.

4. The method for dynamically adjusting the reserved space of a solid-state drive according to claim 1, wherein: The method of dynamically adjusting the OP space according to the current load type and the set OP space threshold comprises the following steps: S41. Use system monitoring tools to determine the current load type; S42. Input the final OP space threshold into the storage management interface so that the system automatically reallocates space and controls the upper limit of the OP space threshold for sequential write-intensive workloads to 25%, the upper limit of the OP space threshold for mixed loads to 30%, and the upper limit of the OP space threshold for random write-intensive workloads to 35%.

5. The method for dynamically adjusting the reserved space of a solid-state drive according to claim 1, wherein: After the single operation is completed, the load trend is calculated using a sliding window algorithm, and the recovery method of the original OP is obtained according to the load trend, including the following steps: S51. Obtain load data for the past seven days, including the average daily write volume and peak IOPS, and use a sliding window algorithm to calculate the load trend. If the recent load continues to decrease, the OP space is quickly reduced. If the load is stable, the OP space is slowly reduced. If the load fluctuates frequently, the current OP threshold is maintained.

6. The method for dynamically adjusting the reserved space of a solid-state drive according to claim 5, characterized in that: Obtaining load data for the past seven days, including the average daily write volume and peak IOPS, and using a sliding window algorithm to calculate the load trend, if the recent load continues to decrease, then quickly reduce the OP space; if the load is stable, then slowly reduce the OP space; if the load fluctuates frequently, then maintain the current OP threshold. The following steps are included: S511. The window is fixed at 7 days and updated daily. The moving average of the average daily write volume and the standard deviation of the peak IOPS in the window are calculated. The standard deviation of the peak IOPS in the window is used as the IOPS volatility. If the moving average of the write volume decreases by more than 15% for three consecutive days and the peak IOPS volatility is within 10%, it is judged that the recent load has been continuously decreasing. The OP space is quickly reduced by 3% to 5% each time, with a single interval of 2 hours. If the moving average fluctuation of the average daily write volume is within 5% and the peak IOPS volatility is within 15%, it is judged that the recent load is stable. The OP space is slowly reduced by 1% to 2% each time, with a single interval of 1 to 2 days. If the moving average of the average daily write volume exceeds 10% or the peak IOPS volatility exceeds 20%, it is judged that the recent load is fluctuating frequently, and the current OP threshold is maintained.

Images