Solid state disk performance improvement method with reserved space dynamic adjustment function

Abstract

A solid state disk performance improvement method with a reserved space dynamic adjustment function comprises the following steps: continuously monitoring a plurality of operation parameters of a solid state disk in a computer system, including a write amplification factor, a Trim command trigger frequency, a data write-in concentration ratio and an average I / O delay value, and carrying out weighted analysis to generate a demand index; and deducing a suggested upper limit value U, a suggested lower limit value L and a suggested intermediate value M of the reserved space as adjustment bases according to the demand indexes. The system formulates an adjustment strategy according to the comparison relationship between the residual capacity and the reserved space and automatically adjusts the residual capacity gt; when U is equal to U, the reserved space is equal to U; u > = residual capacity gt; m, the reserved space changes along with the residual capacity according to a first ratio which is obtained by dividing the variation of the reserved space by the variation of the residual capacity and is equal to 1; mgt; the reserved space is greater than or equal to L, the reserved space is changed along with the residual capacity according to a second ratio obtained by dividing the variation of the reserved space by the variation of the residual capacity to be less than 1, and the reserved space is just reduced to L when the residual capacity is used up, so that reserved resources are dynamically allocated under different use situations.

Description

TECHNICAL FIELD

[0001] The present application relates to a solid state disk and its storage management technology, in particular to a solid state disk performance improvement method with a reserved space dynamic adjustment function, which automatically adjusts the reserved space according to operation parameters and remaining capacity to improve the write performance, resource utilization efficiency and stability of the solid state disk. BACKGROUND

[0002] The existing solid state disk (SSD) usually sets a fixed reserved space (Over-Provisioning, OP) when it is shipped, so as to improve the write amplification effect, prolong the service life of the NAND flash memory, and maintain stable read and write performance. The size of the fixed reserved space is usually configured according to the preset proportion of the manufacturer, and is irrelevant to the actual demand of the user, so that the performance may be insufficient or the capacity may be wasted in different use scenarios. Moreover, although increasing the reserved space can effectively reduce the write amplification effect and prolong the service life, for the user, too large reserved space will directly reduce the available capacity, causing a psychological gap that "buying a 1TB solid state disk, the actual available space is greatly reduced", and then affecting the user's feeling of reasonable allocation of resources. Therefore, in the design of the SSD, not only the performance and the service life should be considered, but also the user's demand and psychological acceptance of the available capacity should be considered to avoid excessive occupation of resources.

[0003] In other words, the existing method cannot simultaneously consider performance, service life and user's available resource feeling for the setting of the reserved space. Therefore, how to automatically and dynamically adjust the reserved space according to multiple operation parameters and remaining capacity during the operation of the SSD, so as to balance the write performance, NAND service life and user's available capacity psychological feeling, has become an important technical issue to improve the overall performance of the SSD and the user experience.

[0004] Therefore, the present application team proposes a solid state disk performance improvement method with a reserved space dynamic adjustment function, which adjusts the reserved space according to multiple operation parameters and remaining capacity, so as to balance the write performance and NAND service life, and also consider the user's psychological feeling of available capacity, and balance the performance, service life and reasonable allocation of resources in different use situations, greatly improving the deficiencies of the existing technology. SUMMARY

[0005] One of the purposes of the present application is to provide a solid state disk performance improvement method with a reserved space dynamic adjustment function, which can automatically and dynamically adjust the reserved space according to multiple operation parameters and remaining capacity of the solid state disk, so as to balance the write performance, NAND service life and user's available capacity psychological feeling, and overcome the problems of insufficient performance, accelerated service life consumption and resource waste caused by fixed or simple dynamic reserved space in the prior art.

[0006] To achieve the above object, the present application provides a solid state disk performance improvement method with reserved space dynamic adjustment function, wherein: a plurality of operating parameters of a solid state disk in a computer system are continuously monitored and weighted analyzed to generate a demand index, the plurality of operating parameters at least including a write amplification factor, a Trim command triggering frequency, a data write concentration, and an average I / O delay value, and according to the demand index, a recommended upper limit value U and a recommended lower limit value L of the reserved space are derived, a middle value M between the recommended upper limit value U and the recommended lower limit value L, and the computer system compares a remaining capacity of the solid state disk with the reserved space to adjust according to an adjustment strategy; wherein, when the remaining capacity of the solid state disk is greater than U during use, the reserved space is maintained at the recommended upper limit value; when the remaining capacity of the solid state disk is less than or equal to U and greater than M, the reserved space changes with the remaining capacity at a first change rate of 1 relative to the remaining capacity; when the remaining capacity of the solid state disk is less than or equal to M and greater than or equal to L, the reserved space changes with the remaining capacity at a second change rate less than 1 relative to the remaining capacity, and when the remaining capacity is about to be used up, the reserved space is just reduced to the recommended lower limit value, and when the remaining space is released in reverse, the first change rate and the second change rate are also used for dynamic adjustment of the reserved space.

[0007] Preferably, when the solid state disk detects that the I / O load is low or enters an idle mode, the reserved space ratio is adjusted.

[0008] In summary, the solid state disk performance improvement method with reserved space dynamic adjustment function proposed by the present application can automatically calculate and adjust the reserved space according to a plurality of operating parameters and a remaining capacity, which is more accurate than the existing fixed or simple dynamic OP design in balancing write performance, NAND lifespan, and user's psychological perception of available capacity. By setting a recommended upper limit value U, a recommended lower limit value L, and a middle value M, and dynamically adjusting the reserved space according to the first change rate and the second change rate based on the remaining capacity, the write amplification effect can be effectively reduced, the NAND lifespan can be extended, and stable performance can be maintained when the remaining capacity is close to exhaustion. At the same time, when the solid state disk detects that the I / O load is low or enters an idle mode, the reserved space ratio is dynamically adjusted according to the aforementioned rules, further taking into account resource utilization efficiency and user demand for available capacity, to improve overall solid state disk performance, stability, and user experience. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 The method flowchart of the preferred embodiment of the present application.

[0010] Explanation of reference numerals: S1-S3 - steps. DETAILED DESCRIPTION

[0011] For those skilled in the art with ordinary knowledge, the technical content of the present application can be clearly understood by the following description with reference to the accompanying drawings. The following detailed description of the preferred embodiments of the performance improvement method of the solid state disk with dynamic reserved space adjustment function of the present application is made with reference to the flowchart shown in the drawings. Figure 1

[0012] First, in step S1, a plurality of operating parameters of the solid state disk are continuously monitored in a computer system, and the plurality of operating parameters are analyzed by weighting to generate a demand index. The operating parameters at least include: a write amplification factor (WAF) used to reflect the ratio of the actual data write amount to the user write amount, which is an important index for evaluating the write efficiency and NAND life of the solid state disk; a Trim command triggering frequency, which represents the frequency of the SSD controller recycling idle blocks, which can affect the write delay and available capacity management; a data write concentration used to reflect whether the data write is concentrated in part of the blocks, and the concentrated write can increase the block wear and affect the life; and an average I / O delay value (Average I / O Latency) used as a delay index for measuring the overall read and write operation, which is an important reference for user performance experience.

[0013] Through the above parameters, the performance load and data write behavior characteristics of the current solid state disk can be evaluated. It is particularly noted that the demand index is calculated by giving different weights according to the different considerations of each operating parameter, so that different algorithms can give different weights to each parameter according to the differences in their considerations. For example, if the write amplification factor is considered more important, the algorithm should increase the weight of the write amplification factor, and the weights of other parameters should be reduced. After comprehensive evaluation and analysis, the corresponding demand index is generated. It is worth noting that the difference in weight and the selection of algorithm are only cognitive or design considerations, so other algorithms may also consider the Trim command frequency or data write concentration as the main consideration, and the corresponding weight will be adjusted accordingly. In other words, since the weights of each parameter are designed differently according to the demand, the technical core of the present application is not to discuss the algorithm itself, but to use the demand index as a basis and as a subsequent dynamic reserved space adjustment execution, so as to improve the performance and resource management of the solid state disk.

[0014] ​Next, in step S2, based on the demand indicators generated in step S1, a suggested upper limit value U, a suggested lower limit value L, and an intermediate value M between the suggested upper limit value U and the suggested lower limit value L are derived for the reserved space. The suggested upper limit value U represents the maximum reserved space that can provide optimal performance when there is sufficient remaining capacity and high performance requirements; in other words, when the reserved space is set at the suggested upper limit value U, the solid-state drive (SSD) performance is better, and even if it exceeds the suggested upper limit value U, the performance impact is limited. The suggested lower limit value L is the minimum reserved space that can be maintained to minimize wear and tear on the SSD when there is limited remaining capacity or high capacity requirements. Therefore, this value is the minimum lifespan protection threshold that the SSD should not fall below; that is, even if the capacity is insufficient, the reserved space should not be reduced beyond this minimum value to avoid irreversible damage to the NAND and affecting its lifespan. The intermediate value M serves as an adjustment reference between U and L, used to control the dynamic range of the reserved space as the remaining capacity changes. This part is also the main technical core of the present invention, with the aim of achieving a balance between performance, lifespan, and the user's psychological rationality in capacity allocation.

[0015] Next, in step S3, the computer system compares the current remaining capacity of the solid-state drive with the upper, lower, and intermediate values ​​of the reserved space derived in step S2, and automatically adjusts the reserved space according to the following adjustment strategy:

[0016] (1) When the remaining capacity is greater than the recommended upper limit U: the reserved space remains at the recommended upper limit U to ensure that the solid-state drive can provide the best performance; (2) When the remaining capacity is less than or equal to the recommended upper limit U and greater than the median M: the reserved space changes with the remaining capacity at a first ratio equal to 1, that is, the reserved space changes by the same amount as the remaining capacity changes, so that the reserved space can be gradually adjusted as the remaining capacity decreases, so as to balance performance and capacity utilization; (3) When the remaining capacity is less than or equal to the median M and greater than or equal to the recommended lower limit L: the reserved space changes synchronously with the remaining capacity at a second ratio less than 1, and when the remaining capacity is almost exhausted, the reserved space drops to the recommended lower limit L to maintain the bottom line of NAND lifespan. That is, although the reserved space increases or decreases in the same direction as the remaining capacity, its change is less than that of the remaining capacity. Similarly, when the remaining capacity is released in reverse, the reserved space is dynamically adjusted according to the first ratio and the second ratio, so that the reserved space can increase synchronously as the remaining capacity recovers, balancing performance, lifespan and the user's psychological perception of available capacity.

[0017] To illustrate this further, let's assume that step S2 calculates a suggested upper limit U of 40% for SSD capacity allocation, a suggested lower limit L of 20%, and an intermediate value M of 30%. Based on the adjustment strategy in step S3, the reserved space changes with the remaining capacity as follows:

[0018] When the remaining capacity is greater than 40% (U), the reserved space remains at 40%, achieving optimal performance with ideal write latency and write amplification. When the remaining capacity is between 30% (M) and 40% (U), the reserved space changes with the remaining capacity at a ratio equal to 1 (the change in reserved space divided by the change in remaining capacity equals 1). For example, if the remaining capacity decreases from 40% to 35%, the reserved space decreases at the same rate to approximately 35%, maintaining a balance between performance improvement and capacity. When the remaining capacity is between 20% (L) and 30% (M), the reserved space changes with the remaining capacity at a smaller ratio (the change in reserved space divided by the change in remaining capacity less than 1). For example, if the remaining capacity decreases from 30% to 25%, the reserved space only decreases to approximately 27%–26%, a smaller decrease than the decrease in remaining capacity, allowing the reserved space to balance operational performance and NAND lifespan protection. When the remaining capacity approaches 20% (L), the reserved space also drops to the lower limit of 20%, providing a minimum baseline for NAND lifespan protection. Even if the capacity is insufficient, it will not decrease further. Therefore, through the above adjustment strategy, this invention can precisely control the dynamic range of the reserved space based on the remaining capacity and demand indicators. This not only maintains SSD performance and extends NAND lifespan but also considers the user's psychological perception of available capacity, achieving the optimal balance between performance, lifespan, and capacity resources.

[0019] Furthermore, in another embodiment, after the computer system completes the dynamic adjustment strategy setting for reserved space in steps S1 to S3, it continuously monitors the I / O load status of the solid-state drive (SSD). When the SSD I / O load is detected to be low or enters an idle mode, the operation of adjusting the reserved space ratio is triggered. In other words, the adjustment of the reserved space ratio is not performed at a fixed period or at any arbitrary time, but is triggered by the actual load status of the SSD: it is only executed when the I / O load is reduced or idle, so as to avoid interfering with normal performance during high-load write operations. Accordingly, the computer system can effectively schedule the timing of reserved space ratio adjustment, minimizing the impact of the dynamic management process on user operations, while improving the overall stability of SSD performance and lifespan control.

[0020] In summary, the solid-state drive (SSD) performance improvement method with dynamic adjustment of reserved space proposed in this invention continuously monitors multiple operating parameters of the SSD to generate demand indicators and adopts a dynamic adjustment strategy based on the remaining capacity, allowing the reserved space to automatically change according to usage. When the remaining capacity is sufficient and performance demand is high, the reserved space can be maintained at the recommended upper limit value U to provide optimal write performance. When the remaining capacity decreases, the reserved space can be dynamically adjusted according to different ranges of the recommended upper limit value U, the middle value M, and the lower limit value L, taking into account performance, product lifespan, and capacity utilization to ensure the user's psychological perception of available capacity. Furthermore, this invention can also select appropriate times to adjust the reserved space ratio based on I / O load or idle state, avoiding performance interference during high-load operations and improving the accuracy and stability of dynamic management. In summary, this invention can automatically and accurately manage reserved space while taking into account SSD performance, NAND lifespan, and the user's psychological perception of available capacity, achieving the technical effects of improving overall SSD performance, extending storage life, and improving user experience.

[0021] 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, all equivalent changes and modifications made without departing from the scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for improving the performance of a solid-state drive (SSD) with dynamically adjustable reserved space, characterized in that: A computer system continuously monitors and weights multiple operating parameters of a solid-state drive (SSD) to generate a demand index. These multiple operating parameters include at least a write amplification factor, a Trim command trigger frequency, a data write concentration, and an average I / O latency value. Based on the demand index, a suggested upper limit value U and a suggested lower limit value L for the reserved space are derived, along with the midpoint M between the suggested upper limit value U and the suggested lower limit value L. The computer system then adjusts the reserved space by comparing the current remaining capacity of the SSD with the reserved space according to an adjustment strategy. The adjustment strategy is as follows: when the remaining capacity of the solid-state drive is greater than U during use, the reserved space remains at the recommended upper limit; when the remaining capacity of the solid-state drive is less than or equal to U and greater than M, the reserved space changes with the remaining capacity at a first ratio equal to 1; when the remaining capacity of the solid-state drive is less than or equal to M and greater than or equal to L, the reserved space changes with the remaining capacity at a second ratio less than 1, and when the remaining capacity is nearly exhausted, the reserved space drops to the recommended lower limit. When the remaining space is released in the opposite direction, the reserved space is dynamically adjusted using the first ratio and the second ratio.

2. The method for improving solid-state drive performance as described in claim 1, characterized in that, When the solid-state drive detects low I / O load or enters idle mode, the reserved space ratio is adjusted.

Images