Method for determining optimal balance point of performance stability of storage equipment based on capacity adjustment

Abstract

The invention relates to the field of storage, in particular to a method for determining the optimal balance point of the performance stability of storage equipment based on capacity adjustment, and aims to solve the problem that the optimal balance point of the performance stability of the storage equipment cannot be effectively determined through capacity adjustment in the prior art. Firstly, the performance of the solid state disk under the maximum available capacity and the minimum available capacity is tested, then an intermediate value of the maximum capacity and the minimum capacity is calculated, the capacity of the solid state disk is adjusted to the intermediate value, and performance data of the intermediate value is obtained through testing. And comparing the performance data of the intermediate value with the performance data of the previous two endpoints, and judging the performance stability change trend. If the performance becomes better, taking the intermediate value as a new boundary, and continuing to carry out the bipartite test; and if the performance corresponding to the intermediate value is not as good as that of a certain end point, abandoning the interval in which the end point with relatively poor performance is located, taking the end point with relatively good performance and the intermediate value as a new interval boundary, and continuing the bipartite test. And the process is continuously repeated until the optimal performance balance point is found.

Description

Technical Field

[0001] This invention relates to the field of storage, specifically a method for determining the optimal balance point for the performance stability of storage devices based on capacity adjustment. Background Technology

[0002] With the rapid development of information technology, storage devices are becoming increasingly important in various systems. The performance stability of storage devices is crucial for the efficient operation of systems, reliable data storage, and fast access. In practical applications, there is a complex relationship between the capacity and performance of storage devices, and different capacity configurations often have a significant impact on performance stability.

[0003] On the one hand, increasing storage capacity may lead to a decrease in the read and write performance of storage devices. For example, in some large-capacity solid-state drives (SSDs), as the amount of stored data increases, the data read and write speeds decrease significantly, and performance stability deteriorates. On the other hand, insufficient capacity configuration may not meet the system's data storage and processing needs, similarly affecting stable performance. Currently, although there is some research on SSD performance optimization, most of it focuses on hardware upgrades and software algorithm improvements, lacking a systematic and effective method to determine the optimal balance point for storage device performance stability through capacity adjustment. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for determining the optimal balance point of storage device performance stability based on capacity adjustment, thereby solving the problem that existing technologies lack an effective way to determine the optimal balance point of storage device performance stability through capacity adjustment.

[0005] To solve the aforementioned technical problem, the technical solution adopted by the present invention is as follows: First, the performance of the solid-state drive was tested at its maximum and minimum available capacity, resulting in two data points on sequential read / write bandwidth stability.

[0006] First binary search: Calculate the median value between the maximum and minimum capacity, adjust the SSD capacity to this median value, and perform sequential read / write bandwidth stability tests again to obtain the corresponding performance data.

[0007] Multiple Binary Search Iterations: Based on the comparison between the current test results and the performance data of the two previous endpoints, the trend of performance stability is determined. If the performance stability corresponding to the median value is worse than that of the two endpoints, then the median value is used as the new boundary to continue the binary search test; if the performance corresponding to the median value is worse than that of a certain endpoint, then the interval containing the worse-performing endpoint is discarded, and the better-performing endpoint and the median value are used as the new interval boundary to continue the binary search test. This process is repeated continuously to gradually narrow the capacity range until the capacity point with the best performance stability is found or a capacity interval with the best performance stability is determined.

[0008] Determining the optimal performance curve: After multiple capacity adjustments and tests using the dichotomy method, a series of performance data and corresponding performance curves were obtained under different capacity configurations. These performance curves were comprehensively analyzed, taking into account the stability of performance indicators, performance fluctuations, and other practical application requirements, to determine a performance curve that represents the optimal balance point for the storage device in terms of performance stability. The capacity configuration corresponding to this curve is the optimal capacity configuration for the storage device in terms of performance stability.

[0009] A method for determining the optimal balance point for the performance stability of storage devices based on capacity adjustment has the following advantages: 1. Efficiency Improvement: Traditional testing methods require trying different capacity configurations one by one, which is time-consuming and labor-intensive. This solution uses a binary search iterative testing method, which significantly reduces the number of tests. For example, when dealing with large-capacity storage devices, if the traditional method requires hundreds of tests, the binary search iterative method, by halving the test interval each time, only requires about ten tests to lock in the optimal range, quickly find the performance-capacity balance point, and adapt to the efficient exploration needs of complex storage scenarios.

[0010] 2. Performance Optimization: Precise capacity configuration optimization significantly improves storage device performance. It reduces read and write latency during business operations, enabling faster query and write responses, such as in database transaction processing; it increases data throughput, meeting the high-frequency data read and write requirements of big data analytics, ensuring the high efficiency and stability of business systems, reducing lag, timeouts, and other failures, and enhancing business continuity and user experience.

[0011] 3. Versatility and Scalability: The solution is based on scientific testing and analysis processes and is not limited by device type. It is applicable to both single-machine storage such as SSDs and HDDs, as well as distributed storage clusters; it can also adapt to diverse business scenarios such as databases, big data analysis, and file storage. By adjusting parameters (such as capacity ratio and performance indicator weights), it can be adapted to new scenarios, possessing good versatility and scalability potential, reducing the cost of repeated development for different scenarios.

[0012] In summary, a method for determining the optimal balance point of storage device performance and stability based on capacity adjustment helps to quickly find the optimal capacity configuration for storage devices in terms of performance and stability. Attached Figure Description

[0013] Figure 1 This is a flowchart of the method described in Example 1. Detailed Implementation

[0014] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0015] Example 1 This embodiment discloses a method for determining the optimal balance point of storage device performance stability based on capacity adjustment, such as... Figure 1 As shown, it includes the following steps: 1) Capacity range initialization: First, determine the total capacity of the storage device, total_cap, and set the initial capacity test range. The minimum usable capacity, min_cap, is taken as a preset percentage of the total capacity (e.g., 10%), i.e., min_cap = total_cap. 0.1, maximum available capacity max_cap is the maximum configurable service capacity supported by the device.

[0016] The storage performance was tested at the minimum and maximum available capacities, and the performance data P1 (corresponding to min_cap) and P2 (corresponding to max_cap) were recorded. The performance metrics covered key parameters such as read and write speed, IOPS (input and output operands), and response latency.

[0017] 2) Binary search iterative test: Calculate the midpoint capacity mid_cap of the current test interval, adjust the storage device to the midpoint capacity, perform performance testing and record data point P3, and compare the performance of P3 with P1 and P2.

[0018] If P3 is better than P1 and P2, the original interval is divided into sub-intervals, namely [min_cap, mid_cap] and [mid_cap, max_cap], with the middle capacity mid_cap as the new boundary. The sub-intervals with better performance trends are tested first (e.g., according to the performance growth gradient, the interval that is closer to the middle capacity and is expected to improve performance is selected).

[0019] If P3 is better than P1 but worse than P2, discard the interval corresponding to the minimum capacity min_cap and update the test interval to [mid_cap, max_cap].

[0020] If P3 is better than P2 but worse than P1, discard the interval corresponding to the maximum capacity max_cap and update the test interval to [min_cap, mid_cap].

[0021] If P3 is worse than P1 and P2, discard the worst performing interval (discard the interval outside the corresponding boundary based on the worse of P1 and P2).

[0022] Determine if the termination conditions are met. Termination conditions include the test interval difference being equal to a preset minimum capacity threshold (e.g., 1GB), or the number of iterations reaching a preset maximum. If not met, repeat the binary search iterative test steps; if met, proceed to the performance analysis phase.

[0023] 3) Performance analysis and configuration determination: Summarize all test data points P1, P2... Pn (n is the number of iterations), and plot the capacity-performance curve, with the horizontal axis representing storage capacity and the vertical axis representing comprehensive performance indicators.

[0024] Based on the capacity-performance curve, the capacity configuration corresponding to the optimal performance balance point is determined. This balance point needs to take into account both performance and capacity utilization. If there is a performance peak in the curve, the capacity corresponding to the peak point is selected. If the performance tends to stabilize as the capacity increases, the optimal performance capacity is selected based on the business capacity requirements, provided that the requirements are met.

[0025] The above description is merely the basic principle and preferred embodiment of the present invention. Improvements and substitutions made by those skilled in the art based on the present invention are within the scope of protection of the present invention.

Claims

1. A method for stabilizing the optimal balance point of storage device performance stability based on capacity adjustment, characterized in that: Includes the following steps: S01. Set the maximum and minimum available capacity of the solid-state drive (SSD), and perform sequential read / write bandwidth stability tests at the maximum and minimum available capacities of the SSD, respectively, and record the performance data. S02, First binary search: Calculate the median value between the maximum and minimum capacity, adjust the SSD capacity to the median value, and perform sequential read / write bandwidth stability test again to obtain the performance data corresponding to the median value; S03. Multiple Binary Iterations: Compare the performance data corresponding to the intermediate value with the performance data of the previous two endpoints to determine the trend of performance stability. If the performance stability corresponding to the intermediate value is better than that of the two endpoints, then use the intermediate value as the new boundary and continue to perform binary testing. If the performance of the intermediate value is worse than that of a certain endpoint, then discard the interval containing the poorly performing endpoint, and use the high-performing endpoint and the intermediate value as the new interval boundary to continue the binary search test. Repeat this process until the termination condition is met; S04. Determine the optimal performance curve: Perform a comprehensive analysis of the performance data and corresponding performance curves obtained from multiple binary searches, and consider the needs of actual application scenarios to determine a performance curve that represents the optimal balance point of the storage device in terms of performance stability. The capacity configuration corresponding to the performance curve is the optimal capacity configuration of the storage device in terms of performance stability.

2. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 1, characterized in that: The maximum available capacity is the maximum configurable service capacity supported by the storage device, and the minimum available capacity is a preset percentage of the total capacity of the storage device.

3. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 2, characterized in that: The preset ratio is 10%.

4. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 1, characterized in that: In step S03, when continuing the binary search test with the median value as the new boundary, the sub-intervals with better performance trends are tested first.

5. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 1, characterized in that: If the performance of the intermediate value is worse than that of the two endpoints, discard the worst-performing interval and continue the binary search test on the remaining interval.

6. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 1, characterized in that: The comprehensive analysis process of the performance data and corresponding performance curves obtained from multiple binary search iterations is as follows: All test data points P1, P2, ..., Pn are summarized, where n is the number of iterations. A capacity-performance curve is plotted, with storage capacity on the horizontal axis and comprehensive performance indicators on the vertical axis. Based on the capacity-performance curve, the capacity configuration corresponding to the optimal performance balance point is determined. This optimal balance point must consider both performance and capacity utilization. If a performance peak exists in the capacity-performance curve, the capacity corresponding to the peak point is selected. If performance tends to stabilize with capacity growth, the optimal capacity is selected based on business capacity requirements, provided those requirements are met.

7. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 1, characterized in that: Performance data includes read / write speed, IOPS, and response latency.

8. The optimal balance point stabilization method for storage device performance stability based on capacity adjustment according to claim 1, characterized in that: Termination conditions include the test interval difference being equal to the preset minimum capacity threshold or the number of iterations reaching the preset maximum.

Images