A read disturbance detection method, a read disturbance detection device, a medium and a product

Abstract

A read disturbance detection method, a read disturbance detection device, a medium and a product, the method comprising: a read disturbance detection device obtains a data access record set of multiple virtual machines, and identifies read disturbance hotspot blocks based on these data access records, and further analyzes the access behavior of these hotspot blocks, and finally determines the serious read disturbance blocks. This layer-by-layer analysis method can not only detect read disturbance problems in a timely manner, but also accurately evaluate the degree of read disturbance, overcoming the limitations of related technologies that only rely on static resource allocation or simple threshold judgment, so as to better cope with the dynamic and changeable data access mode in the cloud environment.

Description

Technical Field

[0001] The present application relates to the field of storage technology, and in particular to a read disturbance detection method, a read disturbance detection device, a medium and a product. Background Art

[0002] In a cloud computing environment, multiple virtual machines share physical storage resources. This resource sharing model significantly improves resource utilization efficiency. However, when multiple virtual machines intensively access storage devices at the same time, the read performance of the storage devices often decreases, affecting the normal operation of the business.

[0003] Currently, the industry generally adopts resource isolation or performance monitoring methods to reduce read interference between virtual machines. For example, resource isolation is achieved by limiting the access rate of virtual machines to storage devices, or allocating a fixed I / O bandwidth to each virtual machine. In addition, resource allocation can be dynamically adjusted by monitoring storage access patterns and response times to reduce performance interference during runtime.

[0004] However, related technologies mainly focus on dealing with read interference through static resource allocation or simple dynamic adjustment, which often does not work well when dealing with high concurrency and dynamically changing data access patterns. In particular, when the virtual machine access pattern changes frequently or sudden high access loads occur, it is difficult to accurately and timely identify the hot blocks of read interference, resulting in a slow response to the performance optimization of storage devices. Summary of the invention

[0005] The present application provides a read disturbance detection method, a read disturbance detection device, a medium and a product for accurately identifying hot spots of read disturbance.

[0006] In a first aspect, the present application provides a read interference detection method, which is applied to a read interference detection device, the method comprising: obtaining a data access set of each of a plurality of virtual machines, the data access set comprising a plurality of data access records, the data access record comprising a data access time, a data access block, and a data response time of the data access block; based on the data access record, determining a first number of different virtual machines accessing each data block within a preset time period and a first average response time of each data block; determining a data block for which the first number exceeds a preset number threshold or the first average response time exceeds a preset time threshold as a read interference hotspot block; based on the data access record, determining A second number of different virtual machines accessing the read interference hotspot block within a certain unit time and a second average response time of the target virtual machine, the target virtual machine being used to represent the virtual machine accessing the read interference hotspot block within the unit time; determining the read interference degree of the read interference hotspot block according to the second number, the preset benchmark concurrency number, the second average response time and the preset benchmark response time; determining the read interference hotspot block whose read interference degree exceeds the preset degree threshold as a severe read interference block; and storing the identification code, read interference degree, read interference start time, read interference end time of the severe read interference block and the virtual machine identification list accessing the severe read interference block in the read interference log.

[0007] By adopting the above technical solution, the read interference detection device obtains the data access records of multiple virtual machines, identifies the read interference hotspot blocks based on these data access records, and further analyzes the access behavior of these hotspot blocks to finally determine the serious read interference blocks. This progressive analysis method can not only detect read interference problems in a timely manner, but also accurately evaluate the degree of read interference, overcoming the limitations of related technologies that only rely on static resource allocation or simple threshold judgment, so as to better cope with the dynamic and changeable data access patterns in cloud environments.

[0008] In combination with some embodiments of the first aspect, in some embodiments, based on the data access record, determining the first number of different virtual machines accessing each data block within a preset time length and the first average response time of each data block specifically includes: determining the target data access record within the preset time length according to the data access moment; establishing a data block index table based on the target data access block in the target data access record, the data block index table including an identification code of each target data block accessed; counting the number of different virtual machines accessing each target data block in the data block index table within the preset time length to obtain the first number; calculating the arithmetic mean of the target data response time of the target data block within the preset time length to obtain the first average response time.

[0009] By adopting the above technical solution, the read interference detection device establishes a data block index table, organizes discrete data access records into a structured form, and greatly improves the efficiency of subsequent statistical analysis. Afterwards, the read interference detection device counts the number of different virtual machines that access each target data block in the data block index table within a preset time length and the arithmetic mean of the target data response time length of the target data block within the preset time length, and obtains the first number of different virtual machines that access each data block within the preset time length and the first average response time length of each data block, so that the read interference detection device can quickly respond to changes in the read interference situation.

[0010] In combination with some embodiments of the first aspect, in some embodiments, the read interference degree of the read interference hotspot block is determined based on the second number, the preset benchmark concurrency number, the second average response time and the preset benchmark response time, specifically including: calculating a first ratio of the second number to the preset benchmark concurrency number, and a second ratio of the second average response time to the preset benchmark response time; based on a preset first weighting coefficient and a preset second weighting coefficient, performing a weighted calculation on the first ratio and the second ratio to obtain a weighted result; and determining the weighted result as the read interference degree of the read interference hotspot block.

[0011] By adopting the above technical solution, the read interference detection device calculates the ratio of the actual concurrent access number to the preset benchmark concurrent number, and the ratio of the actual average response time to the preset benchmark response time, and introduces a weighted coefficient to combine these two ratios, which takes into account both the access pressure and the performance impact, making the final read interference degree assessment more comprehensive and accurate.

[0012] In combination with some embodiments of the first aspect, in some embodiments, after the step of determining the read interference degree of the read interference hotspot block based on the second number, the preset benchmark concurrency number, the second average response time and the preset benchmark response time, the method also includes: obtaining the historical read interference degree of the read interference hotspot block; predicting the future read interference degree of the read interference hotspot block based on the current read interference degree and the historical read interference degree; if the future read interference degree indicates that the read interference degree is increasing, migrating the data of the read interference hotspot block to a preset storage area.

[0013] By adopting the above technical solution, the read disturbance detection device introduces the historical read disturbance level and combines the current read disturbance level to predict future trends, so that potential read disturbance risks can be discovered in advance instead of passively waiting for problems to occur, which greatly reduces the risk of serious performance degradation due to read disturbance and improves the stability and reliability of storage devices.

[0014] In combination with some embodiments of the first aspect, in some embodiments, before the step of migrating the data of the read interference hotspot block to a preset storage area if the future read interference level indicates that the read interference level has increased, the method also includes: determining the rate of change of the future read interference level based on multiple future read interference level values ​​within a future preset time period; if the rate of change of the future read interference level exceeds a preset change rate threshold, determining that the read interference level has increased; if the rate of change of the future read interference level is lower than the preset change rate threshold, determining that the read interference level has decreased.

[0015] By adopting the above technical solution, the read interference detection device calculates the change rate of the future read interference level within a preset time period in the future, and combines it with the preset change rate threshold, so as to accurately judge the change trend of the future read interference level, so that the read interference detection device can discover potential problems earlier, especially when the read interference level rises rapidly in the future, and timely trigger the data migration operation to effectively prevent the performance of the storage device from dropping sharply.

[0016] In combination with some embodiments of the first aspect, in some embodiments, before the step of migrating the data of the read interference hotspot block to a preset storage area if the future read interference level indicates that the read interference level is increasing, the method also includes: obtaining a list of available storage areas in the storage device; calculating the current read load of each available storage area; and determining the available storage area with the smallest current read load as the preset storage area.

[0017] By adopting the above technical solution, the read interference detection device obtains a list of available storage areas in the storage device, calculates the current read load of each available storage area in real time, and selects the available storage area with the smallest current read load as the migration target, thereby ensuring load balancing after data migration, avoiding new hot spot problems that may be caused by simple fixed allocation, and making more reasonable use of resources.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of storing the identification code, read interference degree, read interference start time, read interference end time and virtual machine identification list accessing the severe read interference block in the read interference log, the method also includes: obtaining data access pattern characteristics of the severe read interference block, the data access pattern characteristics include data access frequency, data access time distribution and data access spatial locality; classifying the severe read interference block according to the data access pattern characteristics to obtain a read interference block type; based on the read interference block type, determining a corresponding storage optimization strategy, the storage optimization strategy includes at least one of a partition storage strategy, a cache allocation strategy and a bandwidth allocation strategy.

[0019] By adopting the above technical solution, the read interference detection device obtains the data access pattern characteristics of the serious read interference blocks, and classifies the serious read interference blocks according to these data access characteristics, and selects corresponding storage optimization strategies for different types of read interference blocks, including partition storage, cache allocation, bandwidth allocation and other means. This access pattern-based classification and optimization method enables the read interference detection device to solve different types of read interference problems more targeted, thereby improving the overall performance and resource utilization efficiency of the storage device. For example, for blocks with high frequency access but concentrated time distribution, a cache optimization strategy can be adopted; for blocks with low access frequency but strong spatial locality, a partition storage strategy can be adopted.

[0020] In a second aspect, an embodiment of the present application provides a read interference detection device, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the read interference detection device to perform the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, and when the above-mentioned computer program product runs on a read interference detection device, the above-mentioned read interference detection device executes the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions. When the instructions are executed on a read interference detection device, the read interference detection device executes the method described in the first aspect and any possible implementation manner of the first aspect.

[0023] It can be understood that the read disturbance detection device provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0025] 1. By adopting the above technical solution, the read interference detection device obtains the data access records of multiple virtual machines, identifies the read interference hotspot blocks based on these data access records, and further analyzes the access behavior of these hotspot blocks to finally determine the serious read interference blocks. This progressive analysis method can not only detect read interference problems in a timely manner, but also accurately evaluate the degree of read interference, overcoming the limitations of related technologies that only rely on static resource allocation or simple threshold judgment, so as to better cope with the dynamic and changeable data access patterns in cloud environments.

[0026] 2. By adopting the above technical solution, the read interference detection device establishes a data block index table, organizes discrete data access records into a structured form, and greatly improves the efficiency of subsequent statistical analysis. At the same time, the read interference detection device counts the number of different virtual machines that access each target data block in the data block index table within a preset time period, avoiding redundant calculation of historical data. This processing method significantly reduces the computational overhead while ensuring the accuracy of read interference detection, so that the read interference detection device can quickly respond to changes in read interference conditions.

[0027] 3. By adopting the above technical solution, the read interference detection device obtains a list of available storage areas in the storage device, and calculates the current read load of each available storage area in real time, and selects the available storage area with the smallest current read load as the migration target, thereby ensuring load balancing after data migration, avoiding new hot spot problems that may be caused by simple fixed allocation, and making more reasonable use of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of a read disturbance detection method in an embodiment of the present application;

[0029] Figure 2 is another flowchart of the read disturbance detection method in an embodiment of the present application;

[0030] Figure 3 It is a schematic diagram of a physical device structure of a read disturbance detection device in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0032] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0033] The following is a description of the process of the method provided by this implementation. Figure 1 , is a flow chart of a read disturbance detection method in an embodiment of the present application.

[0034] S101, obtaining data access sets of respective virtual machines, wherein the data access sets include multiple data access records, and the data access records include data access time, data access block, and data response duration of the data access block;

[0035] Among them, a virtual machine refers to a simulated computer system created on a physical computer through virtualization technology, which is used to independently run an operating system and application programs; a data access set refers to a collection of all data access records of a virtual machine's read operations on a storage device within a specific time period; a data access record is used to represent the detailed information of a single data access operation; the data access moment is used to represent the specific time point when the virtual machine initiates a data access request, usually recorded in the form of a timestamp; a data access block refers to a specific data block on the accessed storage device, identified by a unique block identifier; and the data response time refers to the time interval from initiating a data access request to receiving complete data, usually in milliseconds.

[0036] Specifically, the read disturbance detection device continuously collects data access records of all running virtual machines through the monitoring interface and stores them in the database. The read disturbance detection device retrieves the data access set of the virtual machine according to the identifier of the virtual machine.

[0037] Assume that there are three virtual machines (VM1, VM2, and VM3) in a data center. Each virtual machine has its own independent data access set. The data access status of the three virtual machines within a certain minute is as follows:

[0038] Data access set for VM1:

[0039] Data access record 1:

[0040] Data access time: 2024-01-01-10:00:00.123;

[0041] Data access block: Block_A123;

[0042] Data response time: 5ms;

[0043] Data access record 2:

[0044] Data access time: 2024-01-01-10:00:15.456;

[0045] Data access block: Block_B456;

[0046] Data response time: 8ms;

[0047] Data access record 3:

[0048] Data access time: 2024-01-01-10:00:45.789;

[0049] Data access block: Block_A123;

[0050] Data response time: 6ms;

[0051] Data access set for VM2:

[0052] Data access record 1:

[0053] Data access time: 2024-01-01-10:00:05.234;

[0054] Data access block: Block_C789;

[0055] Data response time: 3ms;

[0056] Data access record 2:

[0057] Data access time: 2024-01-01-10:00:30.567;

[0058] Data access block: Block_D012;

[0059] Data response time: 7ms;

[0060] Data access set for VM3:

[0061] Data access record 1:

[0062] Data access time: 2024-01-01-10:00:10.345;

[0063] Data access block: Block_E345;

[0064] Data response time: 4ms

[0065] Data access record 2:

[0066] Data access time: 2024-01-01-10:00:25.678;

[0067] Data access block: Block_F678;

[0068] Data response time: 6ms;

[0069] Data access record 3:

[0070] Data access time: 2024-01-01-10:00:55.901;

[0071] Data access block: Block_E345;

[0072] Data response time: 5ms;

[0073] Based on the above:

[0074] VM1 accessed data 3 times within this 1 minute, VM2 accessed data 2 times, and VM3 accessed data 3 times. Some data blocks may be accessed multiple times (for example, VM1 accessed Block_A123 twice, and VM3 accessed Block_E345 twice).

[0075] S102, based on the data access record, determining a first number of different virtual machines accessing each data block within a preset time period and a first average response time of each data block;

[0076] Among them, the preset duration refers to the time window for hotspot detection, which can usually be configured to minutes or hours; the first number refers to the number of different virtual machines accessing the same data block within the preset duration, which is used to measure the concurrency of data access; the first average response time refers to the average access response time of the data block within the preset duration, which is used to measure the access performance.

[0077] Specifically, first, the read interference detection device determines the time window for hot spot detection and filters out all data access records within the time window. Then, the read interference detection device establishes a data block index table to record all accessed data blocks. For each data block in the data block index table, the read interference detection device counts the number of different virtual machines accessing the data block, and calculates the arithmetic mean of the data response time of all data access records of the data block. This process is performed once for each accessed data block, and finally, the read interference detection device obtains the first number of different virtual machines accessing each data block within a preset time length and the first average response time of each data block.

[0078] Optionally, under normal circumstances, based on the data access record, determining the first number of different virtual machines accessing each data block within a preset time length and the first average response time of each data block can be achieved in the following manner, which is not limited here: determining the target data access record within the preset time length according to the data access moment; establishing a data block index table based on the target data access block in the target data access record, the data block index table including an identification code for each target data block accessed; counting the number of different virtual machines accessing each target data block in the data block index table within the preset time length to obtain the first number; calculating the arithmetic mean of the target data response time of the target data block within the preset time length to obtain the first average response time.

[0079] Following the example in step S101, assuming that the preset duration is 1 minute, all accessed data blocks are: Block_A123, Block_B456, Block_C789, Block_D012, Block_E345, Block_F678, and the first number of different virtual machines accessing each data block within the preset duration and the first average response time of each data block are as follows:

[0080] S103, determining the data block in which the first number exceeds a preset number threshold or the first average response time exceeds a preset time threshold as a read disturbance hotspot block;

[0081] Among them, the preset quantity threshold refers to the standard for the number of concurrent accesses by virtual machines to determine whether a data block is a read interference hotspot block; the preset time threshold refers to the standard for the average response time to determine whether a data block is a read interference hotspot block; a read interference hotspot block refers to a data block where there is a concurrent access conflict among multiple virtual machines or where access performance is degraded.

[0082] Specifically, the read interference detection device compares the first number of each data block (the number of concurrently accessed virtual machines) with a preset number threshold, and at the same time, compares the first average response time with a preset time threshold. When the first number of a data block exceeds the preset number threshold or the first average response time exceeds the preset time threshold, the read interference detection device marks the data block as a read interference hotspot block. This dual judgment standard can more comprehensively identify data blocks that may have read interference problems.

[0083] S104, based on the data access record, determining a second number of different virtual machines accessing the read interference hotspot block within a unit time and a second average response time of a target virtual machine, where the target virtual machine is used to represent a virtual machine accessing the read interference hotspot block within the unit time;

[0084] Among them, unit time refers to the minimum time unit for detailed read interference analysis, usually in seconds or minutes; the second number refers to the number of different virtual machines accessing the same read interference hotspot block within the unit time, which is used to measure the short-term concurrent access intensity; the target virtual machine refers to the virtual machine that has accessed the read interference hotspot block within the unit time; the second average response time refers to the average access response time of the target virtual machine accessing the read interference hotspot block within the unit time, which is used to measure the short-term access performance.

[0085] Specifically, first, the read interference detection device divides the time axis into continuous unit time segments, and then the read interference detection device analyzes the access situation of each read interference hotspot block in each unit time. The read interference detection device counts the number of different virtual machines accessing the read interference hotspot block in each unit time, records the identification information of these virtual machines, and calculates their average response time for accessing the read interference hotspot block, thereby obtaining the second number of different virtual machines accessing the read interference hotspot block in unit time and the second average response time of the target virtual machine.

[0086] S105, determining the read disturbance degree of the read disturbance hotspot block according to the second number, the preset benchmark concurrency number, the second average response time, and the preset benchmark response time;

[0087] Among them, the preset benchmark concurrency number refers to the expected number of concurrent access virtual machines when the storage device is operating normally; the preset benchmark response time refers to the expected data access response time standard when the storage device is operating normally; the read interference degree refers to a quantitative indicator reflecting the severity of the read interference, usually a normalized value.

[0088] Specifically, first, the read interference detection device calculates the ratio of the second number to the preset benchmark concurrent number, and the ratio of the second average response time to the preset benchmark response time. Then, the read interference detection device performs weighted calculation on these two ratios according to the pre-configured weighting coefficient to obtain a comprehensive read interference degree. This calculation method takes into account both the pressure brought by concurrent access and the degree of performance degradation, and can fully reflect the severity of the read interference.

[0089] Optionally, under normal circumstances, determining the read interference degree of the read interference hotspot block based on the second number, the preset benchmark concurrency number, the second average response time and the preset benchmark response time can be achieved in the following manner, which is not limited here: calculating a first ratio of the second number to the preset benchmark concurrency number, and a second ratio of the second average response time to the preset benchmark response time; based on a preset first weighting coefficient and a preset second weighting coefficient, performing a weighted calculation on the first ratio and the second ratio to obtain a weighted result; and determining the weighted result as the read interference degree of the read interference hotspot block.

[0090] S106, determining the read disturbance hotspot block whose read disturbance degree exceeds a preset degree threshold as a severe read disturbance block;

[0091] The preset degree threshold is a standard value used to determine whether the read disturbance is serious; a serious read disturbance block refers to a read disturbance hotspot block whose read disturbance degree exceeds the preset degree threshold, and such a block may significantly affect the performance of the storage device.

[0092] Specifically, the read disturbance detection device compares the read disturbance degree of each read disturbance hotspot block with a preset degree threshold. When the read disturbance degree of a read disturbance hotspot block exceeds the preset degree threshold, the read disturbance detection device marks it as a severe read disturbance block.

[0093] S107: Store the identification code, read disturbance degree, read disturbance start time, read disturbance end time, and a list of virtual machine identifications accessing the severe read disturbance block in a read disturbance log.

[0094] Among them, the identification code refers to the unique identifier of the severe read interference block; the read interference start time refers to the starting time point when the read interference is detected; the read interference end time refers to the time point when the read interference state ends; the virtual machine identification list refers to the set of unique identifiers of all virtual machines that have accessed the severe read interference block; the read interference log refers to a persistent storage file that records information related to read interference.

[0095] Specifically, the read disturbance detection device writes this information into a read disturbance log file in a predetermined format. The read disturbance log can be used for subsequent performance analysis, trend prediction and system optimization, and can also be used as important reference data for system operation and maintenance.

[0096] By adopting the above technical solution, the read interference detection device obtains the data access records of multiple virtual machines, identifies the read interference hotspot blocks based on these data access records, and further analyzes the access behavior of these hotspot blocks to finally determine the serious read interference blocks. This progressive analysis method can not only detect read interference problems in a timely manner, but also accurately evaluate the degree of read interference, overcoming the limitations of related technologies that only rely on static resource allocation or simple threshold judgment, so as to better cope with the dynamic and changeable data access patterns in cloud environments.

[0097] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the read disturbance detection method in an embodiment of the present application.

[0098] After step S105, the following steps may or may not be performed, which is not limited here:

[0099] S201, obtaining the historical read disturbance degree of the read disturbance hotspot block;

[0100] Specifically, the read disturbance detection device queries the read disturbance hotspot block from the read disturbance log, and obtains its historical read disturbance degree within a preset time period in the past.

[0101] S202, predicting the future read disturbance degree of the read disturbance hotspot block according to the current read disturbance degree and the historical read disturbance degree;

[0102] Specifically, the read interference detection device uses a time series analysis method to combine the current read interference level with the historical read interference level to construct a prediction model. The prediction model takes into account the characteristics of the time trend, periodic changes, and sudden fluctuations of the data, and calculates the future read interference level of the read interference hotspot block within a preset time period in the future (such as the next hour) through an algorithm. The prediction results include the predicted values ​​of multiple future time points, forming a prediction sequence.

[0103] S203, determining a change rate of the future read disturb degree according to multiple future read disturb degree values ​​within a future preset time period;

[0104] Among them, the future preset time length refers to the time range for trend analysis; the values ​​of multiple future read disturbance levels refer to the predicted values ​​of future read disturbance levels at multiple future time points in the prediction results; and the change rate refers to the speed at which the future read disturbance level changes over time.

[0105] Specifically, first, the read interference detection device determines the time range for trend analysis, that is, the preset time length in the future, and usually selects the most representative time period in the prediction sequence. Then, the read interference detection device calculates the difference in the future read interference degree between adjacent future time points, and divides it by the time interval to obtain the instantaneous change rate of each future time point. The read interference detection device also calculates the average change rate within the entire future preset time length for overall trend judgment. For the calculation of the change rate, the read interference detection device adopts the numerical differentiation method and performs necessary data smoothing to reduce the impact of noise.

[0106] S204, if the change rate of the future read disturb level exceeds a preset change rate threshold, determining that the read disturb level increases;

[0107] The preset change rate threshold is used to represent a preset change rate judgment standard, and is used to distinguish the significance of the change in the read disturbance level; an increase in the read disturbance level means that the read disturbance situation is worsening.

[0108] Specifically, the read disturbance detection device compares the calculated change rate with a preset change rate threshold. When the change rate is positive and exceeds the preset change rate threshold, the read disturbance detection device marks the state of the read disturbance hotspot block as an increased read disturbance level.

[0109] S205, if the future read disturbance level indicates that the read disturbance level increases, then migrate the data of the read disturbance hotspot block to a preset storage area;

[0110] The data migration refers to the process of copying data from an original storage location to a new storage location; the preset storage area refers to the target storage location determined in step S208.

[0111] Specifically, as shown in step S204, when the future read interference level indicates that the read interference level is increasing, the read interference detection device will formulate a detailed migration plan, including the migration data range, migration time window and migration priority. The read interference detection device will continuously monitor the performance of the source area and the target area during the migration process to ensure that the migration process does not cause a significant performance degradation.

[0112] S206, if the change rate of the future read disturbance level is lower than the preset change rate threshold, determine that the read disturbance level decreases;

[0113] Among them, the decrease in the read disturbance level means that the read disturbance situation is improving.

[0114] Specifically, the read disturbance detection device compares the calculated change rate with a preset change rate threshold. When the change rate is negative and its absolute value is lower than the preset change rate threshold, the read disturbance detection device marks the state of the read disturbance hotspot block as a decreased read disturbance level.

[0115] S207, obtaining a list of available storage areas in the storage device;

[0116] The storage device refers to a physical device used for data storage, such as a hard disk array or a solid-state memory; the available storage area list refers to a collection of free storage areas that can currently be used for data storage.

[0117] Specifically, first, the read disturbance detection device obtains all idle available storage areas in the storage device, including characteristic information such as capacity, location, performance parameters, etc. of these available storage areas. The read disturbance detection device organizes this information into a structured available storage area list, and ensures that each available storage area in the available storage area list meets the minimum performance requirements and availability conditions.

[0118] S208, calculating the current read load of each available storage area;

[0119] Specifically, the read interference detection device monitors the performance of each available storage area in the available storage area list, and collects performance indicators in multiple dimensions including the number of IO requests, data throughput, average response time, etc. The read interference detection device uses a weighted calculation method to integrate these performance indicators into a standardized current read load.

[0120] S209, determining the available storage area with the smallest current read load as the preset storage area;

[0121] The available storage area with the smallest current read load refers to the available storage area with the lowest access pressure among all available storage areas; and the preset storage area refers to the target storage location selected to receive the migration data.

[0122] Specifically, first, the read interference detection device sorts the current read loads of all available storage areas and finds the available storage area with the smallest current read load. The read interference detection device verifies whether the available storage area meets the capacity requirements and checks whether its current state is stable. If the available storage area with the smallest current read load does not meet the capacity requirements, the read interference detection device selects the available storage area with the second smallest current read load and meets the capacity requirements. After selection, the read interference detection device marks the available storage area as a preset storage area and retains its related configuration information to prepare for subsequent data migration.

[0123] S210, determining the read disturbance hotspot block whose read disturbance degree exceeds a preset degree threshold as a severe read disturbance block;

[0124] For details, please refer to step S106, which will not be described in detail here.

[0125] S211, storing the identification code, read disturbance degree, read disturbance start time, read disturbance end time, and a list of virtual machine identifications accessing the severe read disturbance block in a read disturbance log.

[0126] For details, please refer to step S107, which will not be described in detail here.

[0127] S212, obtaining data access pattern characteristics of the severe read interference block, where the data access pattern characteristics include data access frequency, data access time distribution, and data access spatial locality;

[0128] Among them, data access pattern characteristics refer to the feature set that describes the laws of data access behavior; data access frequency refers to the number of times data is accessed per unit time; data access time distribution refers to the distribution law of data access in the time dimension; data access spatial locality refers to the degree to which adjacent data blocks are accessed continuously.

[0129] Specifically, the read interference detection device extracts the historical data of the serious read interference block, and usually analyzes the data of the last 24 hours. The read interference detection device calculates the number of accesses in each time window to obtain the data access frequency, analyzes the concentration and periodicity of the access time to obtain the data access time distribution, and counts the range of continuously accessed data blocks to obtain the data access spatial locality.

[0130] S213, classifying the severe read interference block according to the data access mode characteristics to obtain a read interference block type;

[0131] Specifically, the read interference detection device uses a machine learning classification algorithm, takes the collected data access pattern features as input, and automatically classifies the severe read interference blocks. Common classifications include: high-frequency periodic access type, burst intensive access type, continuous low-frequency access type, etc. The read interference detection device calculates a confidence score for each read interference block type to ensure the reliability of the classification result. Among them, the read interference block type refers to the result of classifying severe read interference blocks based on data access pattern features.

[0132] S214. Determine a corresponding storage optimization strategy based on the read interference block type, where the storage optimization strategy includes at least one of a partition storage strategy, a cache allocation strategy, and a bandwidth allocation strategy.

[0133] Among them, storage optimization strategy refers to the specific technical solution used to improve storage performance; partition storage strategy refers to optimizing access performance by adjusting the physical distribution of data on storage devices; cache allocation strategy is used to indicate how to allocate cache resources for different types of data; bandwidth allocation strategy refers to how to allocate IO bandwidth resources for different access requests.

[0134] Specifically, the read interference detection device selects the most suitable storage optimization strategy from the preset strategy library according to the type of read interference block. For high-frequency periodic access blocks, the read interference detection device may choose to increase the cache quota and adjust the cache replacement algorithm; for burst intensive access blocks, the read interference detection device may choose a dynamic bandwidth allocation strategy; for continuous low-frequency access blocks, the read interference detection device may choose to migrate data to an independent storage partition.

[0135] The following describes the read disturbance detection device in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , is a schematic diagram of a physical device structure of a read disturbance detection device in an embodiment of the present application.

[0136] It should be noted that Figure 3 The structure of the read disturbance detection device shown is only an example and should not bring any limitation to the function and scope of use of the embodiments of the present invention.

[0137] like Figure 3As shown, the read disturbance detection device includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0138] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0139] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0140] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0141] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0142] Specifically, the read disturbance detection device of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the read disturbance detection method provided by the above embodiment is implemented.

[0143] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the read disturbance detection device described in the above embodiment; or may exist independently without being assembled into the read disturbance detection device. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the read disturbance detection device, the read disturbance detection device implements the read disturbance detection method provided in the above embodiment.

[0144] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0145] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0146] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A read disturbance detection method, characterized in that: Applied to a read disturbance detection device, the method comprises: Acquire data access sets of respective virtual machines, wherein the data access sets include multiple data access records, and the data access records include data access time, data access blocks, and data response duration of the data access blocks; Based on the data access record, determining a first number of different virtual machines accessing each data block within a preset time period and a first average response time of each data block; Determine the data block where the first number exceeds a preset number threshold or the first average response time exceeds a preset time threshold as a read disturbance hotspot block; Based on the data access record, determining a second number of different virtual machines accessing the read interference hotspot block within a unit time and a second average response time of a target virtual machine, wherein the target virtual machine is used to represent a virtual machine accessing the read interference hotspot block within the unit time; Determine the read disturbance degree of the read disturbance hotspot block according to the second number, the preset benchmark concurrency number, the second average response time, and the preset benchmark response time; Determine the read disturbance hotspot block whose read disturbance degree exceeds a preset degree threshold as a severe read disturbance block; The identification code, the read disturbance degree, the read disturbance start time, the read disturbance end time of the severe read disturbance block and the virtual machine identification list accessing the severe read disturbance block are stored in a read disturbance log; The method of determining the first number of different virtual machines accessing each data block within a preset time period and the first average response time of each data block based on the data access record specifically includes: determining a target data access record within the preset time period according to the data access moment; establishing a data block index table based on the target data access block in the target data access record, wherein the data block index table includes an identification code of each target data block accessed; counting the number of different virtual machines accessing each target data block in the data block index table within the preset time period to obtain the first number; and calculating the arithmetic mean of the target data response time of the target data block within the preset time period to obtain the first average response time. The method of determining the read interference degree of the read interference hotspot block according to the second number, the preset benchmark concurrency number, the second average response time and the preset benchmark response time specifically includes: calculating a first ratio of the second number to the preset benchmark concurrency number, and a second ratio of the second average response time to the preset benchmark response time; performing weighted calculation on the first ratio and the second ratio based on a preset first weighting coefficient and a preset second weighting coefficient to obtain a weighted result; and determining the weighted result as the read interference degree of the read interference hotspot block.

2. The method according to claim 1, characterized in that After the step of determining the read disturbance degree of the read disturbance hotspot block according to the second number, the preset benchmark concurrency number, the second average response time and the preset benchmark response time, the method further includes: Obtaining the historical read disturbance degree of the read disturbance hotspot block; Predicting the future read disturbance degree of the read disturbance hotspot block according to the current read disturbance degree and the historical read disturbance degree; If the future read disturbance level indicates that the read disturbance level increases, data in the read disturbance hotspot block is migrated to a preset storage area.

3. The method according to claim 2, characterized in that Before the step of migrating the data of the read disturbance hotspot block to a preset storage area if the future read disturbance level indicates that the read disturbance level has increased, the method further includes: Determining a change rate of the future read disturb degree according to multiple future read disturb degree values ​​within a future preset time period; If the change rate of the future read disturb level exceeds a preset change rate threshold, determining that the read disturb level has increased; If the change rate of the future read disturbance level is lower than the preset change rate threshold, it is determined that the read disturbance level decreases.

4. The method according to claim 2, characterized in that: Before the step of migrating the data of the read disturbance hotspot block to a preset storage area if the future read disturbance level indicates that the read disturbance level has increased, the method further includes: Get a list of available storage areas in the storage device; Calculate the current read load for each available storage area; The available storage area with the smallest current read load is determined as the preset storage area.

5. The method according to claim 1, characterized in that After the step of storing the identification code, the read disturbance degree, the read disturbance start time, the read disturbance end time, and the virtual machine identification list accessing the severe read disturbance block in the read disturbance log, the method further includes: Acquire data access pattern characteristics of the severe read interference block, wherein the data access pattern characteristics include data access frequency, data access time distribution, and data access spatial locality; Classifying the severe read interference block according to the data access mode characteristics to obtain a read interference block type; Based on the read interference block type, a corresponding storage optimization strategy is determined, where the storage optimization strategy includes at least one of a partition storage strategy, a cache allocation strategy, and a bandwidth allocation strategy.

6. A read disturbance detection device, characterized in that: The read interference detection device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the read interference detection device to perform the method described in any one of claims 1 to 5.

7. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on a read disturbance detection device, the read disturbance detection device is caused to execute the method according to any one of claims 1 to 5.

8. A computer program product, characterized in that When the computer program product runs on a read disturbance detection device, the read disturbance detection device is enabled to perform the method according to any one of claims 1 to 5.

Images