WO2023050672A1 - 随机写性能调节的方法、装置、电子设备及存储介质 - Google Patents

随机写性能调节的方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2023050672A1
WO2023050672A1 PCT/CN2022/074598 CN2022074598W WO2023050672A1 WO 2023050672 A1 WO2023050672 A1 WO 2023050672A1 CN 2022074598 W CN2022074598 W CN 2022074598W WO 2023050672 A1 WO2023050672 A1 WO 2023050672A1
Authority
WO
WIPO (PCT)
Prior art keywords
hard disk
data set
iops
data
data frames
Prior art date
Application number
PCT/CN2022/074598
Other languages
English (en)
French (fr)
Inventor
宋琪
Original Assignee
苏州浪潮智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州浪潮智能科技有限公司 filed Critical 苏州浪潮智能科技有限公司
Priority to US18/564,565 priority Critical patent/US20240377951A1/en
Publication of WO2023050672A1 publication Critical patent/WO2023050672A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0638—Organizing or formatting or addressing of data
    • G06F3/064—Management of blocks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/061—Improving I/O performance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0653—Monitoring storage devices or systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
    • G06F3/0659—Command handling arrangements, e.g. command buffers, queues, command scheduling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671—In-line storage system
    • G06F3/0673—Single storage device
    • G06F3/0674—Disk device

Definitions

  • the present application relates to the technical field of hard disk control, in particular to a random write performance adjustment method, device, electronic equipment and storage medium.
  • solid-state drives are gradually introduced in the enterprise class due to their high random access performance, low power consumption, and strong environmental stability.
  • enterprise-level SSDs are more concerned with performance consistency, typically with a maximum delay of 500us and a maximum write delay of 10ms. During this period, when there are read and write requests, other operations that take longer will cause blocking, which increases the command response delay.
  • the host initiates a read-write request, the SSD is performing a time-consuming erase operation.
  • the read-write request needs to be delayed until the block is erased to respond to the read-write request, so there is a large delay; when the host initiates a read At the time of the request, the SSD is performing a long-time-consuming write operation. At this time, the read operation must be delayed until the write operation is completed before responding to the read request, so there is a large delay.
  • the embodiments of the present application provide a method, device, electronic device, and storage medium for improving the random write performance of a hard disk, aiming at solving the problem that the random write performance of the hard disk is low, resulting in unstable IOPS performance of the hard disk.
  • the embodiment of the present application provides a method for random write performance adjustment, including the following steps:
  • the IOPS of the hard disk is calculated based on the second data set, and write arbitration adjustment is performed on the IOPS of the hard disk.
  • the method for adjusting random write performance is to calculate the number of data frames of each storage block in the storage unit by balancing the storage units in the remaining space of the hard disk, and calculate the number of data frames of the hard disk through the number of data frames of each storage block.
  • the IOPS of the hard disk can be judged and analyzed, and the IOPS can be adjusted steadily through the write arbitration control module according to the judgment and analysis results to prevent the fluctuation and imbalance of the IOPS and improve the consistency of the IOPS randomly written to the hard disk.
  • the step of obtaining the remaining space of the hard disk target module, and balancing the storage units of the remaining space includes:
  • the first operation is to write data to the storage unit in the first state
  • a second operation is performed, and the second operation is to write data to the storage unit in the second state.
  • the random write performance adjustment method provided by the present application judges whether the number of storage blocks of the storage unit exceeds a preset threshold, performs different operations according to the judgment result, and selects a storage unit with free storage blocks in the storage unit to perform data writing first. input, so that the speed of writing to the storage block is balanced with the speed of erasing.
  • the step of performing convergence processing on the first data set to obtain the second data set includes:
  • the random write performance adjustment method provided by this application calculates the number of data frames in the stable state of the remaining space based on the state of the storage block, obtains the second sub-data set, removes bad blocks in the storage block, and calculates the remaining storage space.
  • the number of data frames of the block reduces the fluctuation caused by reclaiming new storage blocks, and improves the consistency.
  • the data in the second sub-data set is interpolated to obtain the second data set, so that the storage
  • the fluctuation range of the number of data frames in a block decreases and tends to a steady state.
  • the IOPS of the hard disk is calculated based on the second data set, and the step of performing write arbitration adjustment on the IOPS of the hard disk includes:
  • the IOPS of the hard disk is acquired based on the number of data frames, and the IOPS of the hard disk is adjusted.
  • the random write performance adjustment method provided by this application can obtain the corresponding number of data frames according to the stage state of the remaining space, calculate the IOPS of the hard disk according to the number of data frames, and adjust it according to the IOPS in each stage state.
  • the adjustment is smoother and the consistency is relatively improved.
  • the step of obtaining the stage status of the remaining space includes:
  • the IOPS of the hard disk is calculated based on the first stage, the second stage and the third stage.
  • the IOPS of the hard disk is calculated based on the first stage, the second stage and the third stage, and the corresponding number of data frames can be obtained based on different stages According to the number of data frames, the IOPS of the hard disk can be calculated, and the IOPS of different stages can be adjusted, and the corresponding consistency is also improved.
  • the step of adjusting the IOPS of the hard disk includes:
  • the IOPS of the hard disk is adjusted based on the fluctuation situation.
  • the first fluctuation curve formed by the second data set is also fluctuating Yes, but write arbitration adjusts the number of data frames corresponding to the steady state to form a benchmark fluctuation curve, and compare and analyze the ever-changing first fluctuation curve with the benchmark fluctuation curve to determine the level of IOPS of the hard disk. Therefore, the benchmark fluctuation curve The IOPS is adjusted for the standard, so that the first fluctuation curve tends to the reference fluctuation curve, thereby improving the consistency of hard disk random writing.
  • the step of judging the fluctuation between the first fluctuation curve formed by the second data set and the reference fluctuation curve includes:
  • the random write performance adjustment method provided by this application compares and analyzes the fluctuations of the first fluctuation curve and the reference fluctuation curve to determine the level of IOPS of the hard disk, and then adjusts the IOPS accordingly to improve the random write performance of the hard disk. Consistency, the performance of the hard disk is also relatively improved.
  • performing interpolation processing on the data in the second sub-data set, and obtaining the second data set further includes:
  • the data in the second sub-data set is interpolated to establish a coordinate axis, the horizontal axis is time t, and the vertical axis is the number of data frames y in the remaining space, where (xi, yi) is the number of cached data frames at the previous point, ( xi+1, yi+1) is the latest number of data frames, linear interpolation is performed for (xi, yi) and (xi+1, yi+1), and the (x, y) point is the linear interpolation result, which is obtained based on the linear interpolation result The second data set.
  • the embodiment of the present application also provides a random write performance adjustment device, including:
  • a balance processing module configured to obtain the remaining space of the hard disk target module, and perform balance processing on the storage units of the remaining space
  • a data column acquisition module configured to calculate the number of data frames of each storage block in the storage unit based on the balance processing result to obtain the first data set
  • a convergence module configured to perform convergence processing on the first data set to obtain a second data set
  • the adjustment module calculates the IOPS of the hard disk based on the second data set, and performs write arbitration adjustment on the IOPS of the hard disk.
  • the step of performing convergence processing on the first data set to obtain the second data set includes:
  • the data in the second sub-data set is interpolated to establish a coordinate axis, the horizontal axis is time t, and the vertical axis is the number of data frames y in the remaining space, where (xi, yi) is the number of cached data frames at the previous point, ( xi+1, yi+1) is the latest number of data frames, linear interpolation is performed for (xi, yi) and (xi+1, yi+1), and the (x, y) point is the linear interpolation result, which is obtained based on the linear interpolation result The second data set.
  • the device for adjusting random write performance calculates the number of data frames of each storage block in the storage unit by balancing the storage units in the remaining space of the hard disk, and can calculate the number of data frames of the hard disk through the number of data frames of each storage block.
  • the IOPS of the hard disk can be judged and analyzed, and the IOPS can be adjusted steadily through the write arbitration control module according to the judgment and analysis results to prevent the fluctuation and imbalance of the IOPS and improve the consistency of the IOPS randomly written to the hard disk.
  • the above-mentioned balance processing module is specifically used for:
  • a threshold judging module configured to judge whether the number of storage blocks of the storage unit exceeds a preset threshold
  • a first operation module configured to perform a first operation if the number of storage blocks of the storage unit exceeds a preset threshold, and the first operation is to write data to the storage unit in the first state;
  • the second operation module is configured to execute a second operation if the number of storage blocks of the storage unit does not exceed a preset threshold.
  • the first operation is to write data into the storage unit in the second state.
  • the above-mentioned convergence module is specifically used for:
  • the second sub-data set acquisition module is used to calculate the number of data frames in the stable state of the remaining space based on the state of the storage block, and obtain the second sub-data set;
  • the second data set acquisition module is configured to perform interpolation processing on the data in the second sub-data set to obtain the second data set.
  • the above adjustment module is specifically used for:
  • a status acquisition module configured to acquire the phase status of the remaining space, and the number of data frames in the second data set under the phase status
  • An adjustment module configured to acquire the IOPS of the hard disk based on the number of data frames, and adjust the IOPS of the hard disk.
  • the above-mentioned state acquisition module is specifically used for:
  • the first state acquisition module is used to adjust the number of data frames corresponding to the start state to the number of data frames corresponding to the prohibition state based on the write arbitration, and obtain the first stage;
  • the second stage acquisition module is used to adjust the number of data frames corresponding to the forbidden state to the number of data frames corresponding to the stop state based on the write arbitration, and obtain the second stage;
  • the third-stage acquisition module is configured to adjust the number of data frames corresponding to the stop state based on the write arbitration, and acquire the third stage;
  • a calculating module configured to calculate the IOPS of the hard disk based on the first stage, the second stage and the third stage.
  • the above-mentioned state acquisition module is specifically used for:
  • a reference fluctuation curve acquisition module configured to adjust the number of data frames corresponding to the steady state based on the write arbitration to form a reference fluctuation curve
  • a judgment analysis module configured to judge fluctuations between the first fluctuation curve formed by the second data set and the reference fluctuation curve
  • An adjustment module configured to adjust the IOPS of the hard disk based on the fluctuation.
  • the above judgment and analysis module is specifically used for:
  • a first adjustment module configured to adjust the IOPS of the hard disk to a first preset threshold if the first fluctuation curve is greater than the reference fluctuation curve;
  • a second adjustment module configured to adjust the IOPS of the hard disk to a second preset threshold if the first fluctuation curve is smaller than the reference fluctuation curve.
  • an embodiment of the present application provides an electronic device, including a memory and a processor, the memory and the processor are connected to each other in communication, the memory stores computer instructions, and the processor passes Execute the computer instructions, so as to execute the random write performance adjustment method described in the first aspect or any implementation manner of the first aspect.
  • the embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and the computer instructions are used to make the computer execute the first aspect or any of the first aspects.
  • FIG. 1 is a schematic flowchart of a method for adjusting random write performance provided by an embodiment of the present application
  • FIG. 2 is a functional block diagram of a device for adjusting random write performance provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram of a hardware structure of an electronic device provided by an embodiment of the application.
  • Fig. 4 is a schematic diagram of a computer-readable storage medium provided by an embodiment of the application.
  • the random write performance adjustment method provided by the embodiment of the present application may be executed by a random write performance adjustment device, and the random write performance adjustment device may be realized by software, hardware, or a combination of software and hardware.
  • the computer equipment may be a server or a terminal
  • the server in the embodiment of the present application may be a single server, or a server cluster composed of multiple servers
  • the terminal can be smart phone, personal computer, tablet computer, wearable device, intelligent robot and other intelligent hardware devices.
  • the implementation subject is an electronic device as an example for illustration.
  • a method for adjusting random write performance is provided, and the application of the method and electronic equipment is used as an example to illustrate, including the following steps:
  • the number of data frames of each storage block in the storage unit is calculated based on the balance processing result, the number of idle data frames is counted, the first data set is obtained, and the IOPS of the hard disk can be calculated according to the first data set , which is convenient for subsequent analysis and adjustment of the IOPS of the hard disk.
  • the fluctuation between data frames can be observed according to the first data set. If the fluctuation of the data frame is too large, it means that the IOPS of the hard disk is unstable. Therefore, the convergence process is performed on the first data set so that The fluctuation of the first data set is reduced, and the acquisition of the second data set can make the fluctuation of the data frame tend to a stable state.
  • the IOPS of the hard disk can be calculated according to the second data set, and the IOPS of the hard disk can be visually seen.
  • the IOPS is too high, it needs to be adjusted down to Reduce the consumption speed of the storage block. Conversely, when the IOPS is too high, it needs to be increased to increase the consumption speed of the storage block.
  • the method for adjusting random write performance is to calculate the number of data frames of each storage block in the storage unit by balancing the storage units in the remaining space of the hard disk, and calculate the number of data frames of the hard disk through the number of data frames of each storage block.
  • the IOPS of the hard disk can be judged and analyzed, and the IOPS can be adjusted steadily through the write arbitration control module according to the judgment and analysis results to prevent the fluctuation and imbalance of the IOPS and improve the consistency of the IOPS randomly written to the hard disk.
  • the step of "obtaining the remaining space of the hard disk target module, and performing balancing processing on the storage units of the remaining space" in S100 includes:
  • the first operation is to write data into the storage unit in the second state.
  • a preset threshold which can be set, and is currently set to three.
  • the idle state means that no free storage block is selected from the storage unit for filling in data. Which storage unit is in the idle state, then select the storage unit , if there are multiple storage units in the idle state, select the storage unit with a small number of free storage blocks for data writing.
  • the state judgment of the storage unit is skipped, and a storage unit with a small number of free storage blocks is directly selected, that is, a free storage block is selected from the storage unit for data writing.
  • the random write performance adjustment method provided by the present application judges whether the number of storage blocks of the storage unit exceeds a preset threshold, performs different operations according to the judgment result, and selects a storage unit with free storage blocks in the storage unit to perform data writing first. input, so that the speed of writing to the storage block is balanced with the speed of erasing.
  • the step of "converging on the first data set to obtain the second data set" in S300 above includes:
  • LSTM Long Short-Term Memory Chinese: long-term short-term memory network
  • the number of storage blocks changes at any time during the entire process of writing data
  • B1, B2, B3...Bn represent the change of the number of storage blocks, according to the actual bad block of the storage block
  • the remaining space where the number of remaining data frames is calculated is FB1, FB2, FB3...FBn
  • use avgFB1, avgFB2, avgFB3...avgFBn to correspond to the historical average data frame counting space when Bn, make avgFB1 FB1
  • ⁇ value means that only the most recent FBn is kept for averaging, and the idea of LSTM is used for reference, and the results of each calculation are passed to the next step for interpolation processing.
  • the next step is to use the principle of linear interpolation to interpolate the data in the second sub-dataset, and establish a coordinate axis, the horizontal axis is time t, and the vertical axis is the number of data frames y in the remaining space, where (xi, yi) is the cached data of the previous point Number of frames, (xi+1, yi+1) is the latest data frame number, linear interpolation is performed for (xi, yi) and (xi+1, yi+1), (x, y) point is the result of linear interpolation, this
  • the x value uses the center point of the calculation period
  • y is the average value of the vertical axis values of the two points
  • the ratio of the horizontal axis can also be adjusted according to the consistency result.
  • the curve of the second data set formed is more convergent than the curve of the second sub-data set, but the change trend remains the same. Therefore, using the curve formed by the second data set to control the write arbitration of the hard disk IOPS can control the IOPS more for stability.
  • the random write performance adjustment method provided by this application calculates the number of data frames in the stable state of the remaining space based on the state of the storage block, obtains the second sub-data set, removes bad blocks in the storage block, and calculates the remaining storage space.
  • the number of data frames of the block reduces the fluctuation caused by reclaiming new storage blocks, and improves the consistency.
  • the data in the second sub-data set is interpolated to obtain the second data set, so that the storage
  • the fluctuation range of the number of data frames in a block decreases and tends to a steady state.
  • the step of "calculating the IOPS of the hard disk based on the second data set, and performing write arbitration adjustment on the IOPS of the hard disk" in S400 includes:
  • the random write performance adjustment method provided by this application can obtain the corresponding number of data frames according to the stage state of the remaining space, calculate the IOPS of the hard disk according to the number of data frames, and adjust it according to the IOPS in each stage state.
  • the adjustment is smoother and the consistency is relatively improved.
  • the above step of obtaining the stage status of the remaining space includes:
  • start_credit_num (start state) is the number of data frames corresponding to the free storage block that the write arbitration adjustment starts to intervene
  • steady_num steady state
  • forbid_credit_num (forbidden state) is The number of data frames corresponding to the free storage block that starts to prohibit IOPS
  • stop_credit_num (stop state) is the number of data frames corresponding to the free storage block that reduces IOPS to 0.
  • the adjustment process is divided into three stages.
  • the IOPS of the hard disk is calculated based on the first stage, the second stage and the third stage, and the corresponding number of data frames can be obtained based on different stages According to the number of data frames, the IOPS of the hard disk can be calculated, and the IOPS of different stages can be adjusted, and the corresponding consistency is also improved.
  • the above step of adjusting the IOPS of the hard disk includes:
  • the first fluctuation curve formed by the second data set is also fluctuating Yes, but write arbitration adjusts the number of data frames corresponding to the steady state to form a benchmark fluctuation curve, and compare and analyze the ever-changing first fluctuation curve with the benchmark fluctuation curve to determine the level of IOPS of the hard disk. Therefore, the benchmark fluctuation curve The IOPS is adjusted for the standard, so that the first fluctuation curve tends to the reference fluctuation curve, thereby improving the consistency of hard disk random writing.
  • the step of judging the fluctuation between the first fluctuation curve formed by the second data set and the reference fluctuation curve includes:
  • avgFBn+1 When avgFBn+1 is close to the number of data frames of steady_num, it is determined as a steady state. avgFBn+1 is greater than the set steady_num, indicating that the steady-state situation deviates from our preset situation, then increase the preset IOPS to (1+ ⁇ )*IOPS, here tentatively ⁇ is 5%.
  • Bn is less than the data frame number space of forbid_credit_num, it means that the preset IOPS is too high, and needs to be smaller to reduce the consumption speed of the storage block, and reduce the IOPS to (1- ⁇ )*IOPS, here tentatively ⁇ is 10 %.
  • the consistency of the method is increased from 83% to 93%, and the performance is also improved by about 10%.
  • the details can be analyzed and adjusted according to the actual measurement results.
  • the random write performance adjustment method provided by this application compares and analyzes the fluctuations of the first fluctuation curve and the reference fluctuation curve to determine the level of IOPS of the hard disk, and then adjusts the IOPS accordingly to improve the random write performance of the hard disk. Consistency, the performance of the hard disk is also relatively improved.
  • steps in the flow chart of FIG. 1 are displayed sequentially as indicated by the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in FIG. 1 may include multiple steps or stages, and these steps or stages may not necessarily be executed at the same time, but may be executed at different times, and the execution sequence of these steps or stages may also be It is not necessarily performed sequentially, but may be performed alternately or alternately with other steps or at least a part of steps or stages in other steps.
  • this embodiment provides a device for adjusting random write performance, including a balance processing module 1, a data column acquisition module 2, a convergence module 3 and an adjustment module 4:
  • a balance processing module configured to obtain the remaining space of the hard disk target module, and perform balance processing on the storage units of the remaining space
  • a data column acquisition module 2 configured to calculate the number of data frames of each storage block in the storage unit based on the balance processing result to obtain the first data set
  • Convergence module 3 configured to perform convergence processing on the first data set to obtain a second data set
  • the adjustment module 4 calculates the IOPS of the hard disk based on the second data set, and performs write arbitration adjustment on the IOPS of the hard disk.
  • the device for adjusting random write performance calculates the number of data frames of each storage block in the storage unit by balancing the storage units in the remaining space of the hard disk, and can calculate the number of data frames of the hard disk through the number of data frames of each storage block.
  • the IOPS of the hard disk can be judged and analyzed, and the IOPS can be adjusted steadily through the write arbitration control module according to the judgment and analysis results to prevent the fluctuation and imbalance of the IOPS and improve the consistency of the IOPS randomly written to the hard disk.
  • the above balance processing module is specifically used for:
  • a threshold judging module configured to judge whether the number of storage blocks of the storage unit exceeds a preset threshold
  • a first operation module configured to perform a first operation if the number of storage blocks of the storage unit exceeds a preset threshold, and the first operation is to write data to the storage unit in the first state;
  • the second operation module is configured to execute a second operation if the number of storage blocks of the storage unit does not exceed a preset threshold, and the second operation is to write data into the storage unit in the second state.
  • the above convergence module is specifically used for:
  • the second sub-data set acquisition module is used to calculate the number of data frames in the stable state of the remaining space based on the state of the storage block, and obtain the second sub-data set;
  • the second data set acquisition module is configured to perform interpolation processing on the data in the second sub-data set to obtain the second data set.
  • the above adjustment module is specifically used for:
  • a status acquisition module configured to acquire the phase status of the remaining space, and the number of data frames in the second data set under the phase status
  • An adjustment module configured to acquire the IOPS of the hard disk based on the number of data frames, and adjust the IOPS of the hard disk.
  • the above-mentioned status acquisition module is specifically used for:
  • the first state acquisition module is used to adjust the number of data frames corresponding to the start state to the number of data frames corresponding to the prohibition state based on the write arbitration, and obtain the first stage;
  • the second stage acquisition module is used to adjust the number of data frames corresponding to the forbidden state to the number of data frames corresponding to the stop state based on the write arbitration, and obtain the second stage;
  • the third-stage acquisition module is configured to adjust the number of data frames corresponding to the stop state based on the write arbitration, and acquire the third stage;
  • a calculating module configured to calculate the IOPS of the hard disk based on the first stage, the second stage and the third stage.
  • the above-mentioned status acquisition module is specifically used for:
  • a reference fluctuation curve acquisition module configured to adjust the number of data frames corresponding to the steady state based on the write arbitration to form a reference fluctuation curve
  • a judgment analysis module configured to judge fluctuations between the first fluctuation curve formed by the second data set and the reference fluctuation curve
  • An adjustment module configured to adjust the IOPS of the hard disk based on the fluctuation.
  • the above judgment and analysis module is specifically used for:
  • a first adjustment module configured to adjust the IOPS of the hard disk to a first preset threshold if the first fluctuation curve is greater than the reference fluctuation curve;
  • a second adjustment module configured to adjust the IOPS of the hard disk to a second preset threshold if the first fluctuation curve is smaller than the reference fluctuation curve.
  • Each module in the above device for adjusting random write performance can be fully or partially realized by software, hardware and a combination thereof.
  • the above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, and can also be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the corresponding operations of the above-mentioned modules.
  • An embodiment of the present application also provides an electronic device, which has the device for adjusting random write performance shown in FIG. 2 above.
  • FIG. 3 is a schematic structural diagram of an electronic device provided in an optional embodiment of the present application.
  • the electronic device may include: at least one processor 71, such as a CPU (Central Processing Unit, central processing unit), at least one communication interface 73, memory 74, and at least one communication bus 72.
  • the communication bus 72 is used to realize connection and communication between these components.
  • the communication interface 73 may include a display screen (Display) and a keyboard (Keyboard), and the optional communication interface 73 may also include a standard wired interface and a wireless interface.
  • the memory 74 can be a high-speed RAM memory (Random Access Memory, volatile random access memory), or a non-volatile memory (non-volatile memory), such as at least one disk memory.
  • the memory 74 may also be at least one storage device located away from the aforementioned processor 71 .
  • the processor 71 can be combined with the device, the memory 74 stores an application program, and the processor 71 invokes the program code stored in the memory 74 to execute any of the above method steps.
  • the communication bus 72 may be a peripheral component interconnect (PCI for short) bus or an extended industry standard architecture (EISA for short) bus or the like.
  • PCI peripheral component interconnect
  • EISA extended industry standard architecture
  • the communication bus 72 can be divided into an address bus, a data bus, a control bus, and the like.
  • the memory 74 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as flash memory (English: flash memory), hard disk (English: hard disk drive, abbreviated: HDD) or solid-state hard drive (English: solid-state drive, abbreviated: SSD); memory 74 can also include the above-mentioned types combination of memory.
  • volatile memory such as a random access memory (English: random-access memory, abbreviated: RAM)
  • non-volatile memory such as flash memory (English: flash memory), hard disk (English: hard disk drive, abbreviated: HDD) or solid-state hard drive (English: solid-state drive, abbreviated: SSD); memory 74 can also include the above-mentioned types combination of memory.
  • the processor 71 may be a central processing unit (English: central processing unit, abbreviated: CPU), a network processor (English: network processor, abbreviated: NP) or a combination of CPU and NP.
  • CPU central processing unit
  • NP network processor
  • the processor 71 may also optionally include a hardware chip.
  • the aforementioned hardware chip may be an application-specific integrated circuit (English: application-specific integrated circuit, abbreviation: ASIC), a programmable logic device (English: programmable logic device, abbreviation: PLD) or a combination thereof.
  • the above PLD can be a complex programmable logic device (English: complex programmable logic device, abbreviated: CPLD), field programmable logic gate array (English: field-programmable gate array, abbreviated: FPGA), general array logic (English: generic array logic, abbreviation: GAL) or any combination thereof.
  • memory 74 is also used to store program instructions.
  • the processor 71 may invoke program instructions to implement the random write performance adjustment method shown in the embodiment of FIG. 1 of the present application.
  • Fig. 4 is a schematic diagram of the computer-readable storage medium provided by the embodiment of the application. As shown in Fig. 4, the embodiment of the application also provides a non-transitory computer storage medium 601, the computer storage medium 601 stores Executable instructions 610, the computer-executable instructions 610 may implement the random write performance adjustment method in any of the foregoing method embodiments.
  • the storage medium 601 can be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a flash memory (Flash Memory), a hard disk ( Hard Disk Drive, abbreviation: HDD) or solid-state hard drive (Solid-State Drive, SSD), etc.; the storage medium 601 may also include a combination of the above-mentioned types of memory.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)
  • Techniques For Improving Reliability Of Storages (AREA)

Abstract

本申请公开了一种随机写性能调节的方法、装置、电子设备及存储介质，所述方法包括获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理；基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集；对所述第一数据集进行收敛处理，得到第二数据集；基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。本申请通过写仲裁控制模块对IOPS进行稳态调整，防止IOPS的波动不平衡，提高硬盘随机写的IOPS的一致性。

Description

随机写性能调节的方法、装置、电子设备及存储介质
本申请要求在2021年09月29日提交中国专利局、申请号为202111147105.6、发明名称为“随机写性能调节的方法、装置、电子设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及硬盘控制技术领域，具体涉及一种随机写性能调节的方法、装置、电子设备及存储介质。
背景技术
在数据量指数级增长的信息化时代，快速有效的存储和处理数据成为了研究的重要方向。机械硬盘的速度已经无法满足目前CPU的计算速度，固态硬盘以读写速度快、体积小的优势逐渐成为了主流存储器件，广泛应用到生活中的各个领域，固态硬盘的性能也随之受到了重视，由于固态硬盘中NAND的原理决定了，存在如下的特性和限制：典型的，一个Block擦除需要十几毫秒；写一个Block数据大约需要几毫秒；而读一个Block的数据仅仅只需要几十微秒；可选地，同一时间内，单个DIE只能执行一种类型的操作。随着消费类领域的固态硬盘的广泛使用，由于固态硬盘的随机访问性能高，功耗低，环境稳定性强等特点，在企业级也逐步引入了固态硬盘。与消费类固态硬盘更关注峰值性能不同的是，企业级更关心性能的一致性，典型如都延迟最大在500us内，写延迟最大在10ms内。在此期间，当有读写请求时，会因为其他耗时更长的操作引发阻塞，增加了命令响应延迟。当主机发起读写请求时，固态硬盘内部正执行耗时较长的擦除操作，这时读写请求需要延迟到Block擦除才能响应读写请求，因此存在较大的延迟；当主机发起读请求时，固态硬盘内部正在执行耗时较长的写操作，这时读操作必须要延迟到写操作完成才能响应读请求，因此存在较大的延迟。
发明内容
有鉴于此，本申请实施例提供了一种提高硬盘随机写性能的方法、装置、电子设备及存储介质，旨在解决硬盘的随机写性能低，导致硬盘的IOPS性能不稳定的问题。
根据第一方面，本申请实施例提供了一种随机写性能调节的方法，包括如下步骤：
获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理；
基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集；
对所述第一数据集进行收敛处理，得到第二数据集；
基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。
本申请提供的随机写性能调节的方法，通过对硬盘的剩余空间的存储单元进行平衡处 理，计算所述存储单元中各个存储块的数据帧数量，通过各个存储块的数据帧数量可计算出硬盘的IOPS，进而可对硬盘的IOPS进行判断分析，根据判断分析结果通过写仲裁控制模块对IOPS进行稳态调整，防止IOPS的波动不平衡，提高硬盘随机写的IOPS的一致性。
结合第一方面，在第一方面第一实施方式中，获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理步骤中，包括：
判断所述存储单元的所述存储块数量是否超过预设阈值；
若所述存储单元的所述存储块数量超过所述预设阈值，则执行第一操作，所述第一操作为对第一状态的存储单元进行数据写入；
若所述存储单元的所述存储块数量未超过所述预设阈值，则执行第二操作，所述第二操作为对第二状态的存储单元进行数据写入。
本申请提供的随机写性能调节的方法，判断所述存储单元的所述存储块数量是否超过预设阈值，针对判断结果执行不同的操作，选取存储单元中存储块空闲的存储单元先进行数据写入，使得对存储块写入的速度与擦除的速度平衡。
结合第一方面，在第一方面第二实施方式中，对所述第一数据集进行收敛处理，得到第二数据集步骤中，包括：
基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集；
对所述第二子数据集中的数据进行插值处理，得到所述第二数据集。
本申请提供的随机写性能调节的方法，基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集，去除存储块中的坏块，计算出剩余存储块的数据帧数量，减少因为回收出新的存储块而带来的波动，提高了一致性，再对所述第二子数据集中的数据进行插值处理，得到所述第二数据集，使得存储块中的数据帧数量波动范围减小，趋于稳态。
结合第一方面，在第一方面第三实施方式中，基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整步骤中，包括：
获取所述剩余空间的阶段状态，以及所述阶段状态下所述第二数据集中的所述数据帧数量；
基于所述数据帧数量获取所述硬盘的IOPS，对所述硬盘的IOPS进行调整。
本申请提供的随机写性能调节的方法，根据剩余空间的阶段状态可以获取到对应的数据帧数量，根据数据帧数量可计算出硬盘的IOPS，针对每个阶段状态下的IOPS对其进行调整，调整时比较平滑，一致性也相对提高。
结合第一方面第三实施方式，在第一方面第四实施方式中，获取所述剩余空间的阶段状态步骤中，包括：
基于写仲裁调整开始状态对应的数据帧数量到禁止状态对应的数据帧数量，获取第一阶段；
基于所述写仲裁调整禁止状态对应的数据帧数量到停止状态对应的数据帧数量，获取第二阶段；
基于所述写仲裁调整停止状态对应的数据帧数量，获取第三阶段；
基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS。
本申请提供的随机写性能调节的方法，根据基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS，基于不同的阶段可获取到相应的数据帧数量，根据数据帧数量可计算出硬盘的IOPS，针对不同阶段的IOPS进行调整，相应的一致性也提高了。
结合第一方面第四实施方式，在第一方面第五实施方式中，对所述硬盘的IOPS进行调整步骤中，包括：
基于所述写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线；
判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况；
基于所述波动情况对所述硬盘的IOPS进行调整。
本申请提供的随机写性能调节的方法，由于存储块中的数据帧的数量是随着硬盘的写入和擦除操作而在不断变化的，所以第二数据集形成的第一波动曲线也是波动的，但是写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线，将不断变化的第一波动曲线与基准波动曲线进行比较分析，可判断硬盘的IOPS的高低，因此以基准波动曲线为标准对IOPS进行调整，使得第一波动曲线趋于基准波动曲线，进而提高了硬盘随机写的一致性。
结合第一方面第五实施方式，在第一方面第六实施方式中，判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况步骤中，包括：
若所述第一波动曲线大于所述基准波动曲线，则调整所述硬盘的IOPS为第一预设阈值；
若所述第一波动曲线小于所述基准波动曲线，则调整所述硬盘的IOPS为第二预设阈值。
本申请提供的随机写性能调节的方法，根据第一波动曲线与基准波动曲线的波动情况进行对比分析，可判断出硬盘的IOPS的高低，进而对IOPS做相应的调整，提高了硬盘随机写的一致性，硬盘的性能也相对提高。
结合第一方面的第二实施方式，在第一方面第七实施方式中，对所述第二子数据集中的数据进行插值处理，得到所述第二数据集还包括：
对所述第二子数据集中的数据进行插值处理，建立坐标轴，横轴为时间t，纵轴为剩余空间的数据帧数目y，其中(xi，yi)是上一点缓存数据帧数目，(xi+1，yi+1)是最新数据帧数目，对于(xi，yi)和(xi+1，yi+1)进行线性插值，(x，y)点是线性插值结果，基于线性插值结果得到所述第二数据集。
根据第二方面，本申请实施例还提供了一种随机写性能调节的装置，包括：
平衡处理模块，用于获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理；
数据列获取模块，用于基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集；
收敛模块，用于对所述第一数据集进行收敛处理，得到第二数据集；
调整模块，基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。
结合第二方面，在第二方面第八实施方式中，所述对所述第一数据集进行收敛处理，得到第二数据集步骤中，包括：
基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据 集；
对所述第二子数据集中的数据进行插值处理，建立坐标轴，横轴为时间t，纵轴为剩余空间的数据帧数目y，其中(xi，yi)是上一点缓存数据帧数目，(xi+1，yi+1)是最新数据帧数目，对于(xi，yi)和(xi+1，yi+1)进行线性插值，(x，y)点是线性插值结果，基于线性插值结果得到所述第二数据集。
本申请提供的随机写性能调节的装置，通过对硬盘的剩余空间的存储单元进行平衡处理，计算所述存储单元中各个存储块的数据帧数量，通过各个存储块的数据帧数量可计算出硬盘的IOPS，进而可对硬盘的IOPS进行判断分析，根据判断分析结果通过写仲裁控制模块对IOPS进行稳态调整，防止IOPS的波动不平衡，提高硬盘随机写的IOPS的一致性。
结合第二方面，在第二方面第一实施方式中，上述平衡处理模块，具体用于：
阈值判断模块，用于判断所述存储单元的所述存储块数量是否超过预设阈值；
第一操作模块，用于若所述存储单元的所述存储块数量超过预设阈值，则执行第一操作，第一操作即对第一状态的存储单元进行数据写入；
第二操作模块，用于若所述存储单元的所述存储块数量未超过预设阈值，则执行第二操作，第一操作即对第二状态的存储单元进行数据写入。
结合第二方面，在第二方面第二实施方式中，上述收敛模块，具体用于：
第二子数据集获取模块，用于基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集；
第二数据集获取模块，用于对所述第二子数据集中的数据进行插值处理，得到所述第二数据集。
结合第二方面，在第二方面第三实施方式中，上述调整模块，具体用于：
状态获取模块，用于获取所述剩余空间的阶段状态，以及所述阶段状态下所述第二数据集中的所述数据帧数量；
调整模块，用于基于所述数据帧数量获取所述硬盘的IOPS，对所述硬盘的IOPS进行调整。
结合第二方面第三实施方式，在第二方面第四实施方式中，上述状态获取模块，具体用于：
第一状态获取模块，用于基于写仲裁调整开始状态对应的数据帧数量到禁止状态对应的数据帧数量，获取第一阶段；
第二阶段获取模块，用于基于所述写仲裁调整禁止状态对应的数据帧数量到停止状态对应的数据帧数量，获取第二阶段；
第三阶段获取模块，用于基于所述写仲裁调整停止状态对应的数据帧数量，获取第三阶段；
计算模块，用于基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS。
结合第二方面第四实施方式，在第二方面第五实施方式中，上述状态获取模块，具体用于：
基准波动曲线获取模块，用于基于所述写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线；
判断分析模块，用于判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况；
调整模块，用于基于所述波动情况对所述硬盘的IOPS进行调整。
结合第二方面第五实施方式，在第二方面第六实施方式中，上述判断分析模块，具体用于：
第一调整模块，用于若所述第一波动曲线大于所述基准波动曲线，则调整所述硬盘的IOPS为第一预设阈值；
第二调整模块，用于若所述第一波动曲线小于所述基准波动曲线，则调整所述硬盘的IOPS为第二预设阈值。
根据第三方面，本申请实施例提供了一种电子设备，包括存储器和处理器，所述存储器和所述处理器之间互相通信连接，所述存储器中存储有计算机指令，所述处理器通过执行所述计算机指令，从而执行第一方面或者第一方面的任意一种实施方式中所述的随机写性能调节的方法。
根据第四方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质存储计算机指令，所述计算机指令用于使所述计算机执行第一方面或者第一方面的任意一种实施方式中所述的随机写性能调节的方法。
附图说明
为了更清楚地说明本申请具体实施方式或现有技术中的技术方案，下面将对具体实施方式或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施方式，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是应用本申请实施例提供的随机写性能调节的方法的流程示意图；
图2是应用本申请实施例提供的随机写性能调节的装置的功能模块图；
图3是应用本申请实施例提供的电子设备的硬件结构示意图；
图4是应用本申请实施例提供的计算机可读存储介质的示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
需要说明的是，本申请实施例提供的随机写性能调节的方法，其执行主体可以是随机写性能调节的装置，该随机写性能调节的装置可以通过软件、硬件或者软硬件结合的方式实现成为计算机设备的部分或者全部，其中，该计算机设备可以是服务器或者终端，其中，本申请实施例中的服务器可以为一台服务器，也可以为由多台服务器组成的服务器集群，本申请实施例中的终端可以是智能手机、个人电脑、平板电脑、可穿戴设备以及智能机器人等其他智能硬件设备。下述方法实施例中，均以执行主体是电子设备为例来进行说明。
在本申请一个实施例中，如图1所示，提供了一种随机写性能调节的方法，以该方法 应用与电子设备为例进行说明，包括以下步骤：
S100，获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理。
在数据量指数级增长的信息化时代，快速有效的存储和处理数据成为了研究的重要方向。机械硬盘的速度已经无法满足目前CPU的计算速度，固态硬盘以读写速度快、体积小的优势逐渐成为了主流存储器件，广泛应用到生活中的各个领域，因而硬盘的性能也随之受到了重视，但是由于硬盘的写入与擦除的速率不一致，若写入的速度较快，而擦除的速度较慢，那么硬盘中的存储块较少，当硬盘的IOPS(英文：Input/Output Operations Per Second每秒的读写次数)过大时，则会影响硬盘的性能，同样的当写入的速度较慢，而擦除的速度较快，当硬盘的IOPS过小时，也会影响硬盘的数据传输速率，因此需要根据存储块的中的数据帧的状态进行平衡调整，进而确定从那一个存储块中先进行写入操作。
S200，基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集。
在本实施例中，基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，统计出空闲的数据帧数量，获取到第一数据集，根据第一数据集可计算出硬盘的IOPS，便于后续对硬盘的IOPS进行分析调整。
S300，对所述第一数据集进行收敛处理，得到第二数据集。
在本实施例中，根据第一数据集可观测出数据帧之间的波动情况，若数据帧波动的幅度太大，说明根据硬盘的IOPS不稳定，因此对第一数据集进行收敛处理，使得第一数据集的波动减小，获取到第二数据集可以使得数据帧的波动趋于稳定状态。
S400，基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。
在本实施例中，获取到收敛后的第二数据集后，根据第二数据集可计算出硬盘的IOPS，可直观的看到硬盘的IOPS的情况，当IOPS过高时，需要调小以降低存储块的消耗速度，反之当IOPS过高时，需要调大以提高存储块的消耗速度，
本申请提供的随机写性能调节的方法，通过对硬盘的剩余空间的存储单元进行平衡处理，计算所述存储单元中各个存储块的数据帧数量，通过各个存储块的数据帧数量可计算出硬盘的IOPS，进而可对硬盘的IOPS进行判断分析，根据判断分析结果通过写仲裁控制模块对IOPS进行稳态调整，防止IOPS的波动不平衡，提高硬盘随机写的IOPS的一致性。
在本申请一个可选的实施例中，上述S100中的“获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理”步骤中，包括：
(1)判断所述存储单元的所述存储块数量是否超过预设阈值；
(2)若所述存储单元的所述存储块数量超过预设阈值，则执行第一操作，第一操作即对第一状态的存储单元进行数据写入；
(3)若所述存储单元的所述存储块数量未超过预设阈值，则执行第二操作，第一操作即对第二状态的存储单元进行数据写入。
在本实施例中，首先判断空闲存储块的数目最少的存储单元和存储块的数目最多的存储单元是否超过预设阈值，该预设阈值可以设定，当前设定为3个。
如果没有超过该预设阈值，则继续判断存储单元是否处于空闲状态，空闲状态即为暂时没有从该存储单元上选取空闲存储块用来填写数据，哪个存储单元处于空闲状态，则选 择该存储单元，如果有多个存储单元处于空闲状态，则选择空闲存储块数少的存储单元进行数据写入。
如果超过该预设阈值，则跳过对存储单元的状态判断，直接选择空闲存储块数目少的存储单元，即从该存储单元中选取空闲存储块进行数据写入。
如此可以平衡硬盘写入数据和擦除数据的速率，提高了效率。
本申请提供的随机写性能调节的方法，判断所述存储单元的所述存储块数量是否超过预设阈值，针对判断结果执行不同的操作，选取存储单元中存储块空闲的存储单元先进行数据写入，使得对存储块写入的速度与擦除的速度平衡。
在本申请一个可选的实施例中，上述S300中的“对所述第一数据集进行收敛处理，得到第二数据集”步骤中，包括：
(1)基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集；
(2)对所述第二子数据集中的数据进行插值处理，得到所述第二数据集。
当获取到第一数据集后，借鉴LSTM(英文：Long Short-Term Memory中文：长短期记忆网络)的思想计算出新的数据帧数，即得到第二子数据集，减少目标模块因为回收出新的存储块而带来的波动，提高了一致性。
在计算稳态时的剩余空间中，在整个写入数据过程中，存储块数目是随时变化的，令B1，B2，B3…Bn代表存储块数目变化的情况，根据该存储块的实际坏块情况，计算剩余的数据帧数量所在的剩余空间为FB1，FB2，FB3…FBn，使用avgFB1，avgFB2，avgFB3…avgFBn来对应Bn时的历史平均数据帧计数空间，令avgFB1＝FB1，则：
avgFBn+1＝(βavgFBn+FBn+1)/(β+1)      公式1
其中β值的含义是指只保留最近多次的FBn来进行平均，借鉴了LSTM的思想，将每次计算完的结果传给下一步插值处理。
具体如何确定β，我们在实际获取了在随机写从调整到稳态过程的空闲数据帧空间数据，通过计算获得β值使得空闲数据帧数据在稳态过程中经过上述公式计算后波动较小，与实际的测试情况基本相符，经过对比β值在0.9的时候avgFBn+1波动曲线接近原图并且波动范围小，经过公式1计算后的稳态中avgFBn+1的变化很小。
下一步利用线性插值原理对第二子数据集中的数据进行插值处理，建立坐标轴，横轴为时间t，纵轴为剩余空间的数据帧数目y，其中(xi，yi)是上一点缓存数据帧数目，(xi+1，yi+1)是最新数据帧数目，对于(xi，yi)和(xi+1，yi+1)进行线性插值，(x，y)点是线性插值结果，本申请中x值使用了在计算周期的中心点，则y为两点纵轴值的平均值，也可以根据一致性结果对横轴的比例进行调节，
则根据插值原理，可以得出如下公式计算新的数据帧数目。
Figure PCTCN2022074598-appb-000001
每次缓存新数据后，利用上述公式计算，因为计算周期为5ms是固定的，又因为x为周期中心点，所以y为前后两点的平均值。
形成的第二数据集的曲线比第二子数据集的曲线更为收敛，但是变化趋势保持一致，所以利用第二数据集形成的曲线对硬盘的IOPS进行写仲裁控制，可以将IOPS控制的更为稳 定。
本申请提供的随机写性能调节的方法，基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集，去除存储块中的坏块，计算出剩余存储块的数据帧数量，减少因为回收出新的存储块而带来的波动，提高了一致性，再对所述第二子数据集中的数据进行插值处理，得到所述第二数据集，使得存储块中的数据帧数量波动范围减小，趋于稳态。
在本申请一个可选的实施例中，上述S400中“基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整”步骤中，包括：
(1)获取所述剩余空间的阶段状态，以及所述阶段状态下所述第二数据集中的所述数据帧数量；
(2)基于所述数据帧数量获取所述硬盘的IOPS，对所述硬盘的IOPS进行调整。
本申请提供的随机写性能调节的方法，根据剩余空间的阶段状态可以获取到对应的数据帧数量，根据数据帧数量可计算出硬盘的IOPS，针对每个阶段状态下的IOPS对其进行调整，调整时比较平滑，一致性也相对提高。
在本申请一个可选的实施例中，上述获取所述剩余空间的阶段状态步骤中，包括：
(1)基于写仲裁调整开始状态对应的数据帧数量到禁止状态对应的数据帧数量，获取第一阶段；
(2)基于所述写仲裁调整禁止状态对应的数据帧数量到停止状态对应的数据帧数量，获取第二阶段；
(3)基于所述写仲裁调整停止状态对应的数据帧数量，获取第三阶段；
(4)基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS。
start_credit_num(开始状态)的含义是写仲裁调整开始介入的空闲存储块对应数据帧数目，steady_num(稳定状态)是写仲裁调整下稳态时的空闲存储块对应数据帧数目，forbid_credit_num(禁止状态)是开始禁止IOPS的空闲存储块对应数据帧数目，stop_credit_num(停止状态)是将IOPS降为0的空闲存储块对应数据帧数目。
举例说明，调整的过程分为三个阶段，第一个阶段是当第二步中计算的数据帧数目avgFBn+1在start_credit_num＝144对应数据帧数目到forbid_credit_num＝24对应数据帧数目之间，第二个阶段为forbid_credit_num＝24对应数据帧数目到stop_credit_num＝18对应数据帧数目之间，第三阶段为stop_credit_num＝18对应数据帧数目以下。
本申请提供的随机写性能调节的方法，根据基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS，基于不同的阶段可获取到相应的数据帧数量，根据数据帧数量可计算出硬盘的IOPS，针对不同阶段的IOPS进行调整，相应的一致性也提高了。
在本申请一个可选的实施例中，上述对所述硬盘的IOPS进行调整步骤中，包括：
(1)基于所述写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线；
(2)判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况；
(3)基于所述波动情况对所述硬盘的IOPS进行调整。
上述对所述硬盘的IOPS进行调整步骤中计算得到的IOPS后，需要根据公式1计算后的稳态中avgFBn+1与steady_num(稳定状态)数量的数据帧数目比较进行动态调整。
本申请提供的随机写性能调节的方法，由于存储块中的数据帧的数量是随着硬盘的写 入和擦除操作而在不断变化的，所以第二数据集形成的第一波动曲线也是波动的，但是写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线，将不断变化的第一波动曲线与基准波动曲线进行比较分析，可判断硬盘的IOPS的高低，因此以基准波动曲线为标准对IOPS进行调整，使得第一波动曲线趋于基准波动曲线，进而提高了硬盘随机写的一致性。
在本申请一个可选的实施例中，上述判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况步骤中，包括：
若所述第一波动曲线大于所述基准波动曲线，则调整所述硬盘的IOPS为第一预设阈值；
若所述第一波动曲线小于所述基准波动曲线，则调整所述硬盘的IOPS为第二预设阈值。
avgFBn+1接近steady_num数量的数据帧数目时,判定为稳态。avgFBn+1大于设定的steady_num,说明稳态情况离我们的预设情况有偏差,则提升预设的IOPS为(1+δ)*IOPS,这里暂定δ为5％。当Bn小于了forbid_credit_num的数据帧数目空间时,说明预设的IOPS过高了,需要较小以降低存储块的消耗速度,将IOPS降低为(1-ζ)*IOPS,这里暂定ζ为10％。
经实际测验，该方法实现后的一致性从83％提高为93％，性能也提高了10％左右。具体可根据实际测算结果进行分析调整。
本申请提供的随机写性能调节的方法，根据第一波动曲线与基准波动曲线的波动情况进行对比分析，可判断出硬盘的IOPS的高低，进而对IOPS做相应的调整，提高了硬盘随机写的一致性，硬盘的性能也相对提高。
应该理解的是，虽然图1的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图1中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。
如图2所述，本实施例提供一种随机写性能调节的装置，包括平衡处理模块1、数据列获取模块2、收敛模块3和调整模块4：
平衡处理模块1，用于获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理；
数据列获取模块2，用于基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集；
收敛模块3，用于对所述第一数据集进行收敛处理，得到第二数据集；
调整模块4，基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。
本申请提供的随机写性能调节的装置，通过对硬盘的剩余空间的存储单元进行平衡处理，计算所述存储单元中各个存储块的数据帧数量，通过各个存储块的数据帧数量可计算出硬盘的IOPS，进而可对硬盘的IOPS进行判断分析，根据判断分析结果通过写仲裁控制模块对IOPS进行稳态调整，防止IOPS的波动不平衡，提高硬盘随机写的IOPS的一致性。
在本申请一个实施例中，上述平衡处理模块，具体用于：
阈值判断模块，用于判断所述存储单元的所述存储块数量是否超过预设阈值；
第一操作模块，用于若所述存储单元的所述存储块数量超过预设阈值，则执行第一操作，第一操作即对第一状态的存储单元进行数据写入；
第二操作模块，用于若所述存储单元的所述存储块数量未超过预设阈值，则执行第二操作，第二操作即对第二状态的存储单元进行数据写入。
在本申请一个实施例中，上述收敛模块，具体用于：
第二子数据集获取模块，用于基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集；
第二数据集获取模块，用于对所述第二子数据集中的数据进行插值处理，得到所述第二数据集。
在本申请一个实施例中，上述调整模块，具体用于：
状态获取模块，用于获取所述剩余空间的阶段状态，以及所述阶段状态下所述第二数据集中的所述数据帧数量；
调整模块，用于基于所述数据帧数量获取所述硬盘的IOPS，对所述硬盘的IOPS进行调整。
在本申请一个实施例中，上述状态获取模块，具体用于：
第一状态获取模块，用于基于写仲裁调整开始状态对应的数据帧数量到禁止状态对应的数据帧数量，获取第一阶段；
第二阶段获取模块，用于基于所述写仲裁调整禁止状态对应的数据帧数量到停止状态对应的数据帧数量，获取第二阶段；
第三阶段获取模块，用于基于所述写仲裁调整停止状态对应的数据帧数量，获取第三阶段；
计算模块，用于基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS。
在本申请一个实施例中，上述状态获取模块，具体用于：
基准波动曲线获取模块，用于基于所述写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线；
判断分析模块，用于判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况；
调整模块，用于基于所述波动情况对所述硬盘的IOPS进行调整。
在本申请一个实施例中，上述判断分析模块，具体用于：
第一调整模块，用于若所述第一波动曲线大于所述基准波动曲线，则调整所述硬盘的IOPS为第一预设阈值；
第二调整模块，用于若所述第一波动曲线小于所述基准波动曲线，则调整所述硬盘的IOPS为第二预设阈值。
关于随机写性能调节的装置的具体限定以及有益效果可以参见上文中对于随机写性能调节的方法的限定，在此不再赘述。上述随机写性能调节的装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于电子设备中的处理器中，也可以以软件形式存储于电子设备中的存储器中，以便于处理器调用执行 以上各个模块对应的操作。
本申请实施例还提供一种电子设备，具有上述图2所示的随机写性能调节的装置。
如图3所示，图3是本申请可选实施例提供的一种电子设备的结构示意图，如图3所示，该电子设备可以包括：至少一个处理器71，例如CPU(Central Processing Unit，中央处理器)，至少一个通信接口73，存储器74，至少一个通信总线72。其中，通信总线72用于实现这些组件之间的连接通信。其中，通信接口73可以包括显示屏(Display)、键盘(Keyboard)，可选通信接口73还可以包括标准的有线接口、无线接口。存储器74可以是高速RAM存储器(Random Access Memory，易挥发性随机存取存储器)，也可以是非不稳定的存储器(non-volatile memory)，例如至少一个磁盘存储器。存储器74可选的还可以是至少一个位于远离前述处理器71的存储装置。其中处理器71可以结合装置，存储器74中存储应用程序，且处理器71调用存储器74中存储的程序代码，以用于执行上述任一方法步骤。
其中，通信总线72可以是外设部件互连标准(peripheral component interconnect，简称PCI)总线或扩展工业标准结构(extended industry standard architecture，简称EISA)总线等。通信总线72可以分为地址总线、数据总线、控制总线等。
其中，存储器74可以包括易失性存储器(英文：volatile memory)，例如随机存取存储器(英文：random-access memory，缩写：RAM)；存储器也可以包括非易失性存储器(英文：non-volatile memory)，例如快闪存储器(英文：flash memory)，硬盘(英文：hard disk drive，缩写：HDD)或固态硬盘(英文：solid-state drive，缩写：SSD)；存储器74还可以包括上述种类的存储器的组合。
其中，处理器71可以是中央处理器(英文：central processing unit，缩写：CPU)，网络处理器(英文：network processor，缩写：NP)或者CPU和NP的组合。
其中，处理器71还可以可选地包括硬件芯片。上述硬件芯片可以是专用集成电路(英文：application-specific integrated circuit，缩写：ASIC)，可编程逻辑器件(英文：programmable logic device，缩写：PLD)或其组合。上述PLD可以是复杂可编程逻辑器件(英文：complex programmable logic device，缩写：CPLD)，现场可编程逻辑门阵列(英文：field-programmable gate array，缩写：FPGA)，通用阵列逻辑(英文：generic array logic,缩写：GAL)或其任意组合。
可选地，存储器74还用于存储程序指令。处理器71可以调用程序指令，实现如本申请图1实施例中所示的随机写性能调节的方法。
图4是应用本申请实施例提供的计算机可读存储介质的示意图，如图4所示，本申请实施例还提供了一种非暂态计算机存储介质601，所述计算机存储介质601存储有计算机可执行指令610，该计算机可执行指令610可执行上述任意方法实施例中的随机写性能调节的方法。其中，所述存储介质601可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)、随机存储记忆体(Random Access Memory，RAM)、快闪存储器(Flash Memory)、硬盘(Hard Disk Drive，缩写：HDD)或固态硬盘(Solid-State Drive，SSD)等；所述存储介质601还可以包括上述种类的存储器的组合。
虽然结合附图描述了本申请的实施例，但是本领域技术人员可以在不脱离本申请的精神和范围的情况下做出各种修改和变型，这样的修改和变型均落入由所附权利要求所限定的范围之内。

Claims (12)

  1. 一种随机写性能调节的方法，其特征在于，包括如下步骤：
    获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理；
    基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集；
    对所述第一数据集进行收敛处理，得到第二数据集；
    基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。
  2. 根据权利要求1所述的随机写性能调节的方法，其特征在于，获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理步骤中，包括：
    判断所述存储单元的所述存储块数量是否超过预设阈值；
    若所述存储单元的所述存储块数量超过所述预设阈值，则执行第一操作，所述第一操作为对第一状态的存储单元进行数据写入；
    若所述存储单元的所述存储块数量未超过所述预设阈值，则执行第二操作，所述第二操作为对第二状态的存储单元进行数据写入。
  3. 根据权利要求1所述的随机写性能调节的方法，其特征在于，所述对所述第一数据集进行收敛处理，得到第二数据集步骤中，包括：
    基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集；
    对所述第二子数据集中的数据进行插值处理，得到所述第二数据集。
  4. 根据权利要求1所述的随机写性能调节的方法，其特征在于，所述基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整步骤中，包括：
    获取所述剩余空间的阶段状态，以及所述阶段状态下所述第二数据集中的所述数据帧数量；
    基于所述数据帧数量获取所述硬盘的IOPS，对所述硬盘的IOPS进行调整。
  5. 根据权利要求4所述的随机写性能调节的方法，其特征在于，所述获取所述剩余空间的阶段状态步骤中，包括：
    基于写仲裁调整开始状态对应的数据帧数量到禁止状态对应的数据帧数量，获取第一阶段；
    基于所述写仲裁调整禁止状态对应的数据帧数量到停止状态对应的数据帧数量，获取第二阶段；
    基于所述写仲裁调整停止状态对应的数据帧数量，获取第三阶段；
    基于所述第一阶段、所述第二阶段和所述第三阶段计算出所述硬盘的IOPS。
  6. 根据权利要求5所述的随机写性能调节的方法，其特征在于，所述对所述硬盘的IOPS进行调整步骤中，包括：
    基于所述写仲裁调整下稳态时对应的数据帧数量，形成基准波动曲线；
    判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况；
    基于所述波动情况对所述硬盘的IOPS进行调整。
  7. 根据权利要求6所述的随机写性能调节的方法，其特征在于，所述判断所述第二数据集形成的第一波动曲线与所述基准波动曲线的波动情况步骤中，包括：
    若所述第一波动曲线大于所述基准波动曲线，则调整所述硬盘的IOPS为第一预设阈值；
    若所述第一波动曲线小于所述基准波动曲线，则调整所述硬盘的IOPS为第二预设阈值。
  8. 根据权利要求3所述的随机写性能调节的方法，其特征在于，对所述第二子数据集中的数据进行插值处理，得到所述第二数据集还包括：
    对所述第二子数据集中的数据进行插值处理，建立坐标轴，横轴为时间t，纵轴为剩余空间的数据帧数目y，其中(xi，yi)是上一点缓存数据帧数目，(xi+1，yi+1)是最新数据帧数目，对于(xi，yi)和(xi+1，yi+1)进行线性插值，(x，y)点是线性插值结果，基于线性插值结果得到所述第二数据集。
  9. 一种随机写性能调节的装置，其特征在于，包括：
    平衡处理模块，用于获取硬盘目标模块的剩余空间，对所述剩余空间的存储单元进行平衡处理；
    数据列获取模块，用于基于平衡处理结果计算所述存储单元中各个存储块的数据帧数量，得到第一数据集；
    收敛模块，用于对所述第一数据集进行收敛处理，得到第二数据集；
    调整模块，基于所述第二数据集计算出所述硬盘的IOPS，对所述硬盘的IOPS进行写仲裁调整。
  10. 根据权利要求9所述的随机写性能调节的装置，其特征在于，所述对所述第一数据集进行收敛处理，得到第二数据集步骤中，包括：
    基于所述存储块的状态计算所述剩余空间稳定状态下的数据帧数量，得到第二子数据集；
    对所述第二子数据集中的数据进行插值处理，建立坐标轴，横轴为时间t，纵轴为剩余空间的数据帧数目y，其中(xi，yi)是上一点缓存数据帧数目，(xi+1，yi+1)是最新数据帧数目，对于(xi，yi)和(xi+1，yi+1)进行线性插值，(x，y)点是线性插值结果，基于线性插值结果得到所述第二数据集。
  11. 一种电子设备，其特征在于，包括存储器和处理器，所述存储器中存储有计算机指令，所述处理器通过执行所述计算机指令，从而执行权利要求1至8中任一项所述的随机写性能调节的方法。
  12. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机指令，所述计算机指令用于使计算机执行权利要求1至8中任一项所述的随机写性能调节的方法。
PCT/CN2022/074598 2021-09-29 2022-01-28 随机写性能调节的方法、装置、电子设备及存储介质 WO2023050672A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/564,565 US20240377951A1 (en) 2021-09-29 2022-01-28 Method and apparatus for adjusting random write performance, electronic device, and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111147105.6A CN113590050B (zh) 2021-09-29 2021-09-29 随机写性能调节的方法、装置、电子设备及存储介质
CN202111147105.6 2021-09-29

Publications (1)

Publication Number Publication Date
WO2023050672A1 true WO2023050672A1 (zh) 2023-04-06

Family

ID=78242605

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/074598 WO2023050672A1 (zh) 2021-09-29 2022-01-28 随机写性能调节的方法、装置、电子设备及存储介质

Country Status (3)

Country Link
US (1) US20240377951A1 (zh)
CN (1) CN113590050B (zh)
WO (1) WO2023050672A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113590050B (zh) * 2021-09-29 2021-12-31 苏州浪潮智能科技有限公司 随机写性能调节的方法、装置、电子设备及存储介质
CN117827120A (zh) * 2024-01-23 2024-04-05 中电云计算技术有限公司 一种硬盘流量波动优化方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110400573A (zh) * 2018-04-25 2019-11-01 华为技术有限公司 一种数据处理的方法及装置
CN111090398A (zh) * 2019-12-13 2020-05-01 北京浪潮数据技术有限公司 固态硬盘的垃圾回收方法、装置、设备及可读存储介质
CN111708491A (zh) * 2020-05-29 2020-09-25 苏州浪潮智能科技有限公司 一种随机写方法和装置
US20210117317A1 (en) * 2019-10-17 2021-04-22 Kioxia Corporation Memory system and garbage collection control method
CN113590050A (zh) * 2021-09-29 2021-11-02 苏州浪潮智能科技有限公司 随机写性能调节的方法、装置、电子设备及存储介质

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111475115A (zh) * 2020-04-03 2020-07-31 苏州浪潮智能科技有限公司 一种ssd闪存寿命预测的方法、装置、设备及可读介质
CN112416255A (zh) * 2020-11-30 2021-02-26 苏州浪潮智能科技有限公司 一种用户写入速度控制方法、装置、设备和介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110400573A (zh) * 2018-04-25 2019-11-01 华为技术有限公司 一种数据处理的方法及装置
US20210117317A1 (en) * 2019-10-17 2021-04-22 Kioxia Corporation Memory system and garbage collection control method
CN111090398A (zh) * 2019-12-13 2020-05-01 北京浪潮数据技术有限公司 固态硬盘的垃圾回收方法、装置、设备及可读存储介质
CN111708491A (zh) * 2020-05-29 2020-09-25 苏州浪潮智能科技有限公司 一种随机写方法和装置
CN113590050A (zh) * 2021-09-29 2021-11-02 苏州浪潮智能科技有限公司 随机写性能调节的方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
US20240377951A1 (en) 2024-11-14
CN113590050A (zh) 2021-11-02
CN113590050B (zh) 2021-12-31

Similar Documents

Publication Publication Date Title
WO2023050672A1 (zh) 随机写性能调节的方法、装置、电子设备及存储介质
US11467767B2 (en) Storage device throttling amount of communicated data depending on suspension frequency of operation
US20170031822A1 (en) Control method and electronic device
CN103782269B (zh) 有效频率加速操作
US11635904B2 (en) Matrix storage method, matrix access method, apparatus and electronic device
US20140092103A1 (en) Method for adaptively adjusting framerate of graphic processing unit and computer system using thereof
CN108647727A (zh) 不平衡数据分类欠采样方法、装置、设备及介质
US10423719B2 (en) Dynamic loading of tabular data
WO2021238264A1 (zh) 一种随机写方法和装置
CN110825326A (zh) 提升ssd随机读性能的方法、装置、计算机设备及存储介质
WO2019114044A1 (zh) 图像处理方法、装置、电子设备及计算机可读存储介质
CN109669528B (zh) 终端的处理组件控制方法、装置、计算机设备及存储介质
CN117992367A (zh) 一种可变缓存替换管理方法及系统
CN107305531A (zh) 缓存容量限值确定方法和装置以及计算设备
WO2023155695A1 (zh) 处理器、控制方法、设备及介质
CN105740073A (zh) 一种动态控制操作系统进程数量的方法及装置
WO2020103027A1 (zh) 网络功耗调整方法、网络功耗调整装置及移动终端
CN107517273B (zh) 数据迁移的方法、系统、计算机可读存储介质及服务器
JP2023539570A (ja) キャプチャ圧力の制御方法、装置、電子機器及び読み取り可能な記憶媒体
US10838615B2 (en) Method for moving graphical pointer, system for moving graphical pointer, and touch display device
TW201537453A (zh) 電子裝置及基於其儲存裝置吞吐量調整緩衝區大小的方法
CN118466863B (zh) 一种数据存储方法、装置、存储介质及电子设备
KR102466551B1 (ko) 데이터 기록 방법, 장치 및 전자 기기
CN109324797B (zh) 一种桌面图标生成方法、计算机可读存储介质及终端设备
CN115048055A (zh) 固态硬盘分区做数据巡检的方法、装置、计算机设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22874075

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18564565

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22874075

Country of ref document: EP

Kind code of ref document: A1