Hard disk power-on control method and device, computer equipment and storage medium

Abstract

The invention relates to a hard disk power-on control method and device, computer equipment and a storage medium. The method comprises the steps of obtaining the number of to-be-powered hard disks and attribute information of the to-be-powered hard disks; normalizing power-on schemes of the to-be-powered-on hard disks based on the number of the to-be-powered-on hard disks and the attribute information of the to-be-powered-on hard disks to obtain multiple power-on schemes of the hard disks; traversing a plurality of hard disk power-on schemes, and calculating target required power consumption of each hard disk in each hard disk power-on scheme; setting hard disk off-peak starting time according to the target required power consumption corresponding to each hard disk power-on scheme; and obtaining a target power-on demand, screening out a target hard disk power-on scheme from the plurality of hard disk power-on schemes according to the target power-on demand, the hard disk peak shifting starting time and the target demand power consumption, and powering on the hard disk according to the target hard disk power-on scheme. By adopting the method, the reliability of the server system can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of hard disk power-on control, and in particular to a hard disk power-on control method, device, computer equipment and storage medium. Background Art

[0002] In the server system, the power consumption of the server hard disk has large fluctuations and occupancy. The server hard disk is a relatively important device in the server. In the storage server, the number of hard disks included in a single storage server may reach hundreds. The power supply risk of the mechanical hard disk itself is relatively large. When the power supply current is too large, it is easy to cause a crash and burn the hard disk. The power consumption of more mechanical hard disks during data storage throughput also fluctuates greatly. Therefore, improving the security of the power supply to the hard disk plays a vital role in data security and stable operation of the server. Summary of the invention

[0003] Based on this, it is necessary to provide a hard disk power-on control method, device, computer equipment and storage medium that can improve the reliability of the server system in response to the above technical problems.

[0004] In order to solve the above technical problems, in a first aspect, a hard disk power-on control method is provided, the method comprising:

[0005] Get the number of hard disks to be powered on and the attribute information of the hard disks to be powered on;

[0006] Based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, a power-on scheme for the hard disks to be powered on is standardized to obtain multiple power-on schemes for the hard disks;

[0007] Traverse multiple hard disk power-on schemes and calculate the target required power consumption of each hard disk in each hard disk power-on scheme;

[0008] Set the hard disk peak start time according to the target power consumption corresponding to each hard disk power-on solution;

[0009] Obtain a target power-on requirement, select a target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, hard disk off-peak startup time, and target required power consumption, and power on the hard disk according to the target hard disk power-on solution.

[0010] In one embodiment, the power-on schemes for the hard disks to be powered on are standardized based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and multiple hard disk power-on schemes are obtained, including:

[0011] Extracting the type of the hard disk to be powered on and the location information of the hard disk to be powered on from the attribute information of the hard disk to be powered on;

[0012] Set the number of hard disks included in each hard disk power-on plan according to the number of hard disks to be powered on;

[0013] Abstracting the hard disks to be powered on of the same type as nodes, and using the position relationship between the hard disks to be powered on of the same type as paths to build a network of hard disks to be powered on;

[0014] Select a node corresponding to any hard disk to be powered on from the hard disk network to be powered on as the target node, and form a hard disk group with multiple hard disks to be powered on that are connected to the target node path and meet the hard disk quantity requirement, until all the hard disks to be powered on are divided into multiple hard disk groups;

[0015] The power-on sequence of each hard disk in each hard disk group is set in the form of permutations and combinations to obtain multiple hard disk power-on solutions.

[0016] In one embodiment, traversing multiple hard disk power-on schemes and calculating the target required power consumption of each hard disk in each hard disk power-on scheme includes:

[0017] Obtain a preset hard disk power-on model, where the hard disk power-on model includes a target required power consumption calculation formula;

[0018] Input multiple hard disk power-on schemes into the hard disk power-on model, and calculate the target required power consumption of each hard disk in each hard disk power-on scheme according to the hard disk power consumption calculation rule;

[0019] The target required power consumption of each hard disk in each hard disk power-on solution is calculated according to the hard disk power consumption calculation rule, including:

[0020] Obtain the standard power consumption data of each hard disk in each hard disk power-on solution and the preset target required power consumption calculation formula;

[0021] Obtain hard disk capacity data of each hard disk in each hard disk power-on solution, and calculate a capacity impact factor according to the hard disk capacity data of each hard disk;

[0022] Obtaining standard working data and actual working data of each hard disk in each hard disk power-on scheme, and calculating a working state influencing factor according to the standard working data and actual working data of each hard disk;

[0023] Obtaining the operation stage information of each hard disk in each hard disk power-on scheme, and calculating the operation stage influencing factor according to the operation stage information of each hard disk;

[0024] Obtaining work efficiency information of each hard disk in each hard disk power-on scheme, and calculating a work efficiency influencing factor according to the work efficiency information of each hard disk;

[0025] Calculate the target power consumption required for each hard disk based on the preset target power consumption calculation formula, standard power consumption data, capacity impact factor, working status impact factor, operation stage impact factor, and work efficiency impact factor;

[0026] The preset target required power consumption calculation formula is as follows:

[0027]

[0028] Among them, P i1 represents the target required power consumption of hard disk i, P i2 represents the standard power consumption of hard disk i, λ represents the work efficiency influencing factor, |V i | represents the working efficiency of hard disk i, X represents the capacity impact factor, B i Indicates the standard storage capacity of hard disk i, S i represents the used storage capacity of hard disk i, Y represents the working status influencing factor, C i Indicates the standard working data of hard disk i, M i represents the actual working data of hard disk i, Z represents the operating stage influencing factor, and T0 represents the operating stage information of hard disk i.

[0029] In one embodiment, setting the hard disk peak start time according to the hard disk power consumption corresponding to each hard disk power-on scheme includes:

[0030] Obtain the maximum power consumption of each operation phase corresponding to each hard disk power-on solution;

[0031] Set the startup time of each hard disk according to the target power consumption of each hard disk and the maximum power consumption of each operation stage;

[0032] According to the startup time of each hard disk, the staggered startup time corresponding to the hard disk power-on solution to which the hard disk belongs is obtained.

[0033] In one embodiment, the target power-on requirement is obtained, and the target hard disk power-on solution is selected from multiple hard disk power-on solutions according to the target power-on requirement, the hard disk off-peak startup time, and the target required power consumption, including:

[0034] Obtain the target power-on requirement, and use the target power-on requirement as a constraint condition to obtain a hard disk power-on solution to be selected that meets the constraint condition from multiple hard disk power-on solutions;

[0035] Obtain the power consumption corresponding to each hard disk power-on solution to be selected, and the off-peak startup time corresponding to each hard disk power-on solution to be selected;

[0036] Calculate the average power consumption corresponding to each hard disk power-on solution to be selected, and calculate the variance corresponding to the hard disk power-on solution to be selected based on the average power consumption;

[0037] Calculate the priority of each hard disk power-on scheme to be selected based on the scheme corresponding to the hard disk power-on scheme to be selected and the staggered startup time corresponding to the hard disk power-on scheme to be selected;

[0038] The target hard disk power-on scheme is determined according to the priority of each hard disk power-on scheme to be selected.

[0039] In one embodiment, the method further comprises:

[0040] Get the maximum rated power of the power supply and the current system power consumption value;

[0041] Determine the remaining rated power of the power supply based on the difference between the maximum rated power of the power supply and the current system power consumption;

[0042] Detect the peak power of each hard disk after it is powered on, and the time it takes for the peak power of the hard disk to return to the power during stable operation;

[0043] Determine the actual peak power used by each hard disk after it is powered on based on the number of hard disks to be powered on and the peak power;

[0044] The remaining hard disks are powered on in batches according to the time taken for the peak power of the hard disks to return to the power during stable operation, wherein the peak power of each batch of hard disks after power-on, determined according to the actual peak power used by each hard disk after power-on and the power-on plan for each hard disk, is within the remaining rated power range of the power supply.

[0045] In one embodiment, obtaining the number of hard disks to be powered on and the attribute information of the hard disks to be powered on includes:

[0046] In response to the application program acquiring the attribute information of the multiple hard disks on the server backplane; preprocessing the acquired attribute information of the multiple hard disks on the server backplane;

[0047] A machine learning algorithm is used to perform dimension reduction processing on the attribute information of multiple hard disks on the server backplane, and a multi-dimensional feature corresponding to each of the multiple hard disks on the server backplane is obtained;

[0048] The multi-dimensional features corresponding to each hard disk are integrated to generate the health index corresponding to each hard disk;

[0049] The hard disks whose health index is greater than the preset health index value are regarded as hard disks to be powered on.

[0050] In order to solve the above technical problem, in a second aspect, a hard disk power-on control device is provided, the device comprising:

[0051] An acquisition module is used to acquire the number of hard disks to be powered on and the attribute information of the hard disks to be powered on;

[0052] A planning module, used to plan the power-on scheme of the hard disks to be powered on based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and obtain multiple hard disk power-on schemes;

[0053] A calculation module, used for traversing multiple hard disk power-on schemes and calculating the target required power consumption of each hard disk in each hard disk power-on scheme;

[0054] A setting module is used to set the hard disk peak start time according to the target required power consumption corresponding to each hard disk power-on scheme;

[0055] The power-on module is used to obtain the target power-on requirement, select the target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, the hard disk off-peak startup time and the target required power consumption, and power on the hard disk according to the target hard disk power-on solution.

[0056] In order to solve the above technical problem, a third aspect provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: the processor implements the steps of the above first aspect method when executing the computer program.

[0057] In order to solve the above technical problems, in a fourth aspect, the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method of the above first aspect are implemented.

[0058] Different from the prior art, the present application obtains the number of hard disks to be powered on and the attribute information of the hard disks to be powered on; based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the power-on schemes of the hard disks to be powered on are normalized to obtain multiple hard disk power-on schemes; traverses multiple hard disk power-on schemes to calculate the target power consumption of each hard disk in each hard disk power-on scheme; sets the hard disk staggered startup time according to the target power consumption corresponding to each hard disk power-on scheme; obtains the target power-on demand, selects the target hard disk power-on scheme from multiple hard disk power-on schemes according to the target power-on demand, the hard disk staggered startup time and the target power consumption, and powers on the hard disk according to the target hard disk power-on scheme. In this way, the hard disk power-on sequence under different strategies is simulated by deep traversal, and the optimal hard disk power-on scheme is automatically discovered and implemented, which can improve the reliability of the server system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flowchart of a hard disk power-on control method in one embodiment;

[0060] Figure 2 A schematic diagram of a hard disk power-on model in an embodiment;

[0061] Figure 3 is a graph showing the relationship between current, power consumption and time in one embodiment;

[0062] Figure 4 A schematic diagram of a hard disk power-on model in an embodiment;

[0063] Figure 5 is a structural block diagram of a hard disk power-on control device in an embodiment;

[0064] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] With the rapid development of information technology, large-scale storage devices such as servers and data centers are increasingly being used. However, the simultaneous power-on of hard disks may cause the system power consumption to increase instantly, and may even cause the power supply system to fail to start normally due to overload protection (OCP). For example, some JBOD servers have more than 100 hard disks, and the startup power consumption of a single disk is more than 35W. If the peak load is not staggered, the instantaneous startup power consumption will be more than 35*100=3500W. Therefore, finding the optimal solution for staggered power-on of hard disks is of great significance to improving system stability and reducing power consumption.

[0067] In the prior art, the actual power consumption of the hard disk is mainly obtained, and the power-on strategy is dynamically adjusted according to the actual power consumption of the hard disk. However, the prior art ignores the evaluation of the overall effect of peak shifting.

[0068] In view of the above technical problems, in one embodiment, Figure 1 As shown, the present application provides a hard disk power-on control method, which specifically includes the following steps:

[0069] Step S10, obtaining the number of hard disks to be powered on and attribute information of the hard disks to be powered on.

[0070] Step S11 , normalizing power-on schemes for the hard disks to be powered on based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and obtaining a plurality of hard disk power-on schemes.

[0071] In one embodiment, the type of the hard disk to be powered on and the position information of the hard disk to be powered on can be extracted from the attribute information of the hard disk to be powered on; the number of hard disks included in each hard disk power-on scheme is set according to the number of hard disks to be powered on; the hard disks to be powered on of the same type are abstracted as nodes, and the position relationship between the hard disks to be powered on of the same type is used as a path to build a network of hard disks to be powered on; the node corresponding to any hard disk to be powered on is selected from the network of hard disks to be powered on as the target node, and a hard disk group is formed with multiple hard disks to be powered on that are connected to the target node path and meet the hard disk quantity requirement, until all the hard disks to be powered on are divided into multiple hard disk groups; the power-on sequence of each hard disk in each hard disk group is set in the form of permutations and combinations to obtain multiple hard disk power-on schemes.

[0072] Specifically, the number of hard disks to be powered on can be obtained, and the hard disks to be powered on can be evenly divided into several groups. For example, if the total number of hard disks to be powered on is 12, 3 or 4 can be selected as the number of hard disks included in the power-on scheme for each hard disk, and then the hard disk network to be powered on is constructed, and any node is selected as the target node to select the adjacent nodes connected to the node path, and the number of hard disks included in the power-on scheme for each hard disk is set as the threshold for all selected adjacent nodes to form a hard disk group, directly until all hard disks to be powered on are divided into hard disk groups, and multiple hard disk groups can be obtained, which can be arranged in the form of permutations and combinations according to the power-on sequence of each hard disk in the hard disk group. For example, the permutations and combinations in this application refer to the power-on sequence of each hard disk in the hard disk group. For example, when there are 3 hard disks, starting 2 disks first and then starting 1 disk is different from starting 1 disk first and then starting 2 disks. When the number of hard disks to be powered on is N, 2^(n-1) grouping schemes will be generated.

[0073] Step S12, traversing multiple hard disk power-on schemes, and calculating the target required power consumption of each hard disk in each hard disk power-on scheme.

[0074] Specifically, a preset hard disk power-on model can be obtained, which includes a target required power consumption calculation formula; multiple hard disk power-on schemes are input into the hard disk power-on model, and the target required power consumption of each hard disk in each hard disk power-on scheme is calculated according to the hard disk power consumption calculation rules.

[0075] The hard disk power-on model in the specific application can be the data in the hard disk spec, which refers to the current data model of the hard disk power-on stage. It is a common data and can generally be obtained from the hard disk spec. If the hard disk spec does not mention it, you can use an oscilloscope to monitor the 12V and 5V power supplies of the hard disk to obtain the power-on model of the power-on stage. For example, please refer to Figure 2In this application, multiple hard disk power-on schemes are input into the hard disk power-on model. According to the system staggered power-on simulation and power consumption evaluation functions included in the hard disk power-on model, the hard disk power consumption calculation rules are used to calculate the target required power consumption of each hard disk in each hard disk power-on scheme. Finally, the optimal hard disk power-on scheme output model is selected from the multiple hard disk power-on schemes according to the power-on requirements.

[0076] The hard disk power-on model is a data model that establishes the current and power consumption of a single hard disk during the startup phase, from startup to entering the stable phase. Mechanical hard disks require two voltages for power supply, one is 5V and the other is 12V. Figure 3 The figure below shows the current waveform of a certain hard disk during the DC startup phase. Figure 2 The data is stored digitally, and the generated data is shown in Table 1. Table 1 records the real-time 5V current and 12V current at each time point during the hard disk startup phase, with a time accuracy of 0.1 second. Due to the limitation of the number of displays, Table 1 only shows the data at some time points.

[0077] The power-on model can be understood as a quantitative model of (power-on time, instantaneous power consumption), wherein the horizontal axis is time and the vertical axis is instantaneous current (power consumption).

[0078]

[0079] Table 1

[0080] As shown in Table 1, under this model, it takes 10 seconds for the hard disk to complete startup and enter the current stable state after 10 seconds. The peak current of a hard disk during off-peak startup is 2.8A, while the normal operating current is only 0.5A. These data are digitized and graphically displayed as Figure 4 .from Figure 4 It can be seen that the current and power consumption values ​​are not very stable, and there are sudden increases and decreases.

[0081] If 30 hard disks are started at the same time, without peak shifting, an instantaneous current of 2.8*30=84A will be generated, resulting in a power consumption of 12v*84A=1008W. Without peak shifting, the peak current of 84A and the peak power consumption of 1008W will cause an impact on the system, requiring the system to provide a greater current overload capacity and greater system power consumption support. Peak-shifting hard disk power-on is actually a solution that exchanges time for space, by extending the startup time to reduce the spatial range of current and power consumption.

[0082] The target required power consumption of each hard disk in each hard disk power-on solution is calculated according to the hard disk power consumption calculation rule, including:

[0083] Obtain the standard power consumption data of each hard disk in each hard disk power-on solution and the preset target required power consumption calculation formula;

[0084] Obtain the hard disk capacity data of each hard disk in each hard disk power-on scheme, and calculate the capacity impact factor according to the hard disk capacity data of each hard disk; the hard disk capacity data here can be data capacity, and the capacity impact factor can be set according to the degree of influence of the hard disk capacity data on the hard disk power-on. Obtain the standard working data and actual working data of each hard disk in each hard disk power-on scheme, and calculate the working state impact factor according to the standard working data and actual working data of each hard disk; obtain the standard working state data of each hard disk, including standard read and write speed data, standard workload data and standard temperature data; obtain the real-time working state data of each hard disk, including real-time read and write speed data, real-time workload data and real-time temperature data; calculate the working state impact factor according to the standard read and write speed data, standard workload data, standard temperature data, real-time read and write speed data, real-time workload data and real-time temperature data. Obtain the operation stage information of each hard disk in each hard disk power-on scheme, and calculate the operation stage impact factor according to the operation stage information of each hard disk; the maximum load of each stage may be different, and the operation stage impact factor is set according to the operation stage information of each hard disk. Obtain the work efficiency information of each hard disk in each hard disk power-on plan, and calculate the work efficiency impact factor based on the work efficiency information of each hard disk; the work efficiency information can be determined by the real-time read and write speed of the hard disk, the time to complete the work, etc., and the work efficiency impact factor can be set based on the work efficiency information of the hard disk.

[0085] Calculate the target power consumption required for each hard disk based on the preset target power consumption calculation formula, standard power consumption data, capacity impact factor, working status impact factor, operation stage impact factor, and work efficiency impact factor;

[0086] The preset target required power consumption calculation formula is as follows:

[0087]

[0088] Among them, P i1 represents the target required power consumption of hard disk i, P i2 represents the standard power consumption of hard disk i, λ represents the work efficiency influencing factor, |V i | represents the working efficiency of hard disk i, X represents the capacity impact factor, B i Indicates the standard storage capacity of hard disk i, S i represents the used storage capacity of hard disk i, Y represents the working status influencing factor, C i Indicates the standard working data of hard disk i, M irepresents the actual working data of hard disk i, Z represents the operating stage influencing factor, and T0 represents the operating stage information of hard disk i.

[0089] In the present application, the target required power consumption of each hard disk is calculated according to a preset target required power consumption calculation formula, standard power consumption data, capacity influence factor, working status influence factor, operation stage influence factor and work efficiency influence factor, thereby improving the accuracy of calculating the target required power consumption of each hard disk.

[0090] Step S13: setting the hard disk peak-shifting startup time according to the target required power consumption corresponding to each hard disk power-on solution.

[0091] Specifically, the maximum power consumption of each operating stage corresponding to each hard disk power-on plan can be obtained; the startup time of each hard disk can be set according to the target required power consumption of each hard disk and the maximum power consumption of each operating stage; based on the startup time of each hard disk, the staggered startup time corresponding to the hard disk power-on plan to which the hard disk belongs can be obtained.

[0092] In a specific implementation, the present application is provided with obtaining the maximum power consumption of each operating stage corresponding to each hard disk power-on scheme, and starting each hard disk in the same hard disk power-on scheme in stages according to the target required power consumption of each hard disk in each hard disk power-on scheme, and taking the interval between the initial power-on time of the first power-on in any hard disk power-on scheme and the end time of the last hard disk power-on in the scheme as the staggered startup time corresponding to the hard disk power-on scheme, so that the power consumption of the hard disk in different operating stages can be stabilized, which is beneficial to improving the stability of the system.

[0093] Step S14: obtaining a target power-on requirement, selecting a target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, hard disk off-peak startup time, and target required power consumption, and powering on the hard disk according to the target hard disk power-on solution.

[0094] In one embodiment, a target power-on requirement can be obtained, and with the target power-on requirement as a constraint condition, a hard disk power-on scheme to be selected that meets the constraint condition is obtained from a plurality of hard disk power-on schemes; the power consumption corresponding to each hard disk power-on scheme to be selected, and the staggered startup time corresponding to each hard disk power-on scheme to be selected are obtained; the average power consumption corresponding to each hard disk power-on scheme to be selected is calculated, and the variance corresponding to the hard disk power-on scheme to be selected is calculated based on the average power consumption; the priority of each hard disk power-on scheme to be selected is calculated based on the scheme corresponding to the hard disk power-on scheme to be selected and the staggered startup time corresponding to the hard disk power-on scheme to be selected; and the target hard disk power-on scheme is determined according to the priority of each hard disk power-on scheme to be selected.

[0095] Exemplarily, the power consumption evaluation system of the present application can use the peak value (peak value of power consumption), variance, off-peak duration, and variance x duration product of power consumption as evaluation criteria. Among them, the smaller the peak value, the lower the peak value of the system. The smaller the variance, the lower the fluctuation of power consumption. The shorter the off-peak duration, the shorter the time it takes for the system to reach power consumption balance. Among them, the peak value, that is, the peak value of power consumption, refers to the maximum value of power consumption max(power), which is sampled every 0.1 seconds and the maximum value of the sample is taken.

[0096] The variance is calculated based on the average power consumption corresponding to each hard disk power-on solution to be selected, and the variance corresponding to the hard disk power-on solution to be selected is calculated based on the average power consumption. The variance formula is as follows:

[0097]

[0098] Where S represents the square of the variance, A represents the power consumption of each hard disk, x1 represents the power consumption of the first hard disk in the hard disk power-on scheme x, x2 represents the power consumption of the second hard disk in the hard disk power-on scheme x, and x n It represents the power consumption of the nth hard disk in the hard disk power-on scheme x, and n represents the number of hard disks included in the hard disk power-on scheme x.

[0099] Finally, according to the needs and the above evaluation criteria, the optimal solution is output. For example, in some low-power demand scenarios, the lowest peak power consumption is the first priority demand, and the off-peak time is less concerned, so you can give priority to choosing from the solutions with lower variance and peak value. For example, in some warm storage demand scenarios, the hard disk needs to be started as soon as possible within the specified time. At this time, the off-peak time needs to be the first priority demand. From the solutions that meet the off-peak time, choose the solution with lower variance and peak value, which can improve the accuracy of hard disk power-on.

[0100] In one embodiment, a delay scheme can also be set, and the delay file can be combined with the grouping scheme to find all the combinations of staggered power-on methods by deep traversal. Take three hard disks as an example. Three hard disks can be started as a group; one hard disk and two hard disks can be started as a group, totaling two groups, or three groups with one hard disk in each group. Exemplarily, in this case, 81 staggered power-on methods are generated in the case of three hard disks. In this way, n hard disks will generate 9^(n-1) schemes. The delay scheme refers to the delay between every two groups of hard disks. This delay is in a range, and the startup time (spin up time) of the hard disk can be used as this range.

[0101] By utilizing the powerful computing performance of computers, deeply traversing all hard disk staggered startup schemes and finding the optimal staggered startup scheme, the energy efficiency of data centers can be significantly improved, energy consumption during peak hours can be reduced, and the pressure on the power grid can be effectively alleviated. By replacing the hard disk power-on model, staggered power-on simulations can be performed on multiple hard disks, so as to have a clearer grasp of the hard disk compatibility of the system.

[0102] In a specific implementation, the maximum rated power of the power supply and the current system power consumption value can be obtained; the remaining rated power of the power supply is determined based on the difference between the maximum rated power of the power supply and the current system power consumption; the peak power of each hard disk after power-on, and the time taken for the peak power of the hard disk to return to the power during stable operation are detected; the actual peak power used by each hard disk after power-on is determined based on the number of hard disks to be powered on and the peak power; the remaining hard disks are powered on in batches according to the time taken for the peak power of the hard disks to return to the power during stable operation, wherein the peak power of each batch of hard disks after power-on determined according to the actual peak power used by each hard disk after power-on and the power-on plan for each hard disk is within the range of the remaining rated power of the power supply.

[0103] In the specific implementation, firstly, the number of hard disks to be powered on in the next batch is calculated based on the remaining rated power of the power supply and the actual peak power used by each hard disk after being powered on, and the hard disks in this batch are powered on according to the staggered time of the hard disks. After the power-on is completed, if the remaining rated power of the power supply is sufficient to start the next batch of hard disks of the same number, the next batch of hard disks will continue to be powered on at time intervals until all hard disks are powered on. Alternatively, it is detected whether the peak power of the hard disks has returned to stable operation. When the peak power of the previous batch of hard disks has returned to stable operation, the remaining hard disks are powered on in batches. First, the number of hard disks that need to be powered on in the next batch is calculated based on the remaining rated power of the power supply and the actual peak power used by each hard disk after power-on, and the peak power of the hard disks is checked to see if it has returned to stable operation. When the peak power of the previous batch of hard disks returns to stable operation, the hard disks in that batch are powered on. After power-on is completed, the number of hard disks that need to be powered on in the next batch is calculated based on the remaining rated power of the power supply and the actual peak power used by each hard disk after power-on, and the peak power of the hard disks is checked again to see if it has returned to stable operation. When the peak power of the previous batch of hard disks returns to stable operation, the next batch of hard disks is powered on until all hard disks are powered on.

[0104] During normal system operation, the ambient temperature and system power consumption of the chassis power supply can be monitored in real time. Based on the learned ambient temperature value and the corresponding curve relationship between the chassis power supply rated power and temperature, the maximum rated power of the power supply under the ambient temperature can be determined, and then the remaining rated power of the power supply can be calculated based on the maximum rated power of the power supply and the real-time power consumption. When the calculated remaining rated power of the power supply is insufficient, a signal is sent to inform the CPU. After receiving the information, the CPU performs business analysis and transfers or shuts down the business of certain disks or modules (such as hard disks and blowers) based on the analysis results. At the same time, an alarm signal is sent to notify the user to take corresponding measures such as stopping power supply to reduce power consumption until the total power quota does not exceed the standard, ensuring stable operation of the system.

[0105] In one embodiment, the hard disk power-on model is a hard disk power-on model trained by training text, where the training text can be a variety of hard disk power-on solutions generated according to the completed project, and the model is trained by judging whether the final output solution is the optimal solution as the model convergence condition. In one embodiment, the training text can be segmented, at least one word in the training text can be randomly selected and masked to obtain a masked text; the masked words in the masked text can be predicted; the masked words in the masked text can be replaced with the predicted results of the masked words to obtain an enhanced text; if the semantic similarity between the training text and the corresponding enhanced text meets the preset conditions, the enhanced text is added to the training set to obtain an expanded training set; and the model is trained using the expanded training set. Through this method, the text can be expanded by modifying non-keywords and / or keywords under the condition of less impact on the text semantics, and a large number of expanded texts can be generated, and it has strong randomness, and can provide a richer training set for the model, so as to improve the training quality of the model.

[0106] In a specific implementation, a sample set is generated based on the attribute information of multiple hard disks, and the data in the sample set is preprocessed, including preprocessing, standardization, and sample generation. Preprocessing includes data cleaning and data organization, and data cleaning realizes abnormal data removal and missing data processing. Statistical methods or machine learning algorithms are used to reduce the dimension of the initial multidimensional feature data. The machine learning algorithm here can be specifically a dimensionality reduction algorithm, which can map data points in the high-order space to the low-dimensional space. By applying the dimensionality reduction algorithm, the hidden patterns and structures in the attribute information of the hard disk can be found, which is conducive to improving the effect and performance of the model. The most representative one is principal component analysis (PCA). Its basic idea is to project the original n features into k-dimensional space. The orthogonal k-dimensional space is called the principal component, and the new features are transformed from the original features. After weighing the number of features, deleting redundant features, and selecting a more representative feature combination, the multidimensional features corresponding to each hard disk in the multiple hard disks on the server backplane are obtained to improve the model performance. The multi-dimensional features corresponding to each hard disk are fused to generate the health index corresponding to each hard disk. The health index is used to characterize the degradation state of the hard disk over time. Based on the health index, the degradation modeling and prediction method is used to evaluate the life of each hard disk.

[0107] In this application, by modeling and analyzing the degradation characteristics of the hard disk during use, an accurate remaining life prediction result is obtained, and then the remaining life assessment of the hard disk can be realized. In this application, a fusion is performed according to the multi-dimensional features corresponding to each hard disk to generate a health index corresponding to each hard disk; hard disks with health indicators greater than the preset health index value are used as hard disks to be powered on, which can improve the reliability of hard disk power-on.

[0108] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0109] In one embodiment, Figure 5 As shown, a hard disk power-on control device is provided, including: an acquisition module 50, a planning module 51, a calculation module 52, a setting module 53 and a power-on module 54, wherein:

[0110] An acquisition module 50 is used to acquire the number of hard disks to be powered on and attribute information of the hard disks to be powered on;

[0111] A planning module 51, configured to plan power-on schemes for the hard disks to be powered on based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and obtain multiple power-on schemes for the hard disks;

[0112] A calculation module 52, configured to traverse a plurality of hard disk power-on schemes and calculate a target required power consumption of each hard disk in each hard disk power-on scheme;

[0113] A setting module 53 is used to set a hard disk staggered startup time according to the target required power consumption corresponding to each hard disk power-on scheme;

[0114] The power-on module 54 is used to obtain a target power-on requirement, select a target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, the hard disk off-peak startup time and the target required power consumption, and power on the hard disk according to the target hard disk power-on solution.

[0115] In one embodiment, the above device can implement another implementation of a hard disk power-on control method, and the specific steps are as follows:

[0116] Based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the power-on scheme of the hard disks to be powered on is standardized, and multiple hard disk power-on schemes are obtained, including:

[0117] Extracting the type of the hard disk to be powered on and the location information of the hard disk to be powered on from the attribute information of the hard disk to be powered on;

[0118] Set the number of hard disks included in each hard disk power-on plan according to the number of hard disks to be powered on;

[0119] Abstracting the hard disks to be powered on of the same type as nodes, and using the position relationship between the hard disks to be powered on of the same type as paths to build a network of hard disks to be powered on;

[0120] Select a node corresponding to any hard disk to be powered on from the hard disk network to be powered on as the target node, and form a hard disk group with multiple hard disks to be powered on that are connected to the target node path and meet the hard disk quantity requirement, until all the hard disks to be powered on are divided into multiple hard disk groups;

[0121] The power-on sequence of each hard disk in each hard disk group is set in the form of permutations and combinations to obtain multiple hard disk power-on solutions.

[0122] In one embodiment, the above device can implement another implementation of a hard disk power-on control method, and the specific steps are as follows:

[0123] Traversing multiple hard disk power-on schemes, calculating the target required power consumption of each hard disk in each hard disk power-on scheme includes:

[0124] Obtain a preset hard disk power-on model, where the hard disk power-on model includes a target required power consumption calculation formula;

[0125] Input multiple hard disk power-on schemes into the hard disk power-on model, and calculate the target required power consumption of each hard disk in each hard disk power-on scheme according to the hard disk power consumption calculation rule;

[0126] The target required power consumption of each hard disk in each hard disk power-on solution is calculated according to the hard disk power consumption calculation rule, including:

[0127] Obtain the standard power consumption data of each hard disk in each hard disk power-on solution and the preset target required power consumption calculation formula;

[0128] Obtain hard disk capacity data of each hard disk in each hard disk power-on solution, and calculate a capacity impact factor according to the hard disk capacity data of each hard disk;

[0129] Obtaining standard working data and actual working data of each hard disk in each hard disk power-on scheme, and calculating a working state influencing factor according to the standard working data and actual working data of each hard disk;

[0130] Obtaining the operation stage information of each hard disk in each hard disk power-on scheme, and calculating the operation stage influencing factor according to the operation stage information of each hard disk;

[0131] Obtaining work efficiency information of each hard disk in each hard disk power-on scheme, and calculating a work efficiency influencing factor according to the work efficiency information of each hard disk;

[0132] Calculate the target power consumption required for each hard disk based on the preset target power consumption calculation formula, standard power consumption data, capacity impact factor, working status impact factor, operation stage impact factor, and work efficiency impact factor;

[0133] The preset target required power consumption calculation formula is as follows:

[0134]

[0135] Among them, P i1 represents the target required power consumption of hard disk i, P i2 represents the standard power consumption of hard disk i, λ represents the work efficiency influencing factor, |V i | represents the working efficiency of hard disk i, X represents the capacity impact factor, B i Indicates the standard storage capacity of hard disk i, S i represents the used storage capacity of hard disk i, Y represents the working status influencing factor, C iIndicates the standard working data of hard disk i, M i represents the actual working data of hard disk i, Z represents the operating stage influencing factor, and T0 represents the operating stage information of hard disk i.

[0136] In one embodiment, the above device can implement another implementation of a hard disk power-on control method, and the specific steps are as follows:

[0137] According to the hard disk power consumption corresponding to each hard disk power-on scheme, the hard disk peak startup time is set as follows:

[0138] Obtain the maximum power consumption of each operation phase corresponding to each hard disk power-on solution;

[0139] Set the startup time of each hard disk according to the target power consumption of each hard disk and the maximum power consumption of each operation stage;

[0140] According to the startup time of each hard disk, the staggered startup time corresponding to the hard disk power-on solution to which the hard disk belongs is obtained.

[0141] In one embodiment, the above device can implement another implementation of a hard disk power-on control method, and the specific steps are as follows:

[0142] Obtain the target power-on requirement, and select the target hard disk power-on solution from multiple hard disk power-on solutions based on the target power-on requirement, hard disk off-peak startup time, and target required power consumption. The target hard disk power-on solution includes:

[0143] Obtain the target power-on requirement, and use the target power-on requirement as a constraint condition to obtain a hard disk power-on solution to be selected that meets the constraint condition from multiple hard disk power-on solutions;

[0144] Obtain the power consumption corresponding to each hard disk power-on solution to be selected, and the off-peak startup time corresponding to each hard disk power-on solution to be selected;

[0145] Calculate the average power consumption corresponding to each hard disk power-on solution to be selected, and calculate the variance corresponding to the hard disk power-on solution to be selected based on the average power consumption;

[0146] Calculate the priority of each hard disk power-on scheme to be selected based on the scheme corresponding to the hard disk power-on scheme to be selected and the staggered startup time corresponding to the hard disk power-on scheme to be selected;

[0147] The target hard disk power-on scheme is determined according to the priority of each hard disk power-on scheme to be selected.

[0148] In one embodiment, the above device can implement another implementation of a hard disk power-on control method, and the specific steps are as follows:

[0149] The method also includes:

[0150] Get the maximum rated power of the power supply and the current system power consumption value;

[0151] Determine the remaining rated power of the power supply based on the difference between the maximum rated power of the power supply and the current system power consumption;

[0152] Detect the peak power of each hard disk after it is powered on, and the time it takes for the peak power of the hard disk to return to the power during stable operation;

[0153] Determine the actual peak power used by each hard disk after it is powered on based on the number of hard disks to be powered on and the peak power;

[0154] The remaining hard disks are powered on in batches according to the time taken for the peak power of the hard disks to return to the power during stable operation, wherein the peak power of each batch of hard disks after power-on, determined according to the actual peak power used by each hard disk after power-on and the power-on plan for each hard disk, is within the remaining rated power range of the power supply.

[0155] In one embodiment, the above device can implement another implementation of a hard disk power-on control method, and the specific steps are as follows:

[0156] Before obtaining the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the following steps are included:

[0157] In response to the application program acquiring the attribute information of the multiple hard disks on the server backplane; preprocessing the acquired attribute information of the multiple hard disks on the server backplane;

[0158] A machine learning algorithm is used to perform dimension reduction processing on the attribute information of multiple hard disks on the server backplane, and a multi-dimensional feature corresponding to each of the multiple hard disks on the server backplane is obtained;

[0159] The multi-dimensional features corresponding to each hard disk are integrated to generate the health index corresponding to each hard disk;

[0160] The hard disks whose health index is greater than the preset health index value are regarded as hard disks to be powered on.

[0161] The specific definition of the hard disk power-on control device can be found in the definition of the hard disk power-on control method above, which will not be repeated here. Each module in the above hard disk power-on control device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0162] In one embodiment, the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the hard disk power-on control method provided by the above methods.

[0163] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data used in the hard disk power-on control method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a hard disk power-on control method is implemented.

[0164] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0165] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0166] Step S10, obtaining the number of hard disks to be powered on and attribute information of the hard disks to be powered on.

[0167] Step S11 , normalizing power-on schemes for the hard disks to be powered on based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and obtaining a plurality of hard disk power-on schemes.

[0168] Step S12, traversing multiple hard disk power-on schemes, and calculating the target required power consumption of each hard disk in each hard disk power-on scheme.

[0169] Step S13: setting the hard disk peak-shifting startup time according to the target required power consumption corresponding to each hard disk power-on solution.

[0170] Step S14: obtaining a target power-on requirement, selecting a target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, hard disk off-peak startup time, and target required power consumption, and powering on the hard disk according to the target hard disk power-on solution.

[0171] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0172] Based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the power-on scheme of the hard disks to be powered on is standardized, and multiple hard disk power-on schemes are obtained, including:

[0173] Extracting the type of the hard disk to be powered on and the location information of the hard disk to be powered on from the attribute information of the hard disk to be powered on;

[0174] Set the number of hard disks included in each hard disk power-on plan according to the number of hard disks to be powered on;

[0175] Abstracting the hard disks to be powered on of the same type as nodes, and using the position relationship between the hard disks to be powered on of the same type as paths to build a network of hard disks to be powered on;

[0176] Select a node corresponding to any hard disk to be powered on from the hard disk network to be powered on as the target node, and form a hard disk group with multiple hard disks to be powered on that are connected to the target node path and meet the hard disk quantity requirement, until all the hard disks to be powered on are divided into multiple hard disk groups;

[0177] The power-on sequence of each hard disk in each hard disk group is set in the form of permutations and combinations to obtain multiple hard disk power-on solutions.

[0178] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0179] Traversing multiple hard disk power-on schemes, calculating the target required power consumption of each hard disk in each hard disk power-on scheme includes:

[0180] Obtain a preset hard disk power-on model, where the hard disk power-on model includes a target required power consumption calculation formula;

[0181] Input multiple hard disk power-on schemes into the hard disk power-on model, and calculate the target required power consumption of each hard disk in each hard disk power-on scheme according to the hard disk power consumption calculation rule;

[0182] The target required power consumption of each hard disk in each hard disk power-on solution is calculated according to the hard disk power consumption calculation rule, including:

[0183] Obtain the standard power consumption data of each hard disk in each hard disk power-on solution and the preset target required power consumption calculation formula;

[0184] Obtain hard disk capacity data of each hard disk in each hard disk power-on solution, and calculate a capacity impact factor according to the hard disk capacity data of each hard disk;

[0185] Obtaining standard working data and actual working data of each hard disk in each hard disk power-on scheme, and calculating a working state influencing factor according to the standard working data and actual working data of each hard disk;

[0186] Obtaining the operation stage information of each hard disk in each hard disk power-on scheme, and calculating the operation stage influencing factor according to the operation stage information of each hard disk;

[0187] Obtaining work efficiency information of each hard disk in each hard disk power-on scheme, and calculating a work efficiency influencing factor according to the work efficiency information of each hard disk;

[0188] Calculate the target power consumption required for each hard disk based on the preset target power consumption calculation formula, standard power consumption data, capacity impact factor, working status impact factor, operation stage impact factor, and work efficiency impact factor;

[0189] The preset target required power consumption calculation formula is as follows:

[0190]

[0191] Among them, P i1 represents the target required power consumption of hard disk i, P i2 represents the standard power consumption of hard disk i, λ represents the work efficiency influencing factor, |V i | represents the working efficiency of hard disk i, X represents the capacity impact factor, B i Indicates the standard storage capacity of hard disk i, S i represents the used storage capacity of hard disk i, Y represents the working status influencing factor, C i Indicates the standard working data of hard disk i, M i represents the actual working data of hard disk i, Z represents the operating stage influencing factor, and T0 represents the operating stage information of hard disk i.

[0192] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0193] According to the hard disk power consumption corresponding to each hard disk power-on scheme, the hard disk peak startup time is set as follows:

[0194] Obtain the maximum power consumption of each operation phase corresponding to each hard disk power-on solution;

[0195] Set the startup time of each hard disk according to the target power consumption of each hard disk and the maximum power consumption of each operation stage;

[0196] According to the startup time of each hard disk, the staggered startup time corresponding to the hard disk power-on solution to which the hard disk belongs is obtained.

[0197] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0198] Obtain the target power-on requirement, and select the target hard disk power-on solution from multiple hard disk power-on solutions based on the target power-on requirement, hard disk off-peak startup time, and target required power consumption. The target hard disk power-on solution includes:

[0199] Obtain the target power-on requirement, and use the target power-on requirement as a constraint condition to obtain a hard disk power-on solution to be selected that meets the constraint condition from multiple hard disk power-on solutions;

[0200] Obtain the power consumption corresponding to each hard disk power-on solution to be selected, and the off-peak startup time corresponding to each hard disk power-on solution to be selected;

[0201] Calculate the average power consumption corresponding to each hard disk power-on solution to be selected, and calculate the variance corresponding to the hard disk power-on solution to be selected based on the average power consumption;

[0202] Calculate the priority of each hard disk power-on scheme to be selected based on the scheme corresponding to the hard disk power-on scheme to be selected and the staggered startup time corresponding to the hard disk power-on scheme to be selected;

[0203] The target hard disk power-on scheme is determined according to the priority of each hard disk power-on scheme to be selected.

[0204] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0205] The method also includes:

[0206] Get the maximum rated power of the power supply and the current system power consumption value;

[0207] Determine the remaining rated power of the power supply based on the difference between the maximum rated power of the power supply and the current system power consumption;

[0208] Detect the peak power of each hard disk after it is powered on, and the time it takes for the peak power of the hard disk to return to the power during stable operation;

[0209] Determine the actual peak power used by each hard disk after it is powered on based on the number of hard disks to be powered on and the peak power;

[0210] The remaining hard disks are powered on in batches according to the time taken for the peak power of the hard disks to return to the power during stable operation, wherein the peak power of each batch of hard disks after power-on, determined according to the actual peak power used by each hard disk after power-on and the power-on plan for each hard disk, is within the remaining rated power range of the power supply.

[0211] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0212] Before obtaining the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the following steps are included:

[0213] In response to the application program acquiring the attribute information of the multiple hard disks on the server backplane; preprocessing the acquired attribute information of the multiple hard disks on the server backplane;

[0214] A machine learning algorithm is used to perform dimension reduction processing on the attribute information of multiple hard disks on the server backplane, and a multi-dimensional feature corresponding to each of the multiple hard disks on the server backplane is obtained;

[0215] The multi-dimensional features corresponding to each hard disk are integrated to generate the health index corresponding to each hard disk;

[0216] The hard disks whose health index is greater than the preset health index value are regarded as hard disks to be powered on.

[0217] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0218] Step S10, obtaining the number of hard disks to be powered on and attribute information of the hard disks to be powered on.

[0219] Step S11 , normalizing power-on schemes for the hard disks to be powered on based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and obtaining a plurality of hard disk power-on schemes.

[0220] Step S12, traversing multiple hard disk power-on schemes, and calculating the target required power consumption of each hard disk in each hard disk power-on scheme.

[0221] Step S13: setting the hard disk peak-shifting startup time according to the target required power consumption corresponding to each hard disk power-on solution.

[0222] Step S14: obtaining a target power-on requirement, selecting a target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, hard disk off-peak startup time, and target required power consumption, and powering on the hard disk according to the target hard disk power-on solution.

[0223] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0224] Based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the power-on scheme of the hard disks to be powered on is standardized, and multiple hard disk power-on schemes are obtained, including:

[0225] Extracting the type of the hard disk to be powered on and the location information of the hard disk to be powered on from the attribute information of the hard disk to be powered on;

[0226] Set the number of hard disks included in each hard disk power-on plan according to the number of hard disks to be powered on;

[0227] Abstracting the hard disks to be powered on of the same type as nodes, and using the position relationship between the hard disks to be powered on of the same type as paths to build a network of hard disks to be powered on;

[0228] Select a node corresponding to any hard disk to be powered on from the hard disk network to be powered on as the target node, and form a hard disk group with multiple hard disks to be powered on that are connected to the target node path and meet the hard disk quantity requirement, until all the hard disks to be powered on are divided into multiple hard disk groups;

[0229] The power-on sequence of each hard disk in each hard disk group is set in the form of permutations and combinations to obtain multiple hard disk power-on solutions.

[0230] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0231] Traversing multiple hard disk power-on schemes, calculating the target required power consumption of each hard disk in each hard disk power-on scheme includes:

[0232] Obtain a preset hard disk power-on model, where the hard disk power-on model includes a target required power consumption calculation formula;

[0233] Input multiple hard disk power-on schemes into the hard disk power-on model, and calculate the target required power consumption of each hard disk in each hard disk power-on scheme according to the hard disk power consumption calculation rule;

[0234] The target required power consumption of each hard disk in each hard disk power-on solution is calculated according to the hard disk power consumption calculation rule, including:

[0235] Obtain the standard power consumption data of each hard disk in each hard disk power-on solution and the preset target required power consumption calculation formula;

[0236] Obtain hard disk capacity data of each hard disk in each hard disk power-on solution, and calculate a capacity impact factor according to the hard disk capacity data of each hard disk;

[0237] Obtaining standard working data and actual working data of each hard disk in each hard disk power-on scheme, and calculating a working state influencing factor according to the standard working data and actual working data of each hard disk;

[0238] Obtaining the operation stage information of each hard disk in each hard disk power-on scheme, and calculating the operation stage influencing factor according to the operation stage information of each hard disk;

[0239] Obtaining work efficiency information of each hard disk in each hard disk power-on scheme, and calculating a work efficiency influencing factor according to the work efficiency information of each hard disk;

[0240] Calculate the target power consumption required for each hard disk based on the preset target power consumption calculation formula, standard power consumption data, capacity impact factor, working status impact factor, operation stage impact factor, and work efficiency impact factor;

[0241] The preset target required power consumption calculation formula is as follows:

[0242]

[0243] Among them, P i1 represents the target required power consumption of hard disk i, P i2 represents the standard power consumption of hard disk i, λ represents the work efficiency influencing factor, |V i | represents the working efficiency of hard disk i, X represents the capacity impact factor, B i Indicates the standard storage capacity of hard disk i, S i represents the used storage capacity of hard disk i, Y represents the working status influencing factor, C i Indicates the standard working data of hard disk i, M i represents the actual working data of hard disk i, Z represents the operating stage influencing factor, and T0 represents the operating stage information of hard disk i.

[0244] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0245] According to the hard disk power consumption corresponding to each hard disk power-on scheme, the hard disk peak startup time is set as follows:

[0246] Obtain the maximum power consumption of each operation phase corresponding to each hard disk power-on solution;

[0247] Set the startup time of each hard disk according to the target power consumption of each hard disk and the maximum power consumption of each operation stage;

[0248] According to the startup time of each hard disk, the staggered startup time corresponding to the hard disk power-on solution to which the hard disk belongs is obtained.

[0249] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0250] Obtain the target power-on requirement, and select the target hard disk power-on solution from multiple hard disk power-on solutions based on the target power-on requirement, hard disk off-peak startup time, and target required power consumption. The target hard disk power-on solution includes:

[0251] Obtain the target power-on requirement, and use the target power-on requirement as a constraint condition to obtain a hard disk power-on solution to be selected that meets the constraint condition from multiple hard disk power-on solutions;

[0252] Obtain the power consumption corresponding to each hard disk power-on solution to be selected, and the off-peak startup time corresponding to each hard disk power-on solution to be selected;

[0253] Calculate the average power consumption corresponding to each hard disk power-on solution to be selected, and calculate the variance corresponding to the hard disk power-on solution to be selected based on the average power consumption;

[0254] Calculate the priority of each hard disk power-on scheme to be selected based on the scheme corresponding to the hard disk power-on scheme to be selected and the staggered startup time corresponding to the hard disk power-on scheme to be selected;

[0255] The target hard disk power-on scheme is determined according to the priority of each hard disk power-on scheme to be selected.

[0256] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0257] The method also includes:

[0258] Get the maximum rated power of the power supply and the current system power consumption value;

[0259] Determine the remaining rated power of the power supply based on the difference between the maximum rated power of the power supply and the current system power consumption;

[0260] Detect the peak power of each hard disk after it is powered on, and the time it takes for the peak power of the hard disk to return to the power during stable operation;

[0261] Determine the actual peak power used by each hard disk after it is powered on based on the number of hard disks to be powered on and the peak power;

[0262] The remaining hard disks are powered on in batches according to the time taken for the peak power of the hard disks to return to the power during stable operation, wherein the peak power of each batch of hard disks after power-on, determined according to the actual peak power used by each hard disk after power-on and the power-on plan for each hard disk, is within the remaining rated power range of the power supply.

[0263] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0264] Before obtaining the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, the following steps are included:

[0265] In response to the application program acquiring the attribute information of the multiple hard disks on the server backplane; preprocessing the acquired attribute information of the multiple hard disks on the server backplane;

[0266] A machine learning algorithm is used to perform dimension reduction processing on the attribute information of multiple hard disks on the server backplane, and a multi-dimensional feature corresponding to each of the multiple hard disks on the server backplane is obtained;

[0267] The multi-dimensional features corresponding to each hard disk are integrated to generate the health index corresponding to each hard disk;

[0268] The hard disks whose health index is greater than the preset health index value are regarded as hard disks to be powered on.

[0269] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous forwarding path (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0270] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0271] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A hard disk power-on control method, characterized in that: include: Get the number of hard disks to be powered on and the attribute information of the hard disks to be powered on; Based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, a power-on scheme for the hard disks to be powered on is standardized to obtain multiple power-on schemes for the hard disks; Traverse multiple hard disk power-on schemes and calculate the target required power consumption of each hard disk in each hard disk power-on scheme; Setting a hard disk staggered startup time according to the target power consumption corresponding to each hard disk power-on solution; A target power-on requirement is obtained, a target hard disk power-on solution is selected from a plurality of hard disk power-on solutions according to the target power-on requirement, the hard disk off-peak startup time, and the target required power consumption, and the hard disk is powered on according to the target hard disk power-on solution.

2. The method according to claim 1, characterized in that The power-on scheme of the hard disks to be powered on is standardized based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and the multiple hard disk power-on schemes include: Extracting the type of the hard disk to be powered on and the location information of the hard disk to be powered on from the attribute information of the hard disk to be powered on; Setting the number of hard disks included in each hard disk power-on solution according to the number of hard disks to be powered on; Abstracting the hard disks to be powered on of the same type as the hard disks to be powered on as nodes, and using the positional relationship between the hard disks to be powered on of the same type as the hard disks to be powered on as paths to construct a network of hard disks to be powered on; Selecting a node corresponding to any hard disk to be powered on from the hard disk network to be powered on as a target node, and forming a hard disk group with multiple hard disks to be powered on that are connected to the target node path and meet the hard disk quantity requirement, until all the hard disks to be powered on are divided into multiple hard disk groups; The power-on sequence of each hard disk in each hard disk group is set in the form of permutations and combinations to obtain multiple hard disk power-on solutions.

3. The method according to claim 1, characterized in that The traversing multiple hard disk power-on schemes and calculating the target required power consumption of each hard disk in each hard disk power-on scheme comprises: Obtaining a preset hard disk power-on model, wherein the hard disk power-on model includes a target required power consumption calculation formula; Inputting the multiple hard disk power-on schemes into the hard disk power-on model, and calculating the target required power consumption of each hard disk in each hard disk power-on scheme according to the hard disk power consumption calculation rule; Calculating the target required power consumption of each hard disk in each hard disk power-on solution according to the hard disk power consumption calculation rule includes: Obtain the standard power consumption data of each hard disk in each hard disk power-on solution and the preset target required power consumption calculation formula; Obtaining hard disk capacity data of each hard disk in each hard disk power-on solution, and calculating a capacity impact factor according to the hard disk capacity data of each hard disk; Obtaining standard working data and actual working data of each hard disk in each hard disk power-on scheme, and calculating a working state influencing factor according to the standard working data and actual working data of each hard disk; Obtaining the operation stage information of each hard disk in each hard disk power-on scheme, and calculating the operation stage influencing factor according to the operation stage information of each hard disk; Obtaining work efficiency information of each hard disk in each hard disk power-on scheme, and calculating a work efficiency influencing factor according to the work efficiency information of each hard disk; Calculate the target required power consumption of each hard disk according to the preset target required power consumption calculation formula, the standard power consumption data, the capacity impact factor, the working state impact factor, the operating stage impact factor and the working efficiency impact factor; The preset target required power consumption calculation formula is as follows: Among them, P i1 represents the target required power consumption of hard disk i, P i2 represents the standard power consumption of hard disk i, λ represents the work efficiency influencing factor, |V i | represents the working efficiency of hard disk i, X represents the capacity impact factor, B i Indicates the standard storage capacity of hard disk i, S i represents the used storage capacity of hard disk i, Y represents the working status influencing factor, C i Indicates the standard working data of hard disk i, M i represents the actual working data of hard disk i, Z represents the operating stage influencing factor, and T0 represents the operating stage information of hard disk i.

4. The method according to claim 1, characterized in that: The step of setting the hard disk peak start time according to the hard disk power consumption corresponding to each hard disk power-on scheme comprises: Obtain the maximum power consumption of each operation phase corresponding to each hard disk power-on solution; Setting the startup time of each hard disk according to the target required power consumption of each hard disk and the maximum power consumption of each operation stage; According to the startup time of each hard disk, the off-peak startup time corresponding to the hard disk power-on solution to which the hard disk belongs is obtained.

5. The method according to claim 1, characterized in that The obtaining of the target power-on requirement and selecting a target hard disk power-on solution from a plurality of hard disk power-on solutions according to the target power-on requirement, the hard disk off-peak startup time and the target required power consumption include: Obtain the target power-on requirement, and use the target power-on requirement as a constraint condition to obtain a hard disk power-on solution to be selected that meets the constraint condition from multiple hard disk power-on solutions; Obtain the power consumption corresponding to each hard disk power-on solution to be selected, and the off-peak startup time corresponding to each hard disk power-on solution to be selected; Calculating an average value of power consumption corresponding to each hard disk power-on solution to be selected, and calculating a variance corresponding to the hard disk power-on solution to be selected based on the average value of power consumption; Calculating the priority of each hard disk power-on scheme to be selected based on the scheme corresponding to the hard disk power-on scheme to be selected and the staggered startup time corresponding to the hard disk power-on scheme to be selected; The target hard disk power-on scheme is determined according to the priority of each hard disk power-on scheme to be selected.

6. The method according to claim 1, characterized in that The method further comprises: Get the maximum rated power of the power supply and the current system power consumption value; Determining the remaining rated power of the power supply based on the difference between the maximum rated power of the power supply and the current system power consumption; Detect the peak power of each hard disk after it is powered on, and the time it takes for the peak power of the hard disk to return to the power during stable operation; Determine the actual peak power used by each hard disk after it is powered on based on the number of hard disks to be powered on and the peak power; The remaining hard disks are powered on in batches according to the time taken for the peak power of the hard disk to return to the power during stable operation, wherein the peak power of each batch of hard disks after power-on is determined according to the actual peak power used by each hard disk after power-on and the power-on plan for each hard disk, and is within the remaining rated power range of the power supply.

7. The method according to claim 1, characterized in that The step of obtaining the number of hard disks to be powered on and the attribute information of the hard disks to be powered on includes: In response to the application program acquiring the attribute information of the multiple hard disks on the server backplane; preprocessing the acquired attribute information of the multiple hard disks on the server backplane; A machine learning algorithm is used to perform dimension reduction processing on the attribute information of multiple hard disks on the server backplane, and a multi-dimensional feature corresponding to each of the multiple hard disks on the server backplane is obtained; The multi-dimensional features corresponding to each hard disk are integrated to generate the health index corresponding to each hard disk; The hard disks whose health index is greater than the preset health index value are regarded as hard disks to be powered on.

8. A hard disk power-on control device, characterized in that: The device comprises: An acquisition module is used to acquire the number of hard disks to be powered on and the attribute information of the hard disks to be powered on; A planning module, used to plan power-on schemes for the hard disks to be powered on based on the number of hard disks to be powered on and the attribute information of the hard disks to be powered on, and obtain multiple power-on schemes for the hard disks; A calculation module, used for traversing multiple hard disk power-on schemes and calculating the target required power consumption of each hard disk in each hard disk power-on scheme; A setting module, used for setting a hard disk staggered startup time according to a target required power consumption corresponding to each hard disk power-on scheme; The power-on module is used to obtain the target power-on requirement, select the target hard disk power-on solution from multiple hard disk power-on solutions according to the target power-on requirement, the hard disk off-peak startup time and the target required power consumption, and power on the hard disk according to the target hard disk power-on solution.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images