Hard disk starting method, bus adapter, server, device, medium and product

Abstract

The application discloses a hard disk starting method, a bus adapter, a server, equipment, a medium and a product, and relates to the technical field of data storage. A current threshold value corresponding to a current moment is determined based on a system current parameter and a current average parameter, the corresponding current threshold value is dynamically obtained according to different running stages, and the system current parameter at each real-time acquisition moment enjoys its own exclusive current threshold value. In the case that the system current parameter is in a preset range corresponding to the current threshold value, as long as the current is in the region, the hard disk will be subjected to starting control processing, and the stability of the system is ensured. According to a hard disk start-stop strategy and / or a hard disk dynamic performance adjustment strategy, the hard disk is subjected to adjustment of a starting sequence and / or starting control processing of a working state, the hard disk is subjected to starting control through different strategies, the instantaneous current demand of the hard disk is reduced, the current overload is reduced, the accuracy of triggering starting control is improved, and it is ensured that the system always operates in a safe and efficient state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data storage, in particular to a hard disk starting method, a bus adapter, a server, equipment, a medium and a product. BACKGROUND

[0002] When a server system mounts a large number of hard disks, if all the hard disks are started at the same time, the instantaneous current of the hard disks will be superimposed together, resulting in a multi-disk superposition effect, which causes the total instantaneous current to exceed the maximum carrying capacity of the power module of the server system, resulting in system overload problems, and further causing system failure.

[0003] Therefore, how to avoid system overload to improve the hard disk starting efficiency is an urgent problem to be solved by those skilled in the art. SUMMARY

[0004] The purpose of the present application is to provide a hard disk starting method, a bus adapter, a server, equipment, a medium and a product to solve the problem of system overload caused by the superposition of instantaneous current when all hard disks are started at the same time, and to further improve the hard disk starting efficiency.

[0005] To solve the above technical problems, the present application provides a hard disk starting method, comprising:

[0006] obtaining a system current parameter corresponding to the current time in the running stage of the hard disk and a current average parameter corresponding to a preset time before the current time as a reference;

[0007] determining a current threshold corresponding to the current time based on the system current parameter and the current average parameter;

[0008] when the system current parameter is in a preset range corresponding to the current threshold, adjusting the starting order and / or the working state of the hard disk according to the hard disk start-stop strategy and / or the hard disk dynamic performance adjustment strategy to complete the hard disk starting process; wherein the hard disk dynamic performance adjustment strategy is a data transmission rate adjustment strategy of the hard disk.

[0009] On the one hand, determining a current threshold corresponding to the current time based on the system current parameter and the current average parameter comprises:

[0010] performing a second-order derivative processing on the system current parameter to obtain a current change gradient parameter;

[0011] determining a safety margin parameter according to the current change gradient parameter, an adjustment coefficient and a margin factor;

[0012] determining a current threshold corresponding to the current time according to the current average parameter and the safety margin parameter.

[0013] In another aspect, when the system current parameter is in a preset range corresponding to the current threshold, the hard disk is adjusted in starting sequence and / or working state according to the hard disk start-stop strategy and / or the hard disk dynamic performance adjustment strategy, including:

[0014] When the system current parameter exceeds the current threshold and is in a first preset range, the hard disk is adjusted in starting sequence according to the hard disk start-stop strategy;

[0015] When the system current parameter exceeds the current threshold and is in a second preset range, the hard disk is adjusted in starting sequence and working state according to the hard disk start-stop strategy and the hard disk dynamic performance adjustment strategy; wherein the second preset range is greater than the first preset range;

[0016] When the system current parameter does not exceed the current threshold and is in a third preset range, the hard disk is adjusted in working state according to the hard disk dynamic performance adjustment strategy; wherein the third preset range is less than the first preset range.

[0017] In another aspect, when the system current parameter is in a preset range corresponding to the current threshold, the hard disk is adjusted in starting sequence and / or working state according to the hard disk start-stop strategy and / or the hard disk dynamic performance adjustment strategy, including:

[0018] Obtaining the closeness between the system current parameter and the current threshold;

[0019] Adjusting the parameters corresponding to the hard disk start-stop strategy and / or the hard disk dynamic performance adjustment strategy according to the relationship between the closeness and the preset closeness to obtain the adjusted hard disk start-stop strategy and / or the adjusted hard disk dynamic performance adjustment strategy;

[0020] Adjusting the hard disk in starting sequence and / or working state according to the adjusted hard disk start-stop strategy and / or the adjusted hard disk dynamic performance adjustment strategy.

[0021] In another aspect, the generation process of the hard disk start-stop strategy includes:

[0022] Determining a first target hard disk in the hard disk;

[0023] Obtaining a second target hard disk in the first target hard disk, and suspending the starting process of the second target hard disk;

[0024] Resuming the starting control process of the second target hard disk based on the system current parameter falling to a target safety threshold.

[0025] On the other hand, the process of generating the hard disk dynamic performance tuning strategy includes:

[0026] Verify that the hard drive is booting.

[0027] Obtain the critical queue depth corresponding to each of the started hard drives;

[0028] The processed queue depth is obtained by limiting the critical queue depth for each of the already started hard drives.

[0029] The activated hard disk is slowed down based on the processed queue depth to limit the number of commands issued.

[0030] Perform boot control processing on the already booted hard drive after speed reduction.

[0031] On the other hand, the process of generating the hard disk dynamic performance tuning strategy includes:

[0032] Get the time interval between read and write commands for the already started hard drive;

[0033] Idle time is inserted into the read / write command time interval to reduce the duty cycle of the started hard disk;

[0034] The boot control process is then performed on the booted hard drive after the downgrade.

[0035] On the other hand, when performing boot control processing to adjust the boot order and / or working status of the current batch of hard drives according to the hard drive start / stop strategy and / or the hard drive dynamic performance adjustment strategy, the method further includes:

[0036] Get the number of hard drives in the current batch;

[0037] The number of hard drives and the number of power supply modules corresponding to each hard drive are determined based on the number of hard drives.

[0038] Phase dispersion compensation is performed based on the hard disk index and the number of power supply modules to obtain the current phase corresponding to each power supply module.

[0039] Power supply is applied to the hard drives in the current batch according to the current phase described.

[0040] On the other hand, when the first target hard drive is a non-bootable hard drive, the process of acquiring the second target hard drive includes:

[0041] Obtain the slot number, service importance, and power consumption level of the unbooted hard drive;

[0042] Assign a first weighting coefficient, a second weighting coefficient, and a third weighting coefficient according to the slot number, the importance of the service, and the power consumption level;

[0043] The first hard disk parameters corresponding to each of the unstarted hard disks are determined based on the slot number, the first weight coefficient, the importance of the service, the second weight coefficient, the power consumption level, and the third weight coefficient.

[0044] Filter the target first hard disk parameters that are less than the preset hard disk parameters from each of the first hard disk parameters;

[0045] The unbooted hard drive corresponding to the first target hard drive parameters is used as the second target hard drive.

[0046] On the other hand, when all the hard drives are already powered on, and the first target hard drive is a powered-on hard drive, the process of acquiring the second target hard drive includes:

[0047] Get the queue depth, data read / write tasks, and activity status of the started hard drives;

[0048] Within the already started hard drives, a second target hard drive is selected based on the following: the queue depth has not reached the preset queue depth, the data read / write task is a non-real-time data read / write task, and the active state is an idle state.

[0049] On the other hand, the triggering mechanism for restoring the second target hard drive to perform boot control processing includes:

[0050] Obtain the first coefficient of recovery;

[0051] The target safety threshold is determined based on the first recovery coefficient and the current threshold.

[0052] The rate of change of current is obtained by processing the first derivative of the system current parameters.

[0053] When the absolute value of the current change rate is less than or equal to the preset current change rate, and the system current parameter drops to the target safety threshold, the second target hard disk is triggered to perform boot control processing.

[0054] On the other hand, the recovery process of the second target hard drive includes:

[0055] Obtain the peak startup current value of the second target hard drive; sort the peak startup current values ​​from smallest to largest to obtain the sorted second target hard drives; and perform recovery boot according to the sorted second target hard drives;

[0056] Alternatively, the second target hard disk is grouped to obtain grouped second target hard disks; the grouped second target hard disks are then restarted according to a balancing strategy; wherein, the balancing strategy is a strategy in which the sum of the startup currents of any group of the grouped second target hard disks is less than or equal to the system rated current value.

[0057] Alternatively, obtain the slot number of the second target hard drive; perform phase allocation processing based on the slot number and the equivalent hard drives in the same batch corresponding to the second target hard drive to restore the second target hard drive.

[0058] To address the aforementioned technical problems, this application also provides a bus adapter, which includes a processor, a first protocol controller, a second protocol controller, a current acquisition module, a power supply module, and a phase compensation controller.

[0059] The processor is connected to a first protocol controller, a second protocol controller, a current acquisition module, and a power supply module.

[0060] The power supply module and the phase compensation controller are connected;

[0061] Both the second protocol controller and the phase compensation controller are connected to a hard disk;

[0062] The processor is used to execute the steps of the hard disk boot method described above.

[0063] On one hand, the processor includes a first processor and a second processor;

[0064] The first processor is connected to the first protocol controller, the current acquisition module, and the second processor;

[0065] The second processor is connected to the second protocol controller and the power supply module.

[0066] On the other hand, the first processor and the second processor boot according to a first preset boot sequence during the hard disk boot process; wherein, the process of determining the first preset boot sequence includes:

[0067] Obtain the first current peak, the first preset startup time percentage, and the first inherent startup time corresponding to the first processor, and the second current peak, the second preset startup time percentage, and the second inherent startup time corresponding to the second processor; wherein, the sum of the first preset startup time percentage and the second preset startup time percentage is equal to 1;

[0068] A first startup parameter is determined based on the first current peak value and the first inherent startup time; a second startup parameter is determined based on the penalty term coefficient, the first startup parameter, and the critical allowable current; a first startup efficiency is determined based on the first preset startup time percentage and the preset startup time of the first processor; and a first actual startup time percentage corresponding to the first processor is determined based on the first startup efficiency and the second startup parameter.

[0069] A third startup parameter is determined based on the second current peak value and the second inherent startup time; a fourth startup parameter is determined based on the third startup parameter, the penalty term coefficient, the first startup parameter, and the critical allowable current; a second startup efficiency is determined based on the second preset startup time percentage and the preset startup time of the second processor; the first startup efficiency and the second startup efficiency are summed to obtain a third startup efficiency; and a second actual startup time percentage corresponding to the second processor is determined based on the third startup efficiency and the fourth startup parameter.

[0070] The first preset startup sequence of the first processor and the second processor is determined based on the first actual startup time percentage and the second actual startup time percentage.

[0071] On the other hand, the second processor includes at least one sub-processor; the number of sub-processors is the same as the number of hard disk batches; each sub-processor is connected to the first processor and the power supply module; each sub-processor corresponds to a hard disk connection; each sub-processor starts up according to a second preset startup sequence during the hard disk startup process.

[0072] To address the aforementioned technical problems, this application also provides a server, including a host, a hard disk, and the aforementioned bus adapter;

[0073] The host, the bus adapter, and the hard disk are connected in sequence.

[0074] To address the aforementioned technical problems, this application also provides an electronic device, comprising:

[0075] Memory, used to store computer programs;

[0076] A processor for executing the computer program to implement the steps of the hard disk boot method as described.

[0077] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the hard disk boot method as described above.

[0078] To address the aforementioned technical problems, this application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the hard disk boot method.

[0079] The beneficial effect of this application lies in providing a hard drive booting method. First, it acquires the system current parameters corresponding to the current moment of the hard drive's current operating phase. These system current parameters are real-time collected data of the total system current at the current moment. The hard drive booting process includes three operating phases: initial, stable, and operational. This application collects the current total system current data in real-time during any booting phase. It also acquires the average current parameters corresponding to a preset time period prior to the current moment, focusing not only on the current moment's data but also on the average current parameters over a past period from the preset time to the current moment. This ensures that the subsequently determined current threshold is a threshold that combines the changing trends between the current moment and the preset time, guaranteeing monitoring accuracy. Second, it determines the current threshold corresponding to the current moment based on the system current parameters and the average current parameters. Compared to conventional schemes that use data collected at fixed preset time intervals and compared with a fixed threshold for boot control, some hard drives that complete booting prematurely must wait for the next preset time interval to arrive before booting again, which prolongs the entire boot process and reduces boot efficiency. This application acquires data in real time. Startup is initiated only when the acquired data falls within a preset range compared to the corresponding current threshold, eliminating the need to wait for the next preset fixed time interval, thus improving the flexibility of the startup process. Furthermore, the current threshold in this application is not a conventional fixed threshold. Considering the different current changes during the hard drive's startup phase, the current threshold is relatively high in the initial startup phase due to the rapid current increase, and relatively low in the stable phase due to smaller current fluctuations. This application dynamically obtains the corresponding current threshold based on the average current parameter and the current change trend under the system current parameter, according to different operating stages, ensuring that the system current parameter at each real-time acquisition moment has its own dedicated current threshold. Finally, compared to a single current threshold, where small fluctuations in the system current parameter within a single threshold frequently cause it to jump between allowed and prohibited startup, this application ensures system stability by implementing startup control processing as long as the current falls within the preset range corresponding to the current threshold, provided the current is within that range. This application employs a boot control process that adjusts the boot order and / or operating state of hard drives according to hard drive start / stop policies and / or hard drive dynamic performance adjustment policies. Different policies are used to control the hard drive boot process: the start / stop policy adjusts the boot order, and the dynamic performance adjustment policy adjusts the operating state. Compared to conventional solutions where all hard drives are started simultaneously, this application specifically adjusts the boot order to stagger their startup times. Adjusting the operating state reduces the instantaneous current demand of the hard drives, thereby mitigating current overload and improving the flexibility and versatility of boot control. It also enhances the accuracy of triggering boot control, ensuring the system always operates in a safe and efficient state.

[0080] Secondly, the current threshold is obtained in real time at the current moment, closely following the envelope of the actual current curve. This prevents false triggering during sharp current increases and avoids excessive leniency during stable current periods, thus maintaining protection. The selection of hard drive start-stop and dynamic performance adjustment strategies is based on the system current parameters and their proximity to or exceedance of the current threshold. This achieves graded response, avoiding performance waste due to over-defense, and deep current limiting to ensure system survivability under extreme loads. It maximizes the utilization of power supply capacity while ensuring the absolute safety of the storage system. The selection of hard drive start-stop and dynamic performance adjustment strategies uses the degree of proximity to adjust specific parameters under each strategy, generating a smooth adjustment coefficient. This maximizes the number of parallel startups and operating performance, significantly improving the overall system throughput and resource utilization. The generation process of the hard drive start-stop strategy considers the hard drive's startup state, generating the strategy in both non-started and started states, improving the flexibility and diversity of hard drive start-stop strategy generation. Prioritizing the suspension of lower-priority hard drives that have not yet been started, and using a weighted selection process based on three dimensions—slot number, business importance, and power consumption level—combined with weighting coefficients, essentially upgrades hard drive startup control from a simple one-size-fits-all logic to a multi-dimensional, customizable intelligent decision-making system. This system can achieve a balance between ensuring core business operations, maintaining physical security, and preserving system order.

[0081] In addition, this application also provides a bus adapter, server, electronic device, medium, and product that have the same beneficial effects as the hard disk boot method described above. Attached Figure Description

[0082] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0083] Figure 1 A flowchart of a hard disk boot method provided in an embodiment of this application;

[0084] Figure 2 This is a schematic diagram of a conventional SATA HDD connection method provided in an embodiment of this application;

[0085] Figure 3 This is a schematic diagram of the structure of a bus adapter provided in an embodiment of this application;

[0086] Figure 4 A flowchart illustrating another hard disk boot method provided in this application embodiment;

[0087] Figure 5 This is a structural diagram of a hard disk boot device provided in an embodiment of this application;

[0088] Figure 6 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0089] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0090] The core of this application is to provide a hard disk boot method, bus adapter, server, device, media, and product to solve the problem of system overload caused by the superposition of instantaneous currents when all hard disks start simultaneously, thereby reducing hard disk boot efficiency.

[0091] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0092] Non-Volatile Memory Express (NVMe) is a high-performance storage protocol widely used in modern computing systems. It connects directly to the Central Processing Unit (CPU) via the Peripheral Component Interconnect Express (PCIe) interface, eliminating the Advanced Host Controller Interface (AHCI) protocol translation layer of Serial Advanced Technology Attachment (SATA). This reduces input / output (I / O) latency and provides higher data transfer rates than traditional SATA interfaces. NVMe optimizes and simplifies command processing, reducing unnecessary overhead and improving I / O efficiency. Furthermore, NVMe supports a maximum queue depth of 64K, enabling it to handle more I / O requests simultaneously and fully utilize the performance of multi-core processors.

[0093] The SATA interface and its AHCI protocol were originally designed for hard disk drives (HDDs), supporting only a single queue and a queue depth of 32. In server applications, the SATA interface and AHCI protocol are gradually being phased out. Many manufacturers have abandoned the SATA interface in their new generation of storage servers, opting instead for the PCIe interface, which supports the NVMe protocol. However, HDDs, due to their large capacity, mature technology, and low unit storage cost, are still widely used in data centers, video surveillance, historical data archiving, and other high-capacity cold storage scenarios.

[0094] Due to its inherent mechanical structure, a hard disk drive (HDD) incorporates a spindle motor that drives the platters to rotate at high speed, allowing the read / write heads to hover above the platters for data reading and writing. During the startup phase, the spindle motor must overcome static friction and platter inertia, resulting in an instantaneous current that can be more than three times the operating power consumption. The duration and peak value of this current are closely related to the HDD's rotational speed and mechanical design. A typical HDD current curve consists of three phases: an initial phase (approximately 0-0.5 seconds), where the current surges to its peak value to accelerate the spindle motor; a steady phase (approximately 0.5-3 seconds), where the current gradually decreases from the peak to the operating current, and the platters gradually reach their rated speed; and an operating phase (after 3 seconds), where the current remains relatively stable, and the HDD begins normal read / write operations. The peak current during the initial phase can be three times or more than the current during the operating phase. When a server system has a large number of hard drives, if all drives start simultaneously, the combined instantaneous current from multiple drives may exceed the maximum capacity of the system's power module, potentially causing system failure. Therefore, the hard drive startup method provided in this application can solve the aforementioned technical problems.

[0095] Figure 1 A flowchart of a hard disk boot method provided in this application embodiment is shown below. Figure 1 As shown, it includes:

[0096] S11: Obtain the system current parameters corresponding to the current moment of the hard disk at the current operating stage and the average current parameters corresponding to the preset moments before the current moment;

[0097] S12: Determine the current threshold corresponding to the current moment based on the system current parameters and the average current parameters;

[0098] S13: When the system current parameter is within the preset range corresponding to the current threshold, the hard drive is adjusted in accordance with the hard drive start-stop strategy and / or hard drive dynamic performance adjustment strategy to complete the hard drive start-up process.

[0099] Among them, the hard drive dynamic performance adjustment strategy is a strategy for adjusting the hard drive's data transfer rate.

[0100] Specifically, the operational phases of a hard drive mainly consist of the initial phase, the stable phase, and the working phase during the hard drive's boot process. The initial phase is the process after the hard drive is powered on, where the spindle motor drives the platters to accelerate from a standstill until they reach their rated speed (such as 5400 revolutions per minute, 7200 RPM, or 10000 RPM). The stable phase is the state where the hard drive platters have reached their rated speed and are maintaining a constant rotation, the read / write heads are loaded and positioned on the landing zone or servo rails, and the circuit system is initialized and ready to receive commands. The working phase is the active state where the hard drive is executing data transfer commands sent by the host, with the read / write head rapidly moving above the platters (seeking) and reading or writing data.

[0101] A three-stage current mathematical model is established based on the three operational phases described above to achieve millisecond-level load prediction. This three-stage current mathematical model is based on the boot and operation characteristics of HDD hard drives, creating a precise mathematical model.

[0102] In the initial stage ( The expression for current is:

[0103] ; (1)

[0104] in, To generate peak starting current, The attenuation coefficient is used to simulate the rapid current drop during the initial startup phase.

[0105] Stable phase ( Under these conditions, the current expression is:

[0106] ; (2)

[0107] Wherein, is the linear decreasing coefficient, reflecting the trend of the current gradually stabilizing.

[0108] During the work phase ( The current remains at a stable operating current.

[0109] It should be noted that the time thresholds corresponding to the above three operating stages can be flexibly set according to the actual situation, and are not limited here. The three-stage current model (initial stage, stable stage, and working stage) is a priori knowledge model used to describe the typical current behavior when the HDD starts up. Its function is to provide a stage identification basis for the dynamic current management engine; to help determine the value of the adjustment coefficient k in the current threshold formula (the value of k is different in different stages) as a reference benchmark for predicting the current change trend; the model itself does not directly output the threshold, but tells the system which stage it is currently in and how the expected current will change.

[0110] The system current parameters corresponding to the current moment during the hard drive's operation phase are obtained. In this application, the total system current is statistically analyzed. The "current moment" here is a flexibly adjusted moment based on the entire hard drive startup process, compared to the preset fixed time intervals in conventional schemes. This application collects data in real-time. The average current parameters corresponding to preset moments prior to the current moment are obtained. This is to consider the trend of current changes, requiring the acquisition of average current parameters over a past period based on the current moment.

[0111] In step S12, the current threshold is determined based on system current parameters and average current parameters. This improves accuracy compared to conventional solutions that use fixed thresholds for all operating stages or for each specific operating stage. In this application, considering the rapid current increase during the initial hard drive startup phase, the current threshold needs to be appropriately increased to avoid misjudgments. During stable operation, the current threshold is reduced based on the actual current fluctuation range to improve monitoring accuracy. Therefore, this application calculates the current threshold based on real-time acquired system current parameters. Even for hard drives of the same model, actual operating currents can differ due to variations in aging, temperature, and manufacturing tolerances. The real-time threshold is dynamically calculated based on the current trend acquired at the current moment. Real-time acquisition and determination of the threshold can capture microsecond-level current fluctuations.

[0112] Regarding the process of determining the current threshold, the trend of current change can be known by taking the derivative of the system current parameters, and the current threshold at the current moment can be determined by adding the average current parameters.

[0113] In step S13, if the system current parameters are within a preset range corresponding to the current threshold, this preset range includes not only the current threshold but also a hysteresis range. When the system current parameters are within this range, hard drive startup processing will be initiated to ensure system stability. Furthermore, for different degrees of proximity within the preset range, corresponding startup control processing strategies can be set according to these degrees of proximity. Different parameter adjustments under the same startup control processing strategy can also be set based on different degrees of proximity; this is not limited here. Additionally, the number of startup control processing strategies can be one or more; this is not limited here.

[0114] Under a single startup control strategy, multiple startup control strategies can be selected, meaning only one strategy can be chosen at any given time. The selection process is based on the degree to which system current parameters approach or exceed current thresholds. For example, if the system current parameters significantly exceed the threshold and fall within a certain range, a rapid reduction is needed. In this case, the hard drive startup / shutdown strategy aims to pause hardware startup, avoid startup peaks, directly cut off the largest current spike, and eliminate the load at its source. The hard drive startup strategy here can disable the startup of some hardware; this "some hardware" can refer to hard drives that are already running or those that haven't yet started—this is not limited here. The hard drive dynamic performance adjustment strategy limits data transfer rates and reduces dynamic operating current, while still maintaining a basic sustaining current to prevent overheating, refine power management, and avoid stuttering. Of course, other strategies can also be implemented—this is not limited here. The two startup control strategies mentioned above address different dimensions and can complement or operate independently of each other—this is not limited here.

[0115] It's worth noting that the essence of hard drive start / stop strategies is to adjust the boot order, pausing the boot process of some hard drives and then gradually restarting them in the optimized order once the system current parameters decrease. Hard drive dynamic performance tuning strategies adjust the operating state, reducing the data transfer rate of currently active hard drives to decrease instantaneous current demands.

[0116] The beneficial effect of this application's embodiments lies in providing a hard drive booting method. First, it acquires the system current parameters corresponding to the current moment of the hard drive's current operating phase. These system current parameters are real-time collected data of the total system current at the current moment. The hard drive booting phase includes three operating phases: initial, stable, and operational. This application collects the current total system current data in real-time during any booting phase. It also acquires the average current parameters corresponding to a preset time period prior to the current moment, focusing not only on the current moment's data but also on the average current parameters over a past period from the preset time to the current moment. This ensures that the subsequently determined current threshold is a threshold that combines the changing trends between the current moment and the preset time, guaranteeing monitoring accuracy. Second, it determines the current threshold corresponding to the current moment based on the system current parameters and the average current parameters. Compared to conventional schemes that use data collected at fixed preset time intervals and compared with a fixed threshold for boot control processing, some hard drives that complete booting early must wait until the next preset time interval arrives before booting again, which prolongs the entire boot process and reduces boot efficiency. This application acquires data in real time. Startup is initiated only when the acquired data falls within a preset range compared to the corresponding current threshold, eliminating the need to wait for the next preset fixed time interval, thus improving the flexibility of the startup process. Furthermore, the current threshold in this application is not a conventional fixed threshold. Considering the different current changes during the hard drive's startup phase, the current threshold is relatively high in the initial startup phase due to the rapid current increase, and relatively low in the stable phase due to smaller current fluctuations. This application dynamically obtains the corresponding current threshold based on the average current parameter and the current change trend under the system current parameter, according to different operating stages, ensuring that the system current parameter at each real-time acquisition moment has its own dedicated current threshold. Finally, compared to a single current threshold, where small fluctuations in the system current parameter within a single threshold frequently cause it to jump between allowed and prohibited startup, this application ensures system stability by implementing startup control processing as long as the current falls within the preset range corresponding to the current threshold, provided the current is within that range. This application employs a boot control process that adjusts the boot order and / or operating state of hard drives according to hard drive start / stop policies and / or hard drive dynamic performance adjustment policies. Different policies are used to control the hard drive boot process: the start / stop policy adjusts the boot order, and the dynamic performance adjustment policy adjusts the operating state. Compared to conventional solutions where all hard drives are started simultaneously, this application specifically adjusts the boot order to stagger their startup times. Adjusting the operating state reduces the instantaneous current demand of the hard drives, thereby mitigating current overload and improving the flexibility and versatility of boot control. It also enhances the accuracy of triggering boot control, ensuring the system always operates in a safe and efficient state.

[0117] In some embodiments, determining the current threshold at the current moment based on system current parameters and average current parameters includes:

[0118] The current change gradient parameters are obtained by processing the system current parameters using the second derivative.

[0119] The safety margin parameters are determined based on the current change gradient parameters, adjustment coefficients, and redundancy factors.

[0120] The current threshold corresponding to the current moment is determined based on the average current parameter and the safety margin parameter.

[0121] Specifically, the system current parameters are differentiated using the second derivative. The first derivative reflects the rate of current change; a large first derivative indicates drastic current fluctuations, potentially nearing a dangerous level. The second derivative, however, reflects the acceleration of the current rate of change. If the second derivative suddenly increases when the current is just beginning to rise and is still small, it indicates an impending explosive increase (like the surge at motor startup). The system can intervene before the current poses a threat (e.g., by reducing voltage or limiting the duty cycle), achieving a soft landing within milliseconds or even microseconds, rather than waiting for the current to spike before abruptly stopping.

[0122] Because the first derivative only shows the current fluctuations, it's difficult to distinguish between normal startup fluctuations and mechanical faults. However, the second derivative shows the actual startup current changes, allowing monitoring of the continuity of current changes and confirming that the current rise is a genuine load demand, rather than sensor noise.

[0123] The formula for calculating the current threshold is as follows:

[0124] ; (3)

[0125] in, The current threshold value at the current moment. The current average parameters are the values ​​corresponding to preset times prior to the current time. This is the gradient parameter for current change, reflecting the trend of current change. A positive and large value indicates that the current is accelerating, while a negative value indicates that the upward trend is slowing down. To adjust the coefficients, dynamic values ​​are selected based on different operational stages; This is a redundancy factor used to reserve a certain current margin to ensure the safe operation of the system; This is a safety margin parameter.

[0126] It should be noted that the above formula (3) is calculated dynamically in real time, not by calculating a fixed threshold for each stage. It is executed once at each sampling time and outputs the current threshold at the current time. .

[0127] Regarding the adjustment coefficient, the current stage is determined based on the current current value, current change characteristics, and the expectations of the three-stage model. Initial stage (0~0.5 seconds): Choose a larger value (e.g., 1.5~2.0) because the current rises sharply, requiring a higher threshold to avoid false triggering; stabilization phase (0.5~3 seconds): When a moderate value is selected (e.g., 1.0~1.2), the current decrease slows down; operating phase (≥3 seconds): Using a smaller value (such as 0.8~1.0) ensures a stable current and tightens the threshold to improve monitoring accuracy.

[0128] The current threshold provided in this embodiment is obtained in real time at the current moment. The current threshold can closely follow the envelope of the actual current curve, so that it will not be falsely triggered when the current rises sharply, nor will it be too loose and lose its protective function when the current is stable.

[0129] In some embodiments, when the system current parameter is within a preset range corresponding to the current threshold, a startup control process is performed to adjust the startup order and / or working state of the hard drive according to the hard drive start / stop strategy and / or hard drive dynamic performance adjustment strategy, including:

[0130] If the system current parameter exceeds the current threshold but is within the first preset range, the hard drive will be adjusted to control the boot order according to the hard drive start / stop policy.

[0131] If the system current parameter exceeds the current threshold but is within the second preset range, the hard drive will be adjusted in terms of boot order and working status according to the hard drive start-stop strategy and hard drive dynamic performance adjustment strategy; wherein, the second preset range is greater than the first preset range.

[0132] If the system current parameter does not exceed the current threshold and is within the third preset range, the hard drive will be adjusted according to the hard drive dynamic performance adjustment strategy to start the control process; wherein the third preset range is smaller than the first preset range.

[0133] Specifically, when the system current parameter exceeds the current threshold but is within the first preset range, it indicates that the system current parameter exceeds the current threshold. In this case, a hard drive boot strategy is used to adjust the boot order of the hard drives to quickly reduce current overload. However, if the current parameter exceeds the current threshold but is outside the first preset range but within the second preset range, it indicates that the current threshold is exceeded significantly. In this case, the hard drive boot strategy needs to be combined with a hard drive dynamic performance adjustment strategy to adjust the operating state.

[0134] If the system current parameter does not exceed the current threshold and is within the third preset range, it means that the system current parameter is close to the current threshold but does not exceed it. Therefore, the hard drive dynamic performance adjustment strategy is used to adjust the working state to prevent the current from rising further.

[0135] The hard drive start / stop strategy and hard drive dynamic performance adjustment strategy provided in this embodiment are selected based on the system current parameters and the degree to which they approach or exceed the current threshold, to achieve graded response, avoid performance waste caused by excessive defense, and ensure system survivability under extreme loads by deep current limiting; it maximizes the use of the power supply capacity while ensuring the absolute safety of the storage system.

[0136] In some embodiments, when the system current parameter is within a preset range corresponding to the current threshold, a startup control process is performed to adjust the startup order and / or working state of the hard drive according to the hard drive start / stop strategy and / or hard drive dynamic performance adjustment strategy, including:

[0137] To determine the approximation between system current parameters and current thresholds;

[0138] The parameters corresponding to the hard drive start-stop strategy and / or hard drive dynamic performance adjustment strategy are adjusted based on the relationship between the proximity level and the preset proximity level to obtain the adjusted hard drive start strategy and / or hard drive dynamic performance adjustment strategy.

[0139] The boot control process adjusts the boot order and / or operating status of the hard drive based on the adjusted hard drive boot strategy and / or the adjusted hard drive dynamic performance adjustment strategy.

[0140] Specifically, regarding the proximity between system current parameters and current thresholds, the proximity of system current parameters to the current threshold can be set to a negative value, while the proximity of system current parameters exceeding the current threshold can be set to a positive value. The relationship between the obtained positive and negative proximity values ​​and the preset proximity values ​​is used to adjust the parameters corresponding to the hard drive start-up and / or hard drive dynamic performance adjustment strategies, resulting in the adjusted hard drive startup strategy and / or adjusted hard drive dynamic performance adjustment strategy. The specific adjustment process here only adjusts the magnitude of the parameters involved in each strategy; if the proximity is high, the corresponding parameter is adjusted larger. The level of proximity reflects the level of the warning; a warning level results in a slight speed reduction, while an emergency level results in a deep speed reduction. The hard drive startup control is then performed based on the adjusted strategy.

[0141] The hard drive start / stop strategy and hard drive dynamic performance adjustment strategy provided in this embodiment are selected by adjusting the specific parameters under the specific strategy according to the degree of similarity, generating a smooth adjustment coefficient, maximizing the number of parallel startups and running performance, and significantly improving the overall throughput and resource utilization of the system.

[0142] In some embodiments, the process of generating the hard disk start / stop policy includes:

[0143] Identify the first target hard drive in the hard drive list;

[0144] Obtain the second target hard drive from the first target hard drive, and suspend the boot process on the second target hard drive;

[0145] If the system current parameter drops to the target safety threshold, the second target hard drive will be restored for boot control processing.

[0146] Specifically, the boot process for some hard drives is paused. Once the system current parameters drop to the target safety threshold, the gradual boot process for the paused hard drives is resumed. A first target hard drive is identified; this can be either a bootless or already booted hard drive, with subsequent start / stop strategies tailored to its boot status. Among the bootless hard drives, a second target hard drive is selected based on priority. Booting of the second target hard drive is paused, and only after the pause time is extended and the system current parameters drop to the target safety threshold (i.e., stability conditions are met) can the boot process for the second target hard drive be resumed.

[0147] Alternatively, suspend non-critical hard drives among the already started hard drives, determine the second target hard drive by queue depth, read / write tasks, and hard drive activity status, and suspend its startup process. Similarly, after extending the pause time, the system current parameters drop to the target safety threshold, that is, after the stability condition is met, the startup operation on the second target hard drive can be resumed.

[0148] The hard drive start / stop policy generation process provided in this embodiment takes into account the hard drive's startup state, generating the policy in both the non-started and started states, thereby improving the flexibility and diversity of hard drive start / stop policy generation.

[0149] In some embodiments, when the first target hard disk is a non-bootable hard disk, the process of acquiring the second target hard disk includes:

[0150] Obtain the slot number, business importance, and power consumption level of the unbooted hard drive;

[0151] The first, second, and third weighting coefficients are assigned according to the slot number, the importance of the service, and the power consumption level.

[0152] The parameters of the first hard drive corresponding to each unbooted hard drive are determined based on the slot number, the first weight coefficient, the importance of the business, the second weight coefficient, the power consumption level, and the third weight coefficient.

[0153] Filter the target first hard drive parameters that are less than the preset hard drive parameters from each set of first hard drive parameters;

[0154] Use the unbooted hard drive corresponding to the parameters of the first target hard drive as the second target hard drive.

[0155] Specifically, when the first target hard drive is a non-bootable hard drive, the lower-priority non-bootable hard drives are suspended first. This can be set based on the following dimensions: Slot number of the non-bootable hard drive (if the slot numbers are in physical order, the later-booted drives are suspended first); Business importance (e.g., pre-defining certain hard drives for critical data and prioritizing their boot); Power consumption level (hard drives with higher power consumption can have their boot delayed to avoid current accumulation).

[0156] Then, based on the slot number, the importance of the service, and the power consumption level, the corresponding first weight coefficient, second weight coefficient, and third weight coefficient are assigned, thereby obtaining the first hard disk parameter corresponding to each non-booted hard disk. The calculation of the first hard disk parameter corresponding to each non-booted hard disk is as follows: Slot number × first weight coefficient + importance of service × second weight coefficient + power consumption level × third weight coefficient = first hard disk parameter.

[0157] Select the target first hard disk parameter that is less than the preset hard disk parameter from the various first hard disk parameters, and use the unbooted hard disk corresponding to the target first hard disk parameter as the second target hard disk.

[0158] This embodiment prioritizes suspending the lower-priority hard drives that have not yet been started. It uses three dimensions—slot number, business importance, and power consumption level—combined with weighting coefficients for weighted filtering. Essentially, it upgrades hard drive startup control from a simple one-size-fits-all logic to a multi-dimensional, customizable intelligent decision-making system, which can achieve a balance between ensuring core business, maintaining physical security, and maintaining system order.

[0159] In other embodiments, when all hard drives are already powered on and the first target hard drive is a powered-on hard drive, the process of acquiring the second target hard drive includes:

[0160] Get the queue depth, data read / write tasks, and activity status of the started hard drives;

[0161] The second target hard drive is selected from those whose queue depth has not reached the preset queue depth, whose data read / write tasks are non-real-time data read / write tasks, and whose active state is idle.

[0162] Specifically, if all hard drives are powered on but the operating current is too high, the hard drives performing non-critical I / O will be paused. Non-critical hard drives can be determined by queue depth, data read / write tasks, and activity status. These three dimensions use an AND logic relationship: among the powered-on hard drives, the second target hard drive is selected based on its queue depth not reaching the preset queue depth, its data read / write tasks being non-real-time data read / write tasks, and its activity status being idle.

[0163] I / O queue depth is the most intuitive indicator of how busy a hard drive is. A low queue depth (e.g., close to 0) means that the hard drive is currently in a state of slight idleness or waiting for instructions.

[0164] This embodiment provides a method to suspend non-critical I / O operations on hard drives when all hard drives are powered on but the operating current is too high. The system understands the underlying business intent of the hard drives through I / O queues and task attributes. This allows the system to sacrifice the lowest priority background tasks to preserve the highest priority user experience in critical situations of current overload, achieving true intelligent power management.

[0165] In some embodiments, the triggering mechanism for restoring the second target hard disk to perform boot control processing includes:

[0166] Obtain the first coefficient of recovery;

[0167] The target safety threshold is determined based on the first recovery coefficient and the current threshold.

[0168] The rate of change of current is obtained by processing the first derivative of the system current parameters;

[0169] When the absolute value of the current change rate is less than or equal to the preset current change rate, and the system current parameter drops to the target safety threshold, the second target hard drive is triggered to perform boot control processing.

[0170] Specifically, the first recovery coefficient is obtained, the target safety threshold is determined based on the first recovery coefficient and the current threshold, the first derivative of the system current parameter is processed to obtain the current change rate, and two preset conditions are used as stability conditions. That is, when the absolute value of the current change rate is less than or equal to the preset current change rate, and the system current parameter drops to the target safety threshold, the second target hard disk is triggered to perform boot control processing.

[0171] The formula is as follows:

[0172] ; (4)

[0173] ; (5)

[0174] Where A is the first coefficient of recovery, typically ranging from 0.80 to 0.90; For the target safety threshold, The rate of change of current, The preset current change rate ensures that the current has stabilized rather than experiencing a brief drop.

[0175] In the entire process of the triggering mechanism for restoring the second target hard drive for boot control processing provided in this embodiment, the system current must drop below a safe threshold, i.e., after the stability condition is met, before subsequent booting or operation can be resumed to ensure system stability.

[0176] In some embodiments, the recovery process of the second target hard drive includes:

[0177] Obtain the peak startup current value of the second target hard drive; sort the peak startup current values ​​from smallest to largest to obtain the sorted second target hard drive; and perform recovery boot according to the sorted second target hard drive.

[0178] Alternatively, the second target hard disk is grouped to obtain the grouped second target hard disk; the grouped second target hard disk is then restored and booted according to the balancing strategy; wherein, the balancing strategy is a strategy in which the sum of the startup current of any group of the grouped second target hard disks is less than or equal to the system rated current value.

[0179] Alternatively, obtain the slot number of the second target hard drive; perform phase allocation processing on the same hard drives in the same batch corresponding to the second target hard drive based on the slot number, in order to restore the second target hard drive.

[0180] Specifically, the optimized sequence refers to the dynamically adjusted hard drive boot order, with the goal of minimizing the total boot time within current limits. This sequence can be determined in the following ways:

[0181] 1) Sorting based on power consumption: Ranking hard drives by peak startup current. Sort by power consumption from low to high, starting with the lower-power hard drive and then starting the higher-power hard drive to avoid accumulating high current.

[0182] 2) Sequence optimization based on Nash equilibrium (equilibrium strategy): Group the hard drives and determine the optimal boot sequence through iterative calculation, ensuring that the sum of the boot current of any group of hard drives does not exceed the maximum allowable current of the system (system rated current). ).

[0183] 3) Combination of slot number and phase dispersion: Based on the phase allocation preset by the phase compensation controller, hard drive combinations with larger phase misalignment are started first to avoid current peak superposition from a physical level.

[0184] Regarding points 2) and 3) above, they will be mentioned in the following description, and you can refer to the following embodiments.

[0185] The optimized hard drive order provided in this embodiment is achieved through three different strategies to ensure that the total boot time is minimized within current limits, thereby improving boot efficiency. The first strategy typically allows more low-power hard drives to boot simultaneously, making fuller use of the power supply's remaining capacity compared to a strategy that prioritizes larger drives. The second strategy uses algorithmic iteration to find the theoretically fastest boot sequence. It solves the problem of how to mix large and small hard drives, ensuring that the power supply operates at full load every millisecond, thus compressing the total boot time to its limit. The third strategy significantly reduces high-frequency noise and stress impact from the input current, resulting in smoother power module operation, less heat generation, and extended hardware lifespan.

[0186] In some embodiments, the process of generating a hard disk dynamic performance tuning strategy includes:

[0187] Verify that the hard drive is booting.

[0188] Get the critical queue depth for each of the booted hard drives;

[0189] The processed queue depth is obtained by limiting the critical queue depth for each of the already started hard drives.

[0190] The startup hard drive is slowed down based on the processed queue depth to limit the number of commands issued.

[0191] Perform boot control processing on the already booted hard drive after speed reduction.

[0192] Specifically, by limiting the Native Command Queuing (NCQ) depth, the SATA controller reduces concurrent read and write operations on the hard drive by limiting the number of commands issued simultaneously, thereby reducing instantaneous power consumption.

[0193] Let the original maximum queue depth (critical queue depth) be... The reduced queue depth (processed queue depth) is ,but: ;

[0194] in, This is the speed reduction factor, typically ranging from 0.3 to 0.7, which is dynamically adjusted according to the degree of current overload.

[0195] The queue depth limitation method provided in this embodiment directly smooths out the current spikes caused by high-frequency random read / write by limiting the number of commands issued simultaneously. After limiting the queue depth, the hard drive's processing logic becomes closer to sequential read / write, and the movement path of the read / write head arm is shorter and smoother. This reduces the number of times the read / write head accelerates, decelerates, and stops abruptly, directly reducing the power consumption of the voice coil motor, improving data transmission efficiency, and enhancing system stability.

[0196] In some embodiments, the process of generating a hard disk dynamic performance tuning strategy includes:

[0197] Get the time interval between read and write commands for the already started hard drive;

[0198] Idle time is inserted between read and write command intervals to reduce the duty cycle of the already started hard drive;

[0199] Perform boot control processing on the downgraded, already booted hard drive.

[0200] Specifically, idle cycles are inserted by proactively inserting idle time between consecutive read and write commands to reduce the hard drive's duty cycle. Let the original command interval be... Insert idle time Afterwards, the effective data throughput decreased to: ;in, The value is dynamically adjusted according to the degree of current overload, with a typical range of values ​​being: 0.5 to 2 times that.

[0201] Inserting idle cycles is essentially a pulse width modulation strategy on a time slice. Unlike limiting command queue depth (NCQ limiting), which focuses on reducing concurrency, inserting idle cycles focuses on controlling the density of continuous work.

[0202] Combining the above-mentioned speed reduction measures to decrease the data transmission rate, the degree of speed reduction can be determined by the severity of the current overload: if At the warning level ( If so, the speed will decrease slightly. Take 0.7, Pick ;like In emergency level ( If the speed decreases significantly, then the speed will decrease further. Take 0.3, Pick When the current returns to normal ( After that, it gradually returns to the original rate.

[0203] The hard drive dynamic performance adjustment strategy provided in this embodiment, which reduces the hard drive duty cycle by inserting idle periods, avoids the system repeatedly jumping between current-limited and unlimited states. These forced idle periods become miniature heat dissipation windows. Although the time is short, it is enough for heat to be conducted from the core components to the casing or heatsink. In high-temperature environments, this strategy can effectively interrupt the continuous rise in temperature and prevent the hard drive from triggering more severe thermal protection throttling due to overheating. For mechanical hard drives, reducing the continuous power-on time of the motor and read / write heads helps to delay mechanical wear in the long run.

[0204] In some embodiments, when performing boot control processing to adjust the boot order and / or operating status of the current batch of hard drives according to hard drive start / stop policies and / or hard drive dynamic performance adjustment policies, the method further includes:

[0205] Get the number of hard drives in the current batch;

[0206] Determine the hard drive index and the number of power supply modules corresponding to each hard drive based on the number of hard drives.

[0207] Phase dispersion compensation is performed based on the hard disk index and the number of power supply modules to obtain the current phase corresponding to each power supply module.

[0208] Power supply is applied to the hard drives in the current batch according to each current phase.

[0209] Specifically, the hard drive boot control process in the above embodiment is performed on all hard drives in the current batch. However, the number of hard drives in the current batch can be one or more. For multiple hard drives, the phase dispersion technology significantly reduces the peak boot current of multiple hard drives, and the supercapacitor energy storage module can respond to power surges within 20ms.

[0210] The power supply module here can be implemented using supercapacitor banks. Supercapacitor energy storage modules can quickly store and release electrical energy. During the charging phase, its voltage change follows the formula:

[0211] ; (6)

[0212] in, For maximum voltage, The initial voltage, For equivalent series resistance, This represents the capacitance.

[0213] During the discharge phase, the rapid release of electrical energy is achieved through the following formula:

[0214] ; (7)

[0215] The dynamic power allocation algorithm execution unit utilizes phase dispersion compensation technology, and the phase distribution function is:

[0216] ; (8)

[0217] in, The number of power supply modules, For module indexing, The maximum offset. For offset frequency, This is a dynamic phase shift.

[0218] Adjusting the output current phase of the power module avoids current superposition surges, thereby stably distributing power to the SATA controller, NVMe controller, and HDD hard drive, ensuring that all system components receive stable power support under different operating conditions.

[0219] By adjusting the phase of the output current from each power module, the current peaks are staggered in time to avoid superposition. For example, with four HDDs, the startup power phases would be:

[0220] HDD power phase[1]: ;

[0221] HDD power phase[2]: ;

[0222] HDD power phase[3]: ;

[0223] HDD power phase[4]: .

[0224] The collaborative mechanism provided in this embodiment ensures that when the system is starting up with multiple hard drives, when the current approaches the threshold, the startup of some hard drives is dynamically delayed. Power is redistributed through phase dispersion technology to avoid total current overload, and the instantaneous power gap is compensated by the supercapacitor energy storage module.

[0225] Furthermore, this application also provides a bus adapter, which includes a processor 1, a first protocol controller 2, a second protocol controller 3, a current acquisition module 4, a power supply module 5, and a phase compensation controller 6;

[0226] Processor 1 is connected to first protocol controller 2, second protocol controller 3, current acquisition module 4 and power supply module 5;

[0227] The power supply module 5 and the phase compensation controller 6 are connected;

[0228] Both the second protocol controller 3 and the phase compensation controller 6 are connected to the hard drive.

[0229] Processor 1 is used to execute the steps of the hard disk boot method described above.

[0230] Figure 2 This is a schematic diagram of a traditional SATA HDD connection method provided in an embodiment of this application, as shown below. Figure 2 As shown, the SATA interface and the Advanced Host Controller Interface (AHCI) protocol are designed for mechanical hard drives. The host's SATA AHCH controller (Serial Advanced Technology Accessory Advanced Host Controller) connects to the hard drive via the SATA interface, supporting only a single queue and a queue depth of 32. However, newer storage servers no longer use SATA interfaces, opting instead for PCIe interfaces. Nevertheless, HDDs, due to their large capacity, mature technology, and low unit storage cost, are still widely used in data centers, video surveillance, historical data archiving, and other high-capacity cold storage scenarios. Therefore, solving the connectivity problem of HDDs in next-generation servers is crucial. Figure 3 This is a schematic diagram of the structure of a bus adapter provided in an embodiment of this application, as shown below. Figure 3 As shown, it can be roughly divided into the following three parts: dynamic current management engine, controller and quantum power distribution system.

[0231] The current acquisition module collects and monitors the overall current consumption of the Host Bus Adapter (HBA) and all downstream SATA HDDs in real time. It records the current through a status register, and a warning register triggers an interrupt when the current reaches a threshold, providing current monitoring data to the processor. The processor is responsible for HBA initialization and progressive hard drive startup. After startup, it handles hard drive array management, I / O request processing and conversion, and data interaction with the host. The cache provides the processor with a memory context and temporarily stores I / O requests, while the SATA controller executes standard SATA interface protocol transactions to perform read and write operations on the HDDs.

[0232] The HBA adapter acts as a bridge connecting the host and HDDs. Upstream, it connects to the host system via the PCIe bus, fully supporting the NVMe standard protocol and simulating a standard NVMe storage device on the host side. Downstream, it connects to multiple SATA HDD storage devices via the SATA bus, enabling protocol conversion. The NVMe controller conforms to the NVMe specification, receiving host I / O requests through a submission queue, temporarily storing them in a cache before handing them over to system software for processing. After the operation is completed, it sends the result back to the host through a completion queue.

[0233] The dynamic current management engine is the core component for achieving precise current control in this application. It establishes a three-stage current mathematical model to achieve millisecond-level load prediction. When the dynamic current management engine detects that the total system current is close to or exceeds the warning threshold, it immediately activates an intelligent control strategy. By interacting with the processor, it adjusts the HDD boot sequence and operating status to prevent further current increases. The quantum power distribution system plays a crucial role in stabilizing the power supply throughout the architecture.

[0234] For a description of the bus adapter provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the above hard disk boot method.

[0235] In some embodiments, the processor includes a first processor and a second processor;

[0236] The first processor is connected to the first protocol controller 2, the current acquisition module 4, and the second processor;

[0237] The second processor is connected to the second protocol controller 3 and the power supply module 5.

[0238] Specifically, such as Figure 3 As shown, there are two processors here. This is mainly because the processors involve a lot of computing and storage resources during the entire hard drive boot process. The boot efficiency is improved by having two co-processors interact.

[0239] The first processor is characterized by low power consumption and high real-time performance, such as a Cortex M processor; the second processor is characterized by high performance and multi-tasking concurrency, such as a multi-core Cortex A processor. The first processor is connected to the first protocol controller, the current acquisition module, and the second processor; the second processor is connected to the second protocol controller and the power supply module.

[0240] After the adapter is powered on, the first processor runs first, executes the minimum loading program, initializes the HBA, and sets the PCIe interface to the minimum operating state to reduce power consumption and reserve current margin. Then, a progressive strategy is adopted to start the downstream HDD hard drives in sequence, while listening for interrupts from the current acquisition module and adjusting the startup process according to the current change trend. After the hard drive startup is completed, the second processor is woken up.

[0241] The second processor is typically a multi-core Cortex A high-performance embedded processor running a multi-tasking RTOS, serving as the central hub for HBA management and I / O processing. It is responsible for establishing the mapping between the NVMe namespace and SATA HDD devices, decomposing, scheduling, and converting NVMe I / O requests, and feeding back the results to the host after completing data read / write operations.

[0242] The heterogeneous dual-core collaborative processing provided in this embodiment offloads lightweight, resident tasks (such as standby monitoring and simple control) to the first processor (Cortex-M), enabling it to operate in an extremely low-power state; the second processor is only woken up when complex business needs to be processed (hard disk boot process), which ensures the absolute timeliness of critical tasks while retaining the rich functionality of the system.

[0243] In normal circumstances, to improve hard drive boot efficiency, multiple processors are used for parallel processing. However, during parallel processing, the total current of the two processors may exceed a predetermined current value, thus exacerbating the superposition of current peaks. For safety, two processors are used for serial processing, but this reduces boot efficiency. Furthermore, the current of the two processors does not necessarily operate at peak levels throughout the process. In some embodiments, the first and second processors boot according to a first preset boot sequence during hard drive boot; wherein the process of determining the first preset boot sequence includes:

[0244] Obtain the first current peak, the first preset startup time percentage, and the first inherent startup time corresponding to the first processor, and the second current peak, the second preset startup time percentage, and the second inherent startup time corresponding to the second processor; wherein the sum of the first preset startup time percentage and the second preset startup time percentage is equal to 1;

[0245] The first startup parameter is determined based on the first peak current and the first inherent startup time; the second startup parameter is determined based on the penalty term coefficient, the first startup parameter, and the critical allowable current; the first startup efficiency is determined based on the first preset startup time percentage and the preset startup time of the first processor; and the first actual startup time percentage corresponding to the first processor is determined based on the first startup efficiency and the second startup parameter.

[0246] The third startup parameter is determined based on the second current peak value and the second inherent startup time; the fourth startup parameter is determined based on the third startup parameter, the penalty term coefficient, the first startup parameter, and the critical allowable current; the second startup efficiency is determined based on the second preset startup time ratio and the preset startup time of the second processor; the first startup efficiency and the second startup efficiency are summed to obtain the third startup efficiency; the second actual startup time ratio corresponding to the second processor is determined based on the third startup efficiency and the fourth startup parameter.

[0247] The first preset startup sequence of the first processor and the second processor is determined based on the proportion of the first actual startup time and the proportion of the second actual startup time.

[0248] Specifically, the processing method corresponding to the balancing strategy involving the two processors follows the formula as follows:

[0249] ; (9)

[0250] in, It indicates a heterogeneous dual-core processor (first processor and second processor). With core The percentage of preset startup time ( ),and , With core The inherent startup time, With core Peak current at startup These are Lagrange multipliers (penalty term coefficients used to balance time and current constraints). This is the startup efficiency (the reciprocal of the total startup time). To ensure that the current limit is exceeded, the penalty is imposed. .

[0251] When the startup order of the two cores reaches equilibrium, changing the startup time ratio of either core alone cannot further improve efficiency without violating current limits. At this point, the system reaches its fastest startup speed within the safe current range.

[0252] It should be noted that since the first processor is a microprocessor used for issuing instructions and is prioritized for startup, the output for the first and second processors in this embodiment is the actual startup time ratio. The corresponding preset startup sequence is determined based on the specific actual startup time ratio.

[0253] The heterogeneous dual-core collaborative controller works closely with the dynamic current management engine and the quantum power distribution system. It interacts with the dynamic current management engine to acquire real-time current data and current change trends, serving as the basis for startup timing optimization and task scheduling. Simultaneously, it feeds back startup strategies and task execution status to the dynamic current management engine, assisting it in adjusting current warning thresholds and control strategies. Collaborating with the quantum power distribution system, it sends power distribution commands based on system task requirements and current conditions, adjusting the supercapacitor charging and discharging strategy and power output phase to ensure that each component receives a stable and appropriate power supply, jointly guaranteeing the stable and efficient operation of the HBA system.

[0254] The startup timing optimization algorithm for the two processors provided in this embodiment uses Nash equilibrium to determine the optimal startup timing of the two cores (the first processor and the second processor) to avoid the superposition of current peaks during startup and maximize system startup efficiency. Iterative calculations determine the processor startup timing, avoiding the superposition of current peaks and improving collaborative efficiency. When the HBA adapter starts, the heterogeneous dual cores (the first processor and the second processor) need to start in a certain timing sequence, while simultaneously controlling the startup order of the downstream HDD hard drives. The goal is to minimize the maximum allowable current in the system. Under constraints, minimize total startup time and avoid current overload.

[0255] In some embodiments, the second processor includes at least one sub-processor; the number of sub-processors is the same as the number of hard disk batches; each sub-processor is connected to the first processor and the power supply module; each sub-processor corresponds to a hard disk connection; each sub-processor starts up according to a second preset boot sequence during the hard disk boot process.

[0256] Specifically, considering the existence of multiple hard drive batches in the hard drive array, a multi-subprocessor setup is adopted to improve boot efficiency. That is, the second processor includes at least one subprocessor, each corresponding to a specific batch, and each subprocessor is connected to the first processor and the power supply module; each subprocessor corresponds to one hard drive connection. Similarly, each subprocessor boots according to a second preset boot sequence during the hard drive boot process. This second preset boot sequence is obtained by referring to the first preset boot sequence in the above embodiment, i.e., it is also implemented using the Nash equilibrium solution algorithm.

[0257] The startup timing determination of multiple sub-processors provided in this embodiment further avoids the superposition of current peaks when the hard drive starts up in different batches, while also improving system startup efficiency.

[0258] In some embodiments, the entire operation process is based on an innovative design of a heterogeneous dual-processor collaborative architecture and a quantum power distribution system. The specific process and description are as follows:

[0259] Step 1: System power-on and power initialization:

[0260] After the HBA is powered on, the quantum power distribution system is initialized first. The supercapacitor energy storage module begins pre-charging, with the voltage set according to... The exponential law rises to the initial value This stores energy to meet the instantaneous high current demands during the subsequent HDD startup phase. At this time, the dynamic power allocation algorithm execution unit initializes the phase distribution parameters and sets the basic phase offset. .

[0261] Step 2: First processor initialization and PCIe configuration:

[0262] The low-power primary processor (such as Cortex M) boots up and performs HBA minimum system initialization, configuring the PCIe interface to a single-channel, 2.5GT / s minimum operating state to minimize NVMe controller power consumption and reserve more current margin for HDD booting. Simultaneously, a system current monitoring baseline is established. This serves as a reference for subsequent dynamic threshold adjustments.

[0263] Step 3: Progressive HDD boot control:

[0264] The first processor boots the hard disk array in a preset order (e.g., HDD IDs from smallest to largest). During the boot process of each HDD, the current acquisition module collects the total system current in real time. .when Exceeding the current threshold At this time, the system triggers a dynamic response mechanism: suspending subsequent HDD startup; activating the dynamic power allocation algorithm to calculate the optimal phase offset for each component. Adjust the supercapacitor discharge current ,pass The formula controls the discharge rate; power is dynamically allocated based on the Nash equalization algorithm to ensure that the total current is maintained within a safe range.

[0265] Step 4: Workflow of the quantized power distribution system:

[0266] The quantum power distribution system operates continuously throughout the entire HDD boot and system operation process:

[0267] 1) Power demand awareness: Real-time monitoring of the instantaneous power demand of NVMe controller, SATA controller and each HDD;

[0268] 2) Dynamic phase adjustment: The phase of the output current of each power module is dynamically adjusted based on load changes.

[0269] 3) Supercapacitor synergy: Release stored energy during peak current periods and charge during off-peak periods to smooth system power fluctuations;

[0270] 4) Algorithm Iteration Optimization: The phase distribution parameters are updated every 20ms to ensure that the system always runs in the optimal state.

[0271] Figure 4 A flowchart of another hard disk boot method provided in the embodiments of this application is shown below. Figure 4 As shown, the method includes:

[0272] S21: Adapter powered on;

[0273] S22: Initialize the quantum power distribution system;

[0274] S23: The supercapacitor energy storage module is precharged to its initial value;

[0275] S24: The first processor starts up and initializes the adapter;

[0276] S25: Set the minimum operating state of the high-speed peripheral component interconnection interface;

[0277] S26: Establish current monitoring reference value;

[0278] S27: Start the hard disk array in sequence;

[0279] S28: The current acquisition module monitors the system current parameters in real time;

[0280] S29: Determine whether the system current parameter is greater than the current threshold; if yes, proceed to step S30; if no, proceed to step S31.

[0281] S30: Pause startup and activate dynamic power allocation algorithm;

[0282] S32: Adjust the supercapacitor discharge current;

[0283] S33: Calculate the phase offset of each component;

[0284] S34: Dynamically allocate power to hard drive boot;

[0285] S31: Continue booting the next batch of hard drives;

[0286] S35: After all hard drives have finished booting, wake up the second processor;

[0287] S36: Second processor initializes the mapping table;

[0288] S37: Establish a logical mapping between the namespace and the hard disk;

[0289] S38: Receive input / output requests from the host;

[0290] S39: Quantum power distribution system dynamically adjusts power allocation;

[0291] S40: Perform input / output operations;

[0292] S41: The result is fed back to the host via the controller;

[0293] S42: Continuously monitor system power consumption and optimize allocation.

[0294] Furthermore, this application also provides a server, including a host 8, a hard disk 7, and the aforementioned bus adapter;

[0295] The host 8, bus adapter, and hard disk 7 are connected in sequence.

[0296] For a description of the server provided in this application, please refer to the above method embodiments. This application will not repeat the details here, but it has the same beneficial effects as the bus adapter described above.

[0297] The foregoing has described in detail various embodiments corresponding to the hard disk boot method. Based on this, this application also discloses a hard disk boot device corresponding to the above method. Figure 5 This is a structural diagram of a hard disk boot device provided in an embodiment of this application. Figure 5 As shown, the hard disk boot device includes:

[0298] The acquisition module 11 is used to acquire the system current parameters corresponding to the current time of the hard disk at the current operating stage and the average current parameters corresponding to the preset time before the current time as the reference.

[0299] Module 12 is used to determine the current threshold corresponding to the current moment based on the system current parameters and the average current parameters;

[0300] The adjustment module 13 is used to perform startup control processing on the hard drive to adjust the startup order and / or working state according to the hard drive start-stop strategy and / or hard drive dynamic performance adjustment strategy when the system current parameter is within the preset range corresponding to the current threshold, so as to complete the hard drive startup process; wherein, the hard drive dynamic performance adjustment strategy is a strategy to adjust the hard drive data transfer rate.

[0301] Since the embodiments of the device part correspond to the embodiments described above, please refer to the embodiments described in the method part for the embodiments of the device part, and will not be repeated here.

[0302] For a description of the hard disk boot device provided in this application, please refer to the above method embodiments. This application will not repeat the description here, as it has the same beneficial effects as the above hard disk boot method.

[0303] Figure 6 A structural diagram of an electronic device provided in an embodiment of this application, such as... Figure 6 As shown, the device includes:

[0304] Memory 21 is used to store computer programs;

[0305] The third processor 22 is used to implement the steps of the hard disk boot method when executing computer programs.

[0306] The third processor 22 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The third processor 22 may be implemented using at least one hardware form selected from Digital Signal Processors (DSPs), Field-Programmable Gate Arrays (FPGAs), and programmable logic arrays. The third processor 22 may also include a main processor and a coprocessor. The main processor, also known as a central processing unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the third processor 22 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the third processor 22 may also include an Artificial Intelligence (AI) processor, which handles computational operations related to machine learning.

[0307] The memory 21 may include one or more computer-readable storage media, which may be non-transitory. The memory 21 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 21 is used to store at least the following computer program 211, which, after being loaded and executed by the third processor 22, is capable of implementing the relevant steps of the hard disk boot method disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 21 may also include an operating system 212 and data 213, etc., and the storage method may be temporary storage or permanent storage. The operating system 212 may include Windows, Unix, Linux, etc. The data 213 may include, but is not limited to, the data involved in the hard disk boot method, etc.

[0308] In some embodiments, the electronic device may further include a display screen 23, an input / output interface 24, a communication interface 25, a power supply 26, and a communication bus 27.

[0309] Those skilled in the field can understand, Figure 6 The structures shown do not constitute a limitation on electronic devices and may include more or fewer components than those shown.

[0310] The third processor 22 implements the hard disk boot method provided in any of the above embodiments by calling instructions stored in the memory 21.

[0311] For an introduction to the electronic device provided in this application, please refer to the above method embodiments. This application will not repeat the details here, but it has the same beneficial effects as the above hard disk boot method.

[0312] Furthermore, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a third processor 22, implements the steps of the hard disk boot method described above.

[0313] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to conventional technology, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0314] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the above hard disk boot method.

[0315] Furthermore, this application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of a hard disk boot method.

[0316] For an introduction to the computer program product provided in this application, please refer to the above method embodiments. This application will not repeat the details here, but it has the same beneficial effects as the above hard disk boot method.

[0317] The above provides a detailed description of a hard disk boot method, bus adapter, server, device, medium, and product provided in this application. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

[0318] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

Claims

1. A hard disk boot method, characterized in that, include: Obtain the system current parameters corresponding to the current moment of the hard drive's current operating phase and the average current parameters corresponding to the preset moments prior to the current moment; Determine the current threshold corresponding to the current moment based on the system current parameters and the average current parameters; When the system current parameter is within a preset range corresponding to the current threshold, the hard drive is subjected to startup control processing to adjust its startup order and / or working state according to the hard drive start-stop strategy and / or hard drive dynamic performance adjustment strategy to complete the hard drive startup process; specifically, this includes: if the system current parameter exceeds the current threshold and is within a first preset range, then the hard drive is subjected to startup control processing to adjust its startup order according to the hard drive start-stop strategy; if the system current parameter exceeds the current threshold and is within a second preset range, then the hard drive is subjected to startup control processing to adjust its startup order and working state according to the hard drive start-stop strategy and the hard drive dynamic performance adjustment strategy; if the system current parameter does not exceed the current threshold and is within a third preset range, then the hard drive is subjected to startup control processing to adjust its working state according to the hard drive dynamic performance adjustment strategy; wherein, the second preset range is greater than the first preset range; the third preset range is less than the first preset range; and the hard drive dynamic performance adjustment strategy is a strategy for adjusting the hard drive's data transfer rate.

2. The hard disk boot method according to claim 1, characterized in that, Determining the current threshold at the current moment based on the system current parameters and the average current parameters includes: The current change gradient parameters are obtained by performing second-order derivative processing on the system current parameters; The safety margin parameters are determined based on the current change gradient parameters, adjustment coefficients, and redundancy factors. The current threshold corresponding to the current moment is determined based on the average current parameter and the safety margin parameter.

3. The hard disk boot method according to claim 1, characterized in that, When the system current parameter is within a preset range corresponding to the current threshold, the system performs startup control processing to adjust the boot order and / or working state of the hard drive according to the hard drive start / stop strategy and / or hard drive dynamic performance adjustment strategy, including: To determine the degree of closeness between the system current parameters and the current threshold; Based on the relationship between the proximity degree and the preset proximity degree, the parameters corresponding to the hard disk start-stop strategy and / or hard disk dynamic performance adjustment strategy are adjusted to obtain the adjusted hard disk start strategy and / or adjusted hard disk dynamic performance adjustment strategy. The hard drive is subjected to boot control processing that adjusts the boot order and / or working state according to the adjusted hard drive boot strategy and / or the adjusted hard drive dynamic performance adjustment strategy.

4. The hard disk boot method according to claim 1 or 3, characterized in that, The process of generating the hard drive start / stop policy includes: The first target hard disk is identified in the hard disk; Obtain the second target hard drive from the first target hard drive, and suspend the boot process for the second target hard drive; If the system current parameter drops to the target safety threshold, the second target hard disk is restored for boot control processing.

5. The hard disk boot method according to claim 1 or 3, characterized in that, The process of generating the hard disk dynamic performance tuning strategy includes: Verify that the hard drive is booted. Obtain the critical queue depth corresponding to each of the started hard drives; The processed queue depth is obtained by limiting the critical queue depth for each of the already started hard drives. The activated hard disk is slowed down based on the processed queue depth to limit the number of commands issued. Perform boot control processing on the already booted hard drive after speed reduction.

6. The hard disk boot method according to claim 1 or 3, characterized in that, The process of generating the hard disk dynamic performance tuning strategy includes: Get the time interval between read and write commands for the already started hard drive; Idle time is inserted into the read / write command time interval to reduce the duty cycle of the started hard disk; The boot control process is then performed on the booted hard drive after the downgrade.

7. The hard disk boot method according to claim 1, characterized in that, When performing boot control processing to adjust the boot order and / or working status of the current batch of hard drives according to the hard drive start / stop policy and / or the hard drive dynamic performance adjustment policy, the method further includes: Get the number of hard drives in the current batch; The number of hard drives and the number of power supply modules corresponding to each hard drive are determined based on the number of hard drives. Phase dispersion compensation is performed based on the hard disk index and the number of power supply modules to obtain the current phase corresponding to each power supply module. Power supply is applied to the hard drives in the current batch according to the current phase described.

8. The hard disk boot method according to claim 4, characterized in that, When the first target hard drive is a non-bootable hard drive, the process of acquiring the second target hard drive includes: Obtain the slot number, service importance, and power consumption level of the unbooted hard drive; Assign a first weighting coefficient, a second weighting coefficient, and a third weighting coefficient according to the slot number, the importance of the service, and the power consumption level; The first hard disk parameters corresponding to each of the unstarted hard disks are determined based on the slot number, the first weight coefficient, the importance of the service, the second weight coefficient, the power consumption level, and the third weight coefficient. Filter the target first hard disk parameters that are less than the preset hard disk parameters from each of the first hard disk parameters; The unbooted hard drive corresponding to the first target hard drive parameters is used as the second target hard drive.

9. The hard disk boot method according to claim 4, characterized in that, When all hard drives are booted up, and the first target hard drive is a booted hard drive, the process of acquiring the second target hard drive includes: Get the queue depth, data read / write tasks, and activity status of the started hard drives; Within the already started hard drives, a second target hard drive is selected based on the following: the queue depth has not reached the preset queue depth, the data read / write task is a non-real-time data read / write task, and the active state is an idle state.

10. The hard disk boot method according to claim 4, characterized in that, The trigger mechanism for restoring the boot control process of the second target hard drive includes: Obtain the first coefficient of recovery; The target safety threshold is determined based on the first recovery coefficient and the current threshold. The rate of change of current is obtained by processing the first derivative of the system current parameters. When the absolute value of the current change rate is less than or equal to the preset current change rate, and the system current parameter drops to the target safety threshold, the second target hard disk is triggered to perform boot control processing.

11. The hard disk boot method according to claim 10, characterized in that, The recovery process for the second target hard drive includes: Obtain the peak startup current value of the second target hard drive; sort the peak startup current values ​​from smallest to largest to obtain the sorted second target hard drives; and perform recovery boot according to the sorted second target hard drives; Alternatively, the second target hard disk is grouped to obtain grouped second target hard disks; the grouped second target hard disks are then restarted according to a balancing strategy; wherein, the balancing strategy is a strategy in which the sum of the startup currents of any group of the grouped second target hard disks is less than or equal to the system rated current value. Alternatively, obtain the slot number of the second target hard drive; perform phase allocation processing based on the slot number and the equivalent hard drives in the same batch corresponding to the second target hard drive to restore the second target hard drive.

12. A bus adapter, characterized in that, The bus adapter includes a processor, a first protocol controller, a second protocol controller, a current acquisition module, a power supply module, and a phase compensation controller. The processor is connected to a first protocol controller, a second protocol controller, a current acquisition module, and a power supply module. The power supply module and the phase compensation controller are connected; Both the second protocol controller and the phase compensation controller are connected to a hard disk; The processor is configured to perform the steps of the hard disk boot method according to any one of claims 1 to 11.

13. The bus adapter according to claim 12, characterized in that, The processor includes a first processor and a second processor; The first processor is connected to the first protocol controller, the current acquisition module, and the second processor; The second processor is connected to the second protocol controller and the power supply module.

14. The bus adapter according to claim 13, characterized in that, The first processor and the second processor boot according to a first preset boot sequence during the hard disk boot process; wherein, the process of determining the first preset boot sequence includes: Obtain the first current peak, the first preset startup time percentage, and the first inherent startup time corresponding to the first processor, and the second current peak, the second preset startup time percentage, and the second inherent startup time corresponding to the second processor; wherein, the sum of the first preset startup time percentage and the second preset startup time percentage is equal to 1; A first startup parameter is determined based on the first current peak value and the first inherent startup time; a second startup parameter is determined based on the penalty term coefficient, the first startup parameter, and the critical allowable current; a first startup efficiency is determined based on the first preset startup time percentage and the preset startup time of the first processor; and a first actual startup time percentage corresponding to the first processor is determined based on the first startup efficiency and the second startup parameter. A third startup parameter is determined based on the second current peak value and the second inherent startup time; a fourth startup parameter is determined based on the third startup parameter, the penalty term coefficient, the first startup parameter, and the critical allowable current; a second startup efficiency is determined based on the second preset startup time percentage and the preset startup time of the second processor; the first startup efficiency and the second startup efficiency are summed to obtain a third startup efficiency; and a second actual startup time percentage corresponding to the second processor is determined based on the third startup efficiency and the fourth startup parameter. The first preset startup sequence of the first processor and the second processor is determined based on the first actual startup time percentage and the second actual startup time percentage.

15. The bus adapter according to claim 13, characterized in that, The second processor includes at least one sub-processor; the number of sub-processors is the same as the number of hard disk batches; each sub-processor is connected to the first processor and the power supply module; Each subprocessor corresponds to a hard disk connection; each subprocessor starts up according to the second preset boot sequence during the hard disk boot process.

16. A server, characterized in that, Includes a host, a hard disk, and the bus adapter as described in any one of claims 12 to 14 above; The host, the bus adapter, and the hard disk are connected in sequence.

17. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the hard disk boot method as described in any one of claims 1 to 11 when executing the computer program.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the hard disk boot method as described in any one of claims 1 to 11.

19. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the hard disk boot method according to any one of claims 1 to 11.

Images