A device management system, server and method

Abstract

The application discloses a device management system, a server and a method, relates to the technical field of hot plug, and devices can be installed in the direction perpendicular to the server panel through the opening of the server panel. Through the cooperative control of the first power supply module and the second power supply module, the orderly connection and disconnection of power supply and communication are realized when the devices are inserted or pulled out, so that the safe hot plug is realized, the high-power and high-stability power supply after the device is inserted is realized by the two power supply modules, and the application demand of high-performance devices is adapted. The problems that the traditional PCIe form devices face in high-density deployment, such as physical space limitation, complex plug-in and plug-out operation, whole machine power-off maintenance and high link reconnection risk and the like are solved, and the beneficial effects of improving operation and maintenance efficiency, avoiding business interruption, reducing hardware damage probability and guaranteeing data security are achieved.

Description

Technical Field

[0001] This application relates to the field of hot-swappable technology, and more particularly to a device management system, server, and method. Background Technology

[0002] With the rapid development of cloud computing, big data, and artificial intelligence technologies, data centers are facing ever-increasing demands for computing and network processing. Traditional CPU (Central Processing Unit) architectures are inefficient at handling network packets, storage requests, and security functions, leading to performance bottlenecks. Data Processing Units (DPUs) have emerged as dedicated processors specifically designed to offload and accelerate network, storage, and security workloads. However, the current mainstream PCIe (Peripheral Component Interconnect Express) form factor of DPUs, under the space constraints of high-density computing environments in data centers, makes replacement, installation, and deployment difficult, significantly increasing maintenance time and costs. Furthermore, the characteristics of the PCIe form factor necessitate disconnecting and reconnecting all interconnect links when replacing a DPU; this process is not only time-consuming but may also introduce risks such as poor physical connections or misconfigurations.

[0003] Therefore, how to provide a solution to the above-mentioned technical problems is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] This application provides a device management system, server, and method to at least solve the problems faced by PCIe form factor devices in high-density deployments, such as physical space limitations, complex plugging and unplugging operations, the need for complete machine power-off maintenance, and high risk of link reconnection.

[0005] This application provides a device management system. The device includes a power supply interface and a high-speed interface. The server includes a housing, a guide portion, and a backplate. The housing has a housing opening corresponding to the connector position on the backplate. The guide portion is located between the housing opening and the connector. The guiding direction of the guide portion is perpendicular to the plane of the housing opening. The guide portion is configured to guide the device moving along the housing opening to insert and remove it from the server parallel to the backplate. The device management system includes a first power supply module and a second power supply module. The output power of the first power supply module is greater than the output power of the second power supply module. The power supply priority of the first power supply module is higher than that of the second power supply module. The power-off priority of the first power supply module is lower than that of the second power supply module. Electrical priority; when the device engages with the connector, the power supply interface of the device is electrically connected to the first power supply module, and the high-speed interface of the device is electrically connected to the second power supply module; the management module is configured to, when the device is detected to be inserted into the connector, control the first power supply module and the second power supply module to start supplying power to the device in descending order of power supply priority, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, and control the first power supply module and the second power supply module to disconnect the power supply to the device to be removed in descending order of power-off priority; the main control module is configured to establish a communication connection with the device in response to the device insertion signal, and disconnect the communication connection with the device to be removed in response to the device removal signal.

[0006] This application also provides a server, including the device management system described above, and further including a server housing, a guide portion, and a backplate. The housing is provided with a housing opening corresponding to the connector position of the backplate. The guide portion is located between the housing opening and the connector. The guiding direction of the guide portion is perpendicular to the plane where the housing opening is located. The guide portion is configured to guide a device moving along the housing opening to insert into and remove from the server parallel to the backplate.

[0007] This application also provides a device management method applied to a management module in a device management system as described above. The device management method includes: when a device is detected inserted into a connector on a backplane, controlling a first power supply module and a second power supply module in the device management system to start supplying power to the device in descending order of power supply priority, and outputting a device insertion signal so that the host module in the device management system responds to the device insertion signal and establishes a communication connection with the device; wherein, the output power of the first power supply module is greater than the output power of the second power supply module, the power supply priority of the first power supply module is higher than the power supply priority of the second power supply module, and the power-off priority of the first power supply module is lower than the power-off priority of the second power supply module; when the device engages with the connector, the power supply interface of the device is electrically connected to the first power supply module, and the high-speed interface of the device is electrically connected to the second power supply module; when a device to be unplugged is detected, outputting a device unplugging signal so that the host module responds to the device unplugging signal and disconnects the communication connection with the device to be unplugged; controlling the first power supply module and the second power supply module to disconnect the power supply to the device to be unplugged in descending order of power supply priority.

[0008] This application addresses the challenges of hot-swappable devices, which can be installed parallel to the server backplane. A guide section between the server housing and the connector guides the device's movement. Through coordinated control of the first and second power supply modules, power and communication are connected and disconnected in an orderly manner during device insertion or removal, ensuring safe hot-swapping. Simultaneously, the two power supply modules provide high-power, high-stability power after device insertion, meeting the application requirements of high-performance devices. This solves the problems faced by traditional PCIe devices in high-density deployments, such as physical space limitations, complex insertion / removal operations, the need for complete system power-off maintenance, and high risk of link reconnection. It achieves the beneficial effects of improving operational efficiency, avoiding business interruptions, reducing the probability of hardware damage, and ensuring data security. Attached Figure Description

[0009] 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.

[0010] Figure 1 This is a schematic diagram of the structure of an equipment management system provided in an embodiment of this application.

[0011] Figure 2 This is a schematic diagram of another device management system provided in an embodiment of this application.

[0012] Figure 3 This is a device insertion timing control diagram provided in an embodiment of this application. Detailed Implementation

[0013] 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.

[0014] It should be noted that, in the description of this application, 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. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0015] 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.

[0016] Please refer to Figure 1 The device includes a power supply interface and a high-speed interface. The device management system is set up and applied in the server. The server includes a housing, a guide section, and a backplate. The housing has a housing opening corresponding to the connector position on the backplate. The guide section is located between the housing opening and the connector. The guiding direction of the guide section is perpendicular to the plane of the housing opening. The guide section is configured to guide the device moving along the housing opening to insert and remove it from the server parallel to the backplate.

[0017] The equipment management system includes: a first power supply module 1 and a second power supply module 2. The output power of the first power supply module 1 is greater than the output power of the second power supply module 2. The power supply priority of the first power supply module 1 is higher than the power supply priority of the second power supply module 2. The power-off priority of the first power supply module 1 is lower than the power-off priority of the second power supply module 2.

[0018] Management module 3 is configured to, when a device is detected being inserted into a connector on the backplane, control the first power supply module 1 and the second power supply module 2 to start supplying power to the device in order of power supply priority from high to low, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, and control the first power supply module 1 and the second power supply module 2 to disconnect the power supply to the device to be removed in order of power supply priority from high to low.

[0019] The main control module 4 is configured to establish a communication connection with the device in response to the device insertion signal, and to disconnect the communication connection with the device to be removed in response to the device removal signal.

[0020] First, it should be noted that the server housing has a front panel, which is a side panel, and a back panel, which is horizontally positioned within the housing. The front and back panels should be roughly perpendicular to each other. Connectors are located on the back panel for mounting the aforementioned devices. A housing opening is provided on the front panel, and a guide is provided between the connector and the housing opening. The device can be inserted into the housing through the housing opening and guided to the connector position by the guide, allowing the device to connect with the connector. The device is equipped with a power supply interface and a high-speed interface. It is inserted through the housing opening on the side panel, moves horizontally relative to the back panel, and ultimately plugs into the connector on the back panel.

[0021] Understandably, the connector's insertion interface can be oriented towards the housing opening, meaning the interface direction is parallel to the backplate surface. This allows the device to be directly inserted into the connector after being inserted through the housing opening. In this configuration, the guide section can provide guidance to the device, ensuring accurate and smooth insertion.

[0022] The installation configuration of the structure and equipment provided above is an OCP (Open Compute Project) configuration.

[0023] It should be noted that the relevant technologies mainly rely on the following three solutions for DPU power supply: First, PCIe standard power supply (≤75W), which supplies power through the motherboard PCIe slot, suitable for low-power DPUs but unable to meet high-performance requirements; second, a combination of PCIe slot and external power supply (such as 6 / 8-pin power supply), but hot-swapping may cause inrush current due to timing issues; third, OCP NIC power supply (≤75W), using the OCP standard interface, but only suitable for low-power network cards and not for high-performance DPUs. Considering that PCIe form factors are mostly installed in servers with a riser, the installation path is generally parallel to the server panel, requiring the server cover to be opened for disassembly. For high-density servers, opening the cover requires powering off and unplugging the entire node. In contrast, the OCP form factor is installed perpendicular to the server panel, allowing direct unplugging from the panel. The current power supply power under the OCP protocol does not meet the DPU power supply requirements.

[0024] This embodiment takes into account the inconvenience of replacing and installing PCIe-based DPU devices. Therefore, it selects a DPU device with an Open Compute Project (OCP) standard interface (high-speed interface), i.e., an OCP-based DPU device. Furthermore, considering the high power requirements of the DPU itself, a simple standard OCP link cannot meet the power supply requirements. Therefore, a main power supply circuit is added to provide additional power. Specifically, the power supply section of this embodiment adopts a dual 12V independent power supply architecture, divided into a main power supply module (i.e., the first power supply module 1 in this embodiment) and an OCP link power supply module (i.e., the second power supply module 2 in this embodiment). Power isolation design achieves hierarchical power management and safety isolation.

[0025] The first power supply module 1 has a higher output power (providing 300W of service power) and is mainly responsible for providing the main service power to the DPU device to meet its high power requirements. The second power supply module 2 has a relatively lower output power and mainly provides a 12V direct power supply to the PCIe / USB (Universal Serial Bus) physical layer and high-speed clock chip of the DPU device to support the normal operation of these components. Specifically, the first power supply module 1 can adopt a high-power power supply circuit design, capable of providing 300W of service power, while the second power supply module 2 is based on the OCP standard interface and uses its standardized 12V direct power supply to power specific components of the DPU device. At the same time, through cooperation with the first power supply module 1, it meets the overall power supply requirements of the DPU device.

[0026] In this embodiment, the management module 3 serves as the intelligent control center of the entire system. It is responsible for timing control of the power supply and de-energization of the first power supply module 1 and the second power supply module 2 based on the hot-plugging operations of the devices, ensuring the safety and reliability of power management. This can be achieved using a CPLD (Complex Programmable Logic Device). Specifically, the management module 3 detects device insertion and removal signals on the connectors on the backplane and utilizes the logic control function of the CPLD to precisely control the startup and de-energization sequence of the first power supply module 1 and the second power supply module 2 according to the set power supply and de-energization priorities. For example, when device insertion is detected, the first power supply module 1 is started first, and the second power supply module 2 is started only after it has stabilized. When the device needs to be removed, the power supply to the second power supply module 2 is disconnected first, and then the power supply to the first power supply module 1 is disconnected, avoiding device damage or data loss due to improper power supply or de-energization timing.

[0027] The main control module 4 is used to establish and disconnect communication connections with the device, responding to device insertion and removal signals from the management module 3 to achieve communication management of the device. Specifically, the main control module 4 connects to the DPU device via a communication interface. Upon receiving a device insertion signal from the management module 3, it immediately initiates a communication connection procedure to establish a stable communication link with the DPU device for management and control. When a device removal signal is received, the main control module 4 quickly disconnects the communication connection with the DPU device to ensure safe removal of the device while avoiding communication interference and data errors.

[0028] This embodiment selects an OCP-based DPU device, whose installation path is perpendicular to the server panel, allowing it to be directly pulled out from the panel without needing to open the server cover as with a PCIe-based device. This simplifies the device replacement and installation process, making it particularly suitable for high-density server environments and improving equipment maintenance efficiency. By adopting a dual 12V independent power supply architecture, combining the main power supply module and the OCP link power supply module, it not only fully utilizes the advantages of the OCP standard interface but also solves the high-power supply problem of the DPU device by adding a main power supply circuit, ensuring stable operation of the DPU device and maximizing its high-performance computing capabilities. The high-precision power timing control and status management functions of management module 3 can precisely control the power supply and power-off sequence based on the device's hot-plugging operations, effectively avoiding safety hazards such as surge currents caused by timing issues, improving the safety and reliability of the entire device management system, and extending the device's service life.

[0029] The front end of the guide section can be a flared structure to provide guidance and correction for its direction when it comes into contact with the head of the equipment.

[0030] The aforementioned housing or guide section is equipped with a locking element to lock the device in place when it moves to the connector insertion position. There are various types of locking elements; one method involves a hole in the housing with a pin inserted into it, which engages with the inserted device to prevent outward movement. To facilitate smooth insertion while preventing removal, a wedge-shaped structure is provided on the pin or device. This allows insertion without lifting the pin, but requires lifting the pin to control the device's movement out of the housing, thus ensuring stability during operation.

[0031] In one exemplary embodiment, the device management system further includes: an in-situ detection module configured to output a first in-situ signal when the device is inserted into the connector; the management module 3 is further configured to determine the presence of a connector inserted into the backplane based on the first in-situ signal and a second in-situ signal of a target detection pin on the backplane.

[0032] The detection module includes a mechanical switch located in the connector, configured to output a first presence signal when in contact with the device and a first departure signal when not in contact with the device.

[0033] The device management system also includes an input monitoring layer, comprising a presence detection module and an environmental detection module. In this embodiment, the presence detection module is used to detect whether a device is inserted into a connector (which may be a slot on the backplane). Its core component is a mechanical switch located in the slot. When the device is inserted into the slot on the backplane, the mechanical switch is triggered, outputting a first presence signal; when the device is not inserted, it outputs a first absence signal. The backplane with the slot also has a target detection pin (such as the PRSNT# pin) for generating a second presence signal. Specifically, the voltage state of the detection pin is used to confirm whether the device is inserted into the slot. When the device is inserted, the pin voltage changes, thereby outputting a signal indicating that the device is in place.

[0034] Specifically, the detection module includes a high-reliability microswitch installed in a slot. When the DPU device is inserted into the slot, the microswitch is triggered, generating a low-level IN_PRESENT# signal, indicating that the device has been inserted. Simultaneously, the CPLD confirms physical insertion by detecting voltage changes on dedicated detection pins (such as PRESENT_DETECT) on the backplane. For example, when the device is inserted, the pin voltage changes from a pull-up state to a pull-down state, equivalent to outputting a second presence signal. After detecting both the first and second presence signals, the CPLD determines that a device has been inserted into the slot on the backplane. This dual mechanism of mechanical microswitches and electrical detection ensures accurate identification of the device insertion status, effectively avoiding false positives and improving the reliability and stability of device insertion status monitoring.

[0035] As an optional embodiment, the environmental monitoring unit integrates a temperature sensor and a voltage / current monitoring circuit to collect key parameters such as onboard temperature, 12V power supply voltage, and supply current in real time. The temperature sensor monitors the onboard temperature to ensure the equipment operates within a safe temperature range; the voltage / current monitoring circuit monitors the voltage and current of the 12V power supply in real time to ensure the stability and safety of the power supply. This monitoring data can be collected and analyzed by the management module 3 to promptly detect and handle potential faults or anomalies.

[0036] In an exemplary embodiment, the management module 3 is specifically configured to determine that a device is inserted into a connector on the backplane when the first presence signal and the second presence signal on the target detection pin on the backplane are continuous for a preset time period.

[0037] In this embodiment, the preset time period refers to a specific time threshold used to confirm the device insertion status, preventing misjudgments caused by signal jitter. For example, the preset time period can be set to 10 milliseconds; the device insertion is only confirmed as valid when the detected signal remains low for more than this time threshold. The debouncing timer is a timing mechanism used to eliminate signal jitter. It is activated when the CPLD detects a signal change. In this embodiment, the debouncing timer has a timing duration of 10 ms to ensure signal stability and avoid misjudgments caused by brief signal fluctuations.

[0038] Specifically, when the DPU device is inserted into the slot, a mechanical microswitch is triggered, generating a first presence signal (low level). Simultaneously, the CPLD detects that the voltage of the target detection pin (e.g., PRESENT_DETECT) on the backplane changes from a pull-up state to a pull-down state, generating a second presence signal. Upon receiving these two signals, the CPLD starts a 10ms debouncing timer. During these 10ms, if both signals remain low, the CPLD confirms the device insertion is valid; if the signals fluctuate within the 10ms (e.g., a brief high level), it is considered jitter, and the insertion is not confirmed.

[0039] The CPLD performs an AND operation on the mechanical detection signal (first presence signal) and the electrical detection signal (second presence signal). Only when both signals simultaneously meet the condition of remaining low for more than 10ms will the CPLD output a device insertion confirmation signal, notifying the main control module 4 that the device has been inserted and subsequent power supply and communication operations can proceed. This embodiment ensures signal stability and avoids erroneous operations triggered by brief signal fluctuations, such as incorrect power supply or communication connections. This helps reduce system failures and improves the stability and reliability of the entire device management system.

[0040] In an exemplary embodiment, the first power supply module 1 includes a main power supply, a timing control unit, and a monitoring unit. The management module 3 is specifically configured to, when a device is detected inserted into a connector on the backplane, control the main power supply to start and enable the timing control unit, while simultaneously controlling the second power supply module 2 to disconnect power to the device until device initialization is complete. Then, it controls the second power supply module 2 to start powering to the device and outputs a device insertion signal. When a device to be removed is detected, it outputs a device removal signal and controls the first power supply module 1 and the second power supply module 2 to disconnect power to the device to be removed in descending order of power priority. The timing control unit is configured to perform a soft-start operation, gradually increasing its output voltage from an initial value to a target voltage. The monitoring unit is configured to output a power good signal when the output voltage of the timing control unit rises to the target voltage, so that the device initializes based on the power good signal.

[0041] Reference Figure 2As shown, the first power supply module 1 includes a main power supply and a timing control unit (…). Figure 2 (not shown in the image), monitoring unit ( Figure 2 (Not shown in the diagram), main power supply hot-swappable controller, and main power domain of DPU. The power supply path of the first power supply module 1 is that the AC / DC main power supply generates 12V / 300W output, which is connected to the main power supply hot-swappable controller through the power supply interface on the back panel, and then sent to the main power domain of DPU.

[0042] The timing control unit starts under the control of management module 3, performing a soft start operation to gradually increase the output voltage from its initial value to the target voltage. This process is achieved through control circuitry to ensure a smooth voltage rise and prevent damage to the equipment from sudden high current. The monitoring unit monitors the output voltage of the timing control unit in real time. When the output voltage rises to the target voltage, the monitoring unit outputs a power good signal, notifying the equipment that initialization can begin.

[0043] In this embodiment, when the management module 3 detects a device inserted into a connector on the backplane, it controls the main power supply to start and enables the timing control unit, while simultaneously controlling the second power supply module 2 to disconnect power to the device. The timing control unit begins a soft-start operation, and the monitoring unit monitors the output voltage. When the output voltage rises to the target voltage, the monitoring unit outputs a power good signal, and the device initializes based on this signal. After initialization, the management module 3 controls the second power supply module 2 to start supplying power to the device and outputs a device insertion signal. When the management module 3 detects a device to be removed, it outputs a device removal signal and controls the first power supply module 1 and the second power supply module 2 to disconnect power to the device in descending order of power priority.

[0044] In this embodiment, the smooth start-up process of the power supply is ensured by the soft start-up operation of the timing control unit and the voltage monitoring of the monitoring unit, avoiding damage to the equipment caused by instantaneous high current. Simultaneously, the output of a good power signal ensures that the equipment initializes under stable power conditions, improving the reliability of the power supply. The management module 3 precisely controls the power supply and de-energization sequence of the first power supply module 1 and the second power supply module 2 when the equipment is inserted and removed, preventing equipment damage or data loss due to timing issues. This precise timing control improves the stability and reliability of the entire system.

[0045] In one exemplary embodiment, the timing control unit includes a soft-start circuit or a hot-swap controller.

[0046] In this embodiment, the soft-start circuit consists of a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) and an RC (Resistor-Capacitor) circuit to control the voltage rise process. By adjusting the time constant of the RC circuit, the time it takes for the voltage to rise from 0V to the target voltage (e.g., 12V) can be controlled. Specifically, the MOSFET acts as a switching element, controlling the power supply's on / off state; the RC circuit controls the voltage rise slope. By selecting appropriate resistor and capacitor values, the time for the voltage to rise from 0V to 12V can be controlled within 5ms. For example, choosing a suitable RC time constant allows the voltage to smoothly rise from 0V to 12V within 5ms.

[0047] Alternatively, a dedicated hot-swap controller can be used to control the voltage rise rate. The hot-swap controller integrates voltage control circuitry, and the voltage rise time can be set via external configuration (such as resistors and capacitors). For example, the hot-swap controller can be configured to rise the voltage from 0V to 12V within 5ms.

[0048] The monitoring unit monitors the 12V voltage in real time, and the timing control unit outputs the voltage when it reaches the normal value. When the voltage reaches 10.8V or higher, a PWR_GOOD signal is sent to the CPLD. The monitoring unit includes a voltage detection circuit and a comparator to accurately measure the voltage and determine whether a set threshold has been reached. For example, the power monitoring circuit can use a voltage divider circuit to divide the 12V voltage to a suitable range, and then compare it with a 10.8V reference voltage via a comparator. When a voltage of 10.8V or higher is detected, the comparator outputs a high-level signal, which is sent to the CPLD as the PWR_GOOD signal.

[0049] A soft-start circuit or hot-swap controller controls the voltage rise rate to ensure a smooth power supply startup process, preventing damage to equipment caused by instantaneous high current. The power monitoring circuit monitors the 12V voltage in real time and activates the circuit once the voltage reaches its normal value. The above steps involve sending a PWR_GOOD signal to the CPLD. This mechanism ensures that the device initializes under stable power conditions, avoiding initialization failures or abnormal situations caused by unstable power supply, thus improving system reliability.

[0050] In an exemplary embodiment, the second power supply module 2 includes a main power supply and a shielding switch disposed on the power supply path between the main power supply and the connector; the management module 3 is specifically configured to control the main power supply to start when a device is detected inserted into the connector on the backplane, and simultaneously output a first control signal until the device initialization is completed, and then output a second control signal and a device insertion signal in sequence; when a device to be removed is detected, it outputs a device removal signal, and controls the first power supply module 1 and the second power supply module 2 to disconnect the power supply to the device to be removed in descending order of power priority; the shielding switch is configured to disconnect in response to the first control signal and connect in response to the second control signal.

[0051] Reference Figure 2 The second power supply module 2 includes an AC / DC main power supply and an OCP link power supply (i.e., Figure 2 The system includes the power supply link (link power), shielded switch, and DPU OCP link power domain, as well as power isolation diodes. The shielded switch is a switching element located on the main power supply and slot power supply path, used to control the on / off state of the power supply path of the second power supply module 2. The shielded switch can respond to control signals issued by the management module 3, achieving precise control of the power supply path. The shielded switch can use MOSFETs or other suitable switching elements, enabling rapid response to control signals issued by the management module 3. When the management module 3 detects the connector inserted into the backplane, it controls the main power supply to start and outputs a first control signal to open the shielded switch, temporarily disconnecting the power supply from the second power supply module 2 to the device. At this time, the device is initialized by the first power supply module 1 alone. After the device initialization is complete, the management module 3 sequentially outputs a second control signal and a device insertion signal. The second control signal closes the shielded switch, restoring the power supply from the second power supply module 2 to the device; the device insertion signal notifies the main control module 4 that the device has been inserted and initialized, allowing subsequent communication connections and other operations. When the management module 3 detects a device to be unplugged, it outputs a device unplugging signal and controls the first power supply module 1 and the second power supply module 2 to disconnect the power supply to the device in descending order of power priority.

[0052] Specifically, after confirming the DPU device is inserted into the slot, the CPLD starts the main power supply by setting the PWR_EN signal to a high level. This signal is connected to the enable pin of the backplane power controller. The main power supply (12V) is gradually powered on through a soft-start circuit to avoid inrush current. Simultaneously with starting the main power supply, the CPLD sets the OCP_MASK# signal to active (high level), which controls a MOSFET switch (i.e., a shield switch) to disconnect the 12V power supply to the OCP link. Thus, the OCP link (such as PCIe) is powered down until the DPU initialization is complete, preventing link signals from being driven before the DPU is ready.

[0053] After receiving the initialization completion signal, the CPLD first removes the shielding of the OCP link (pulling the OCP_MASK# signal low) and restores the 12V power supply. Then, the CPLD sends a hot-plug interrupt signal to the host via the System Management Bus (SMBus) (triggering a PCIe Hot-Plug Interrupt), notifying the host that a new device has been inserted. The host then begins enumerating DPU devices, establishing PCIe links, loading drivers, and completing the logical connections.

[0054] Combination Figure 3 Upon 12V input (instantaneous insertion), the DPU device connects to the 12V external power interface on the backplane. The main power input serves as the startup source for the entire system, triggering subsequent power management processes. The internal power management IC (PMIC) is activated and begins managing various voltage levels on the backplane. After a successful soft-start of the main power supply module, a PWR_GOOD signal is returned. The PMIC completes its self-test and confirms stable output voltage before issuing a high-level PWR_GOOD signal. Execution of st0: The CPLD disables the OCP power supply request, triggered simultaneously with the DPU's connection to the 12V external power interface. The system actively disables the OCP link's power supply request, ensuring that the DPU's core modules, such as the CPU and memory, are initialized first.

[0055] After the main power supply stabilizes, the DPU device starts up and executes its firmware initialization process, including clock initialization, DDR (Double Data Rate) memory self-test, and internal register configuration. Specifically, this includes the following steps: st1: Start CPU core initialization: load firmware and configure registers; st2: Complete DDR memory self-test: check capacity / errors to ensure reliable data path; IO module initialization: configure basic peripherals (such as UART (Universal Asynchronous Receiver / Transmitter) and I2C (Inter-Integrated Circuit) controllers). After the DPU system loads and hardware initialization is complete, a SYS_READY pulse (st4) is sent, indicating that the core system is ready. The CPLD releases the OCP shield (st5). Upon receiving SYS_READY, the system allows the OCP link to request power, and then performs link training and data path establishment according to the OCP protocol specification, starting DPU link training: the CPU and DPU card negotiate communication parameters (rate, equalization) through a high-speed bus (such as PCIe). Link training completes: both parties confirm that signal integrity meets the standards. Establishing a data path: Data transmission permissions are granted via the DPU driver. After the link is established, the DPU driver detects the completion signal and, through the driver under the host, triggers the DPUFW to establish the data transmission channel. The DPU card can then read and write to the host memory or interact with the host CPU. The system is ready; all modules have stable power supply, hardware and software initialization is complete, and the data path is unobstructed. The system enters standby or running state, ready to respond to upper-layer commands and workloads.

[0056] As an optional embodiment, the link negotiation process is described below: The high-speed bus negotiation process mainly includes: pre-configuring accelerated training mode, storing device parameter templates in the CPLD's internal Flash storage, and indexing them by DPU model (e.g., {DeviceID:0x1A2B, Template:{Rate:32GT / s,Equalization:CTLE=6dB,DFE=5tap,TxPre=3dB}}). When the DPU enters the training process, the CPLD reads the DeviceID from its EEPROM (Electrically Erasable Programmable Read-Only Memory) via I2C and retrieves the pre-set matching template: if the DeviceID is successfully matched, the parameters are directly loaded into the PHY (Physical Layer) controller; if no match is found, the default template is enabled, and the conservative parameter 8GT / s is used for minimum equalization initialization training. By loading pre-configurations, the full-rate scan in standard training is skipped, reducing the step-by-step ramp-up time of LTSSM (Link Training and Status State Machine) control; deep collaboration between the power system and communication link: forced timing synchronization is performed: after the CPLD removes the OCP link power supply shield ( This immediately triggers the DPU physical layer to send a Beacon signal to the Host and synchronously loads pre-configured parameters (rate / equalization). The power supply system during the training startup phase ( Locking out output voltage fluctuations ( Through the coordinated action of a synchronous buck controller and a low-dropout linear regulator, the power supply ripple of the PHY chip is kept within a small range, preventing voltage disturbances from causing training failures. During training, a linked protection mechanism is implemented to prevent training failures. If the link training fails consecutively, the CPLD executes a three-level response: degrade power supply mode (switching to current-limiting mode); reset the DPU's PCIe controller (triggering the PERST# pulse); and fall back to the lowest rate template for retry.

[0057] The training failure flag signal from the HostPCIe controller is detected. If three failure signals are detected, the gate voltage of the current-limiting MOSFET is controlled to hard-limit the output current to a low current, putting the device into a low-power mode. The data rate can also be reduced accordingly to ensure the device is not directly blocked. This is because repeated link training failures indicate an underlying hardware anomaly (such as impedance mismatch or clock jitter). Maintaining high current in this situation could lead to: a sharp increase in PHY chip junction temperature; amplified power rail noise; and potential damage to PCB traces. A power retry mechanism reduces performance to ensure functionality, allowing for subsequent problem repair. If the retry still fails, the CPLD records the error code, then controls the lighting of the fault LED, and disables the OCP link power supply, isolating the device hardware.

[0058] In an exemplary embodiment, the management module 3 is specifically configured to, when a device is detected being inserted into a connector on the backplane, control the first power supply module 1 and the second power supply module 2 to start supplying power to the device in descending order of power supply priority, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, a first control signal, and a power-off request signal in sequence; and when a feedback signal is received, control the first power supply module 1 to disconnect the power supply to the device to be removed. The main control module 4 is specifically configured to, in response to the device insertion signal, establish a communication connection with the device; in response to the device removal signal, stop data transmission with the device to be removed and uninstall the driver; the device to be removed is configured to, in response to the power-off request signal, perform a safe shutdown operation and output a feedback signal after the safe shutdown operation is completed.

[0059] In this embodiment, power supply priority refers to the order in which the first power supply module 1 and the second power supply module 2 are started. The power supply priority of the first power supply module 1 is higher than that of the second power supply module 2. That is, when the device is inserted, the first power supply module 1 is started first, followed by the second power supply module 2. The device insertion signal is a signal output by the management module 3, which notifies the main control module 4 that the device has been inserted into the connector on the backplane and a communication connection can be established. The device removal signal is a signal output by the management module 3, which notifies the main control module 4 that the device is about to be removed, and data transmission needs to be stopped and the driver uninstalled. The shutdown request signal is a signal output by the management module 3, which notifies the device to be removed to perform a safe shutdown operation to ensure that the device is in a safe state before being removed. The feedback signal is a signal output by the device to be removed after performing the safe shutdown operation, which notifies the management module 3 that the device is ready to be removed and the power supply to the first power supply module 1 can be disconnected. The safe shutdown operation refers to a series of operations performed by the device to be removed after receiving the shutdown request signal, including saving data, closing applications, and shutting down the system, to ensure that the device is not lost or damaged when it is removed.

[0060] When management module 3 detects a device inserted into the connector on the backplane, it first controls the first power supply module 1 to start supplying power to the device, then controls the second power supply module 2 to start supplying power to the device, and outputs a device insertion signal to notify the main control module 4 that the device has been inserted, according to the power supply priority from high to low. When management module 3 detects a device to be removed, it outputs a device removal signal, a first control signal, and a power-off request signal in sequence. The device removal signal notifies the main control module 4 that the device is about to be removed. The main control module 4 responds to this signal by stopping data transmission with the device to be removed and uninstalling the driver. The first control signal controls the shielding switch of the second power supply module 2 to open, disconnecting the power supply from the second power supply module 2 to the device. The power-off request signal notifies the device to be removed to perform a safe shutdown operation. After performing the safe shutdown operation, the device to be removed outputs a feedback signal to management module 3. Upon receiving the feedback signal, management module 3 controls the first power supply module 1 to disconnect the power supply to the device, completing the device removal process.

[0061] The main control module 4 responds to the device insertion signal, establishes a communication connection with the device, and ensures that the device can communicate and transmit data normally. The main control module 4 also responds to the device removal signal, stops data transmission with the device to be removed, and uninstalls the driver associated with the device to ensure that removing the device will not affect the system.

[0062] The device to be unplugged responds to the power-off request signal and performs a safe shutdown operation, including saving data, closing applications, and shutting down the system, to ensure that the device is in a safe state when it is unplugged. After the safe shutdown operation is completed, the device to be unplugged outputs a feedback signal to the management module 3, notifying the management module 3 that the device is ready to be unplugged.

[0063] The management module 3 sequentially outputs a device unplug signal, a first control signal, and a shutdown request signal, ensuring that the device performs a safe shutdown operation before being unplugged. This orderly process avoids data loss or device damage caused by sudden device unplugging, improving the safety of device unplugging. Upon receiving the device unplug signal, the main control module 4 stops data transmission with the device to be unplugged and uninstalls the driver, ensuring the reasonable release of system resources and avoiding resource waste and potential system conflicts. Through a feedback signal mechanism, the management module 3 only disconnects the power supply to the first power supply module after confirming that the device to be unplugged has been safely shut down. This mechanism ensures the reliability of the entire unplugging process, avoiding device damage or data loss due to power interruption, and improving the overall stability and reliability of the system.

[0064] Specifically, during the unplugging process, the CPLD prioritizes disconnecting the OCP link power supply, and only cuts off the main power supply after the DPU service is safely shut down. A timeout management mechanism is also added, setting a 200ms initialization timeout threshold. If the DPU fails to provide an initialization completion signal on time, an error handling procedure is triggered. The timeout threshold is set based on the following factors: minimum hardware requirement: 10ms (physical power failure); software backup requirement: 160ms (service safe exit); signal transmission requirement: 5ms; redundancy requirement: 25m; service safe exit is assessed based on experience, and the overall data is based on empirical evaluation.

[0065] In this embodiment, manual triggering of the unplugging occurs when the user presses the EjectButton on the front panel. This button generates a low-level EJECT_REQ# signal to the CPLD. Upon receiving the unplugging request, the CPLD first notifies the host system (via SMBus) to begin disconnecting the PCIe link. The host stops data transmission with the DPU and uninstalls the driver. Then, the CPLD sets the OCP_MASK# signal, cutting off the 12V power supply to the OCP link to ensure the link is powered off. The CPLD sends a shutdown request signal (PWR_DOWN#, active low) to the DPU. Upon receiving this signal, the DPU begins the safe shutdown process: suspending all data processing tasks; writing the data in the cache to non-volatile memory (such as NVMe SSD); closing the network port and stopping packet transmission and reception; after completing the shutdown, the DPU sends a SHUTDOWN_OK signal (0x55) to the CPLD via I²C.

[0066] In an exemplary embodiment, the management module 3 is specifically configured to, when a device is detected being inserted into a connector on the backplane, control the first power supply module 1 and the second power supply module 2 to start supplying power to the device in descending order of power supply priority, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, a first control signal, and a power-off request signal in sequence; and when a feedback signal is received, control the main power supply to be turned off; the timing control unit is further configured to, after the main power supply is turned off, perform a soft shutdown operation to gradually reduce its own output voltage to the initial value.

[0067] In this embodiment, after receiving SHUTDOWN_OK (a signal output by the device to be unplugged after completing a safe shutdown operation, notifying the management module 3 that the device is ready to be unplugged and the main power can be turned off), the CPLD sets the PWR_EN signal to a low level and shuts off the main power (12V). The main power controller performs a power-off operation, which typically includes a soft shutdown process (voltage ramp-down) to protect the device from the impact of a momentary power outage.

[0068] The management module 3 sequentially outputs a device unplug signal, a first control signal, and a shutdown request signal, ensuring a safe shutdown operation is performed before the device is unplugged. This orderly process avoids data loss or device damage caused by sudden unplugging, improving the safety of device unplugging. The timing control unit performs a soft shutdown operation after the main power is turned off, gradually reducing the output voltage to its initial value instead of a sudden power outage. This soft shutdown process protects the device from the impact of sudden power outages, extending its lifespan.

[0069] In addition, during the unplugging process, if the DPU does not return SHUTDOWN_OK within a predetermined time (e.g., 100ms), the CPLD will trigger the hardware watchdog to reset the DPU and force a power-off to prevent the system from hanging, avoid system instability or data loss due to equipment failure, and enhance the system's fault tolerance.

[0070] In one exemplary embodiment, a prompting module is also included; the management module 3 is further configured to output a drive signal when the main power supply is turned off; the prompting module is configured to respond to the drive signal to prompt information corresponding to the unplugging operation.

[0071] In this embodiment, the prompting module refers to the module used to prompt the user with information related to the device being unplugged. It can use various methods, such as LED indicators, sound prompts, or display screens, to notify the user of the device's status. When the management module 3 controls the main power supply to shut down, it outputs a drive signal to the prompting module. The output of the drive signal indicates that the device is in a safe unplugged state, and the prompting module performs corresponding prompting operations based on this signal.

[0072] In this embodiment, after the main power supply is turned off, the management module 3 (CPLD) illuminates an LED indicator by outputting a drive signal, causing it to flash green to indicate to the user that it is safe to unplug the DPU. By using an indicator module (such as an LED indicator) to intuitively remind the user that the device can be safely unplugged, the user does not need to check the device status separately; they can directly unplug the device based on the flashing indicator, improving the convenience of operation and user experience.

[0073] In an exemplary embodiment, the first power supply module 1 further includes a backplane interface connected to the main power supply. The backplane interface is equipped with a detection resistor and a protection switch. The first end of the detection resistor is connected to the input terminal of the backplane interface, the first end of the protection switch is connected to the second end of the detection resistor, and the second end of the protection switch is connected to the output terminal of the backplane interface. The management module 3 is further configured to control the protection switch to open when the current flowing through the detection resistor meets the safety protection conditions. In this embodiment, the backplane interface refers to the interface connecting the first power supply module 1 to the backplane, used to transmit power from the main power supply to the devices on the backplane. The backplane interface includes an input terminal and an output terminal. The first end of the detection resistor is connected to the input terminal of the backplane interface, and the second end is connected to the first end of the protection switch. The detection resistor is used to detect the current flowing through the backplane interface. The first end of the protection switch is connected to the second end of the detection resistor, and the second end is connected to the output terminal of the backplane interface. Under the control of the management module 3, the protection switch opens or closes the circuit according to the current of the detection resistor. When the management module 3 detects that the current flowing through the detection resistor meets the safety protection conditions (such as the current exceeding a preset threshold), it controls the protection switch to open, cutting off the circuit and protecting the device from overcurrent damage. The power supply hot-swap controller includes inrush current suppression and overcurrent protection. Inrush current suppression is achieved by controlling the voltage rise slope (e.g., a 5ms ramp-up from 0V to 12V) to avoid high current surges during power connection. A high-precision current sensing resistor is deployed at the backplane interface, and the voltage difference across the sensing resistor is amplified by a differential amplifier, converting it into a measurable voltage signal. CPLD logic controls the comparator signal, which in turn controls the blocking switch, ensuring unidirectional current flow, preventing reverse current, protecting the device from current surges, and improving the overall reliability of the system.

[0074] In summary, this invention provides secure hot-swapping, supporting millisecond-level DPU insertion and removal, preventing data loss or hardware damage, and overcoming the inconvenience of inserting and removing existing PCIe-based DPU cards. Traditional PCIe-based DPU cards are prone to data loss and hardware damage during insertion and removal. This technology offers millisecond-level insertion and removal, intelligently protecting data and hardware, thus eliminating inconvenience. It also provides high-power support for the DPU card, expanding OCP power supply to 300W to meet the needs of high-performance DPUs. High-performance DPUs require more power, which traditional OCP power supply cannot provide. Expanding to 300W, with optimized design and stable power supply, meets their performance requirements. Simultaneously, it ensures standardization compatibility, extending the OCP protocol to ensure compatibility with existing data center architectures. Data center architectures rely on standard protocols. The OCP protocol extension makes the DPU compatible with existing architectures, facilitating deployment and maintenance.

[0075] This application also provides a server, including the device management system provided in any of the above embodiments, and further including a server housing, a guide portion, and a backplate. The housing has a housing opening corresponding to the connector position of the backplate. The guide portion is located between the housing opening and the connector, and the guiding direction of the guide portion is perpendicular to the plane of the housing opening. The guide portion is configured to guide a device moving along the housing opening to insert into and remove from the server parallel to the backplate. For more information about the server, please refer to the above embodiments.

[0076] To address the aforementioned technical problems, this invention also provides a device management method, applied to a management module in a device management system as described in any of the embodiments above. The device management method includes: when a device is detected inserted into a connector on a backplane, controlling a first power supply module and a second power supply module in the device management system to start supplying power to the device in descending order of power supply priority, and outputting a device insertion signal so that the host module in the device management system responds to the device insertion signal and establishes a communication connection with the device; wherein, the output power of the first power supply module is greater than the output power of the second power supply module, the power supply priority of the first power supply module is higher than the power supply priority of the second power supply module, and the power-off priority of the first power supply module is lower than the power-off priority of the second power supply module; when the device engages with the connector, the device's power supply interface is electrically connected to the first power supply module, and the device's high-speed interface is electrically connected to the second power supply module; when a device to be unplugged is detected, outputting a device unplugging signal so that the host module responds to the device unplugging signal and disconnects the communication connection with the device to be unplugged; controlling the first power supply module and the second power supply module to disconnect the power supply to the device to be unplugged in descending order of power supply priority.

[0077] Embodiments of this application also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the device management method embodiments described above.

[0078] Embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above-described device management method embodiments when it is run.

[0079] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0080] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described device management method embodiments.

[0081] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described device management method embodiments.

[0082] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0083] The above provides a detailed description of the device management system, server, and method provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. An equipment management system, characterized in that, The device includes a power supply interface and a high-speed interface. The server includes a housing, a guide section, and a backplate. The housing has a housing opening corresponding to the connector position on the backplate. The guide section is located between the housing opening and the connector. The guiding direction of the guide section is perpendicular to the plane of the housing opening. The guide section is configured to guide the device moving along the housing opening to insert and remove it from the server parallel to the backplate. The device management system includes: A first power supply module and a second power supply module, wherein the output power of the first power supply module is greater than that of the second power supply module, the power supply priority of the first power supply module is higher than that of the second power supply module, and the power-off priority of the first power supply module is lower than that of the second power supply module; when the device engages with the connector, the power supply interface of the device is electrically connected to the first power supply module, and the high-speed interface of the device is electrically connected to the second power supply module; the first power supply module provides service power to the device, and the second power supply module provides power to the PCIe / USB physical layer and high-speed clock chip of the device; The management module is configured to, when the device is detected to be inserted into the connector, control the first power supply module and the second power supply module to start supplying power to the device in order of power supply priority from high to low, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, and control the first power supply module and the second power supply module to disconnect the power supply to the device to be removed in order of power-off priority from high to low. The main control module is configured to establish a communication connection with the device in response to the device insertion signal, and to disconnect the communication connection with the device to be removed in response to the device removal signal. The first power supply module includes a main power supply, a timing control unit, and a monitoring unit; The management module is specifically configured to, when a device is detected being inserted into a connector on the backplane, control the main power supply to start and enable the timing control unit, and simultaneously control the second power supply module to disconnect the power supply to the device until the device initialization is complete, control the second power supply module to start supplying power to the device and output a device insertion signal; when a device to be unplugged is detected, output a device unplugging signal, and control the first power supply module and the second power supply module to disconnect the power supply to the device to be unplugged in descending order of power-down priority; The timing control unit is configured to perform a soft-start operation to gradually increase its output voltage from an initial value to a target voltage. The monitoring unit is configured to output a power good signal when the output voltage of the timing control unit rises to the target voltage, so that the device initializes based on the power good signal. The second power supply module includes the main power supply and a shielding switch disposed on the power supply path between the main power supply and the connector; The management module is specifically configured to control the main power supply to start when the device is detected to be inserted into the connector on the backplane, and output a first control signal until the device initialization is completed. Then, it outputs a second control signal and a device insertion signal in sequence. When a device to be removed is detected, it outputs a device removal signal and controls the first power supply module and the second power supply module to disconnect the power supply to the device to be removed in the order of power-down priority from high to low. The shielding switch is configured to disconnect in response to the first control signal and connect in response to the second control signal; The management module is specifically configured to, when a device is detected to be inserted into a connector on the backplane, control the first power supply module and the second power supply module to start supplying power to the device in order of power supply priority from high to low, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, the first control signal and a power-off request signal in sequence; and when a feedback signal is received, control the first power supply module to disconnect the power supply to the device to be removed. The main control module is specifically configured to establish a communication connection with the device in response to the device insertion signal, and to stop data transmission with the device to be unplugged in response to the device removal signal, and to uninstall the driver. The device to be unplugged is configured to perform a safe shutdown operation in response to the power-off request signal, and to output the feedback signal after the safe shutdown operation is completed.

2. The equipment management system according to claim 1, characterized in that, The equipment management system also includes: The presence detection module is configured to output a first presence signal when the device is inserted into the connector; The management module is also configured to determine the presence of a connector on the backplane based on the first presence signal and a second presence signal of the target detection pin on the backplane.

3. The equipment management system according to claim 2, characterized in that, The management module is specifically configured to determine that a device is inserted into a connector on the backplane when the first presence signal and the second presence signal on the target detection pin on the backplane are continuous for a preset time period.

4. The equipment management system according to claim 1, characterized in that, The management module is specifically configured to, when a device is detected to be inserted into a connector on the backplane, control the first power supply module and the second power supply module to start supplying power to the device in order of power supply priority from high to low, and output a device insertion signal; when a device to be removed is detected, output a device removal signal, the first control signal and a power-off request signal in sequence; and when a feedback signal is received, control the main power supply to be turned off. The timing control unit is also configured to perform a soft shutdown operation when the main power supply is turned off, so that its output voltage gradually decreases to the initial value.

5. The equipment management system according to claim 4, characterized in that, The first power supply module also includes a backplane interface connected to the main power supply. The backplane interface is provided with a detection resistor and a protection switch. The first end of the detection resistor is connected to the input end of the backplane interface, the first end of the protection switch is connected to the second end of the detection resistor, and the second end of the protection switch is connected to the output end of the backplane interface. The management module is also configured to control the protection switch to open when the current flowing through the detection resistor meets the safety protection conditions.

6. A server, characterized in that, The device management system according to any one of claims 1-5 further includes a server housing, a guide portion, and a backplate. The housing is provided with a housing opening corresponding to the connector position of the backplate. The guide portion is located between the housing opening and the connector. The guiding direction of the guide portion is perpendicular to the plane where the housing opening is located. The guide portion is configured to guide a device moving along the housing opening to insert into and remove from the server parallel to the backplate.

7. A method for managing equipment, characterized in that, The management module is applied in the equipment management system as described in any one of claims 1-5, and the equipment management method includes: When a device is detected inserted into the connector on the backplane, the first and second power supply modules in the device management system are controlled to start supplying power to the device in descending order of power supply priority, and output a device insertion signal so that the host module in the device management system responds to the device insertion signal and establishes a communication connection with the device. The output power of the first power supply module is greater than that of the second power supply module, the power supply priority of the first power supply module is higher than that of the second power supply module, and the power-off priority of the first power supply module is lower than that of the second power supply module. When the device engages with the connector, the device's power supply interface is electrically connected to the first power supply module, and the device's high-speed interface is electrically connected to the second power supply module. When a device to be unplugged is detected, a device unplug signal is output so that the host module responds to the device unplug signal and disconnects the communication connection with the device to be unplugged. The first power supply module and the second power supply module are controlled to disconnect the power supply to the device to be unplugged in order of power priority from high to low. The first power supply module includes a main power supply, a timing control unit, and a monitoring unit; When a device is detected plugged into a connector on the backplane, the first and second power supply modules in the device management system are controlled to start supplying power to the device in descending order of power supply priority, and a device insertion signal is output so that the host module in the device management system can respond to the device insertion signal and establish a communication connection with the device. When a device to be unplugged is detected, a device unplugging signal is output so that the host module can respond to the device unplugging signal and disconnect the communication connection with the device to be unplugged. The process of controlling the first and second power supply modules to disconnect power to the device to be unplugged in descending order of power supply priority includes: When a device is detected being inserted into a connector on the backplane, the main power supply is activated and the timing control unit is enabled. Simultaneously, the second power supply module is deactivated to supply power to the device until the device initialization is complete. Then, the second power supply module is activated to supply power to the device and outputs a device insertion signal. When a device to be removed is detected, a device removal signal is output. The first and second power supply modules are deactivated to supply power to the device to be removed in descending order of power-down priority. Specifically, the timing control unit performs a soft start operation to gradually increase its output voltage from the initial value to the target voltage. When the output voltage of the timing control unit rises to the target voltage, the monitoring unit outputs a power good signal so that the device can initialize based on the power good signal. The second power supply module includes the main power supply and a shielding switch disposed on the power supply path between the main power supply and the connector; When a device is detected plugged into a connector on the backplane, the first and second power supply modules in the device management system are controlled to start supplying power to the device in descending order of power supply priority, and a device insertion signal is output so that the host module in the device management system can respond to the device insertion signal and establish a communication connection with the device. When a device to be unplugged is detected, a device unplugging signal is output so that the host module can respond to the device unplugging signal and disconnect the communication connection with the device to be unplugged. The process of controlling the first and second power supply modules to disconnect power to the device to be unplugged in descending order of power supply priority includes: When a device is detected inserted into a connector on the backplane, the main power supply is activated, and a first control signal is output. This continues until the device initialization is complete. A second control signal and a device insertion signal are then output sequentially. When a device to be removed is detected, a device removal signal is output. The first and second power supply modules are then controlled to disconnect power to the device to be removed, according to the power-down priority from high to low. The shielding switch disconnects in response to the first control signal and connects in response to the second control signal. When a device is detected plugged into a connector on the backplane, the first and second power supply modules in the device management system are controlled to start supplying power to the device in descending order of power supply priority, and a device insertion signal is output so that the host module in the device management system can respond to the device insertion signal and establish a communication connection with the device. When a device to be unplugged is detected, a device unplugging signal is output so that the host module can respond to the device unplugging signal and disconnect the communication connection with the device to be unplugged. The process of controlling the first and second power supply modules to disconnect power to the device to be unplugged in descending order of power supply priority includes: When a device is detected being inserted into a connector on the backplane, the first power supply module and the second power supply module are controlled to start supplying power to the device in descending order of power supply priority, and a device insertion signal is output. When a device to be removed is detected, a device removal signal, the first control signal, and a power-off request signal are output in sequence. When a feedback signal is received, the first power supply module is controlled to disconnect the power supply to the device to be removed. Specifically, the main control module establishes a communication connection with the device in response to the device insertion signal, stops data transmission with the device to be unplugged in response to the device removal signal, uninstalls the driver, performs a safe shutdown operation in response to the power-off request signal from the device to be unplugged, and outputs the feedback signal after the safe shutdown operation is completed.

Images