Power consumption adjusting method and device and electronic equipment

Abstract

The invention provides a power consumption adjusting method, which comprises the following steps of: responding to a condition that a first power supply source of electronic equipment stops supplying power and the electric quantity of a second power supply source of the electronic equipment is less than or equal to a preset electric quantity threshold value, and acquiring use information of multiple pieces of hardware in the electronic equipment; the use information at least represents the use degree of the hardware in a preset time range; determining at least one piece of target hardware according to each piece of use information; and the power consumption of each target hardware is reduced.

Description

Technical Field

[0001] This disclosure relates to the field of power control for electronic devices, and more specifically, to a power consumption adjustment method, apparatus, and electronic device. Background Technology

[0002] With the development of information technology, electronic devices, especially servers, data centers, and edge computing nodes, play an increasingly crucial role in the operation of society. To ensure that these devices can continue to provide services or complete critical tasks in the face of unexpected situations such as power outages, they are typically equipped with uninterruptible power supplies (UPS) and backup battery packs as secondary power sources. In emergency situations powered by these secondary sources, their energy reserves are usually limited. Therefore, effectively reducing the overall power consumption of electronic devices to maximize their battery life has become a significant technical challenge.

[0003] To address this, the industry has proposed several power management technologies. These technologies typically activate a preset, static power-saving strategy after detecting a system switch to backup power. They lack the ability to perceive and analyze the actual workload, and cannot achieve differentiated and targeted power control. As a result, their power-saving strategies come at the cost of excessive performance loss in some scenarios, while being ineffective in others. Summary of the Invention

[0004] In view of this, the present disclosure provides a power consumption adjustment method, apparatus and electronic device.

[0005] One aspect of this disclosure provides a power consumption adjustment method, comprising: in response to a first power source of an electronic device stopping power supply and a second power source of the electronic device having a power level less than or equal to a preset power threshold, acquiring usage information of multiple hardware components in the electronic device; the usage information at least characterizing the degree of use of the hardware within a preset time range; determining at least one target hardware component based on the usage information; and reducing the power consumption of each target hardware component.

[0006] According to embodiments of this disclosure, determining at least one target hardware includes: selecting hardware that is used at a frequency less than or equal to a preset frequency threshold within a preset time range as the target hardware.

[0007] According to embodiments of this disclosure, reducing the power consumption of various target hardware includes: in response to the target hardware having corresponding power consumption control parameters, reducing the power consumption of the target hardware based on the power consumption control parameters; wherein the power consumption control parameters include at least one of the target hardware's operating frequency and / or operating voltage, the target hardware's I / O bandwidth, and the target hardware's operating mode.

[0008] According to embodiments of this disclosure, reducing the power consumption of various target hardware includes: in response to the target hardware not having corresponding power consumption control parameters, reducing the frequency of use of the target hardware based on the current operating requirements of the electronic device, or disabling the function of the target hardware.

[0009] According to embodiments of this disclosure, a second power source supplies power to electronic devices after the first power source is powered off, and the power supply of the second power source is less than or equal to a preset power threshold.

[0010] According to embodiments of this disclosure, the information used also includes historical power consumption information of the hardware within a preset time range when the electronic device is powered by a second power source, reducing the power consumption of each target hardware, including: reducing the power consumption of the target hardware based on the historical power consumption information.

[0011] According to embodiments of this disclosure, reducing the power consumption of each target hardware includes: determining the power consumption reduction margin of the target hardware based on the remaining power of the second power source; and reducing the power consumption of the target hardware based on the power consumption reduction margin.

[0012] According to embodiments of this disclosure, reducing the power consumption of each target hardware includes: in response to the existence of multiple first hardwares with the same function in each target hardware, reducing the power consumption of at least some of the first hardwares among the multiple first hardwares, or turning off at least some of the first hardwares.

[0013] Another aspect of this disclosure provides a power consumption adjustment device, comprising: a first acquisition module, configured to acquire usage information of multiple hardware devices in an electronic device in response to a first power source of an electronic device stopping power supply and a second power source of the electronic device having a power level less than or equal to a preset power threshold; the usage information at least characterizes the degree of use of the hardware within a preset time range; a first determination module, configured to determine at least one target hardware device based on the usage information; and a first reduction module, configured to reduce the power consumption of each target hardware device.

[0014] Another aspect of this disclosure provides an electronic device including: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the power adjustment method of any of the foregoing embodiments.

[0015] Another aspect of this disclosure provides a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to perform a power consumption adjustment method according to any of the foregoing embodiments.

[0016] Another aspect of this disclosure provides a computer program product, including a computer program / instructions, characterized in that the computer program / instructions, when executed by a processor, implement the operation of the power consumption adjustment method of any of the foregoing embodiments. Attached Figure Description

[0017] The above and other objects, features and advantages of this disclosure will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0018] Figure 1 A flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0019] Figure 2 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically;

[0020] Figure 3 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically;

[0021] Figure 4 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically;

[0022] Figure 5 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically;

[0023] Figure 6 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically;

[0024] Figure 7 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically;

[0025] Figure 8 A block diagram schematically illustrates a power consumption adjustment device according to an embodiment of the present disclosure; and

[0026] Figure 9 A block diagram of an electronic device suitable for implementing the methods described above, according to embodiments of the present disclosure, is illustrated schematically. Detailed Implementation

[0027] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0028] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0029] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0030] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0031] In the embodiments disclosed herein, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.

[0032] Embodiments of this disclosure provide a power consumption adjustment method, comprising: in response to a first power source of an electronic device stopping power supply and a second power source of the electronic device having a power level less than or equal to a preset power threshold, acquiring usage information of multiple hardware components in the electronic device; the usage information at least characterizing the degree of use of the hardware within a preset time range; determining at least one target hardware component based on the usage information; and reducing the power consumption of each target hardware component.

[0033] Figure 1 A flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0034] like Figure 1 As shown, the power consumption adjustment method may include at least operations S110 to S130.

[0035] In operation S110, in response to the first power source of the electronic device stopping power supply and the power of the second power source of the electronic device being less than or equal to a preset power threshold, usage information of multiple hardware components in the electronic device is obtained; the usage information at least characterizes the degree of use of the hardware within a preset time range.

[0036] The primary power source can be mains power or an external power grid. The secondary power source can be an energy storage device, such as an uninterruptible power supply (UPS) or a backup battery bank, used to provide emergency power when the primary power source is interrupted. The preset power threshold is a critical value used to trigger power consumption adjustment strategies; for example, it could be 50% or 30% of the total power, or any value configured by the system administrator based on business importance. Usage information is statistical data used to assess the activity level of each hardware component, which may include, but is not limited to, hardware access counts, data throughput, percentage of active time, or number of executed operation commands. The preset time range is a historical time window for collecting and analyzing usage information, such as the past month, two months, or one quarter.

[0037] For example, consider a server deployed in an edge computing environment. The server's primary power source is the data center's mains power, and its secondary power source is a UPS. When the mains power fails, the server switches to UPS power. After running for a period of time, the UPS power level is monitored to drop to 30%. At this point, usage information for each hardware component within the server over the past two months is obtained. For example, the average load rate of the central processing unit (CPU), the total number of read / write operations on the non-volatile memory host controller interface specification (NVMe) solid-state drive, and the total execution time of computing tasks on the graphics processing unit (GPU) are obtained.

[0038] In operation S120, at least one target hardware is identified based on various usage information. By analyzing and comparing the acquired usage information of each hardware, one or more hardware devices with relatively low usage are identified and selected as the targets for power reduction operations.

[0039] For example, based on the usage information obtained in operation S110, if the analysis results indicate that the server is handling compute-intensive services with relatively low demand for storage devices, then the Non-Volatile Memory Host Controller Interface (NVMe) solid-state drive, which has relatively low usage, can be identified as the target hardware. In another scenario, if the analysis results indicate that the server's workload has low demand for graphics processing capabilities, then the Graphics Processing Unit (GPU) can be identified as the target hardware.

[0040] In operation S130, power consumption of each target hardware is reduced. For one or more target hardware identified in operation S120, control commands are sent to their corresponding power management units to limit their power consumption. For example, a lower power consumption limit can be set directly, or the hardware can be triggered to enter a preset power-saving operating state, thereby reducing actual power consumption. Specifically, continuing the previous example, after identifying an NVMe solid-state drive as target hardware, a power management command is sent to the controller of the NVMe solid-state drive through the out-of-band management controller (BMC) or operating system-level management tools to adjust its power consumption limit from the default 15 watts to a lower value, such as 7 watts. For another target hardware identified in the aforementioned scenario, a GPU that has been in a low-usage state for a long time, a command can be sent to it to enter a deep idle state, thereby reducing its power consumption to the lowest level.

[0041] Compared to some power limiting strategies, such as indiscriminately reducing CPU frequency under power constraints without considering the actual workload characteristics of the server, this "one-size-fits-all" approach severely impacts the performance of critical services when handling compute-intensive tasks, while the energy-saving effect of reducing CPU power consumption is very limited when handling I / O-intensive tasks. This disclosure, by analyzing the usage information of multiple heterogeneous hardware devices over a historical time range under emergency power conditions, can accurately identify non-critical or low-usage hardware under specific workload modes. Furthermore, this method can selectively and purposefully reduce the power consumption of these identified target hardware devices. For example, it can limit the power consumption of storage devices under compute-intensive loads, while prioritizing the limitation of power consumption of certain CPU cores or GPUs under I / O-intensive loads. This workload-sensitive differentiated power control strategy achieves fine-grained power management, avoids unnecessary damage to the performance of critical tasks, and thus maximizes the execution efficiency of core services while achieving the same energy-saving goals, more effectively extending the effective operating time of devices under limited power conditions.

[0042] Figure 2 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0043] like Figure 2 As shown, based on the aforementioned embodiments, operation S120 may include operation S210.

[0044] In operation S210, hardware whose usage frequency within a preset time range is less than or equal to a preset frequency threshold is selected as target hardware. "Usage frequency" refers to a metric used to quantify the frequency of activity or invocation of a hardware component within a preset time range. This metric can take different forms depending on the hardware type. For example, for storage hardware, it could be the number of read / write operations per unit time; for network hardware, it could be the number of data packets sent / received per unit time; and for computing hardware, it could be the percentage of time it spends in an active state relative to the total time. The "preset frequency threshold" is a benchmark value used to determine whether hardware is in a low-usage state. Based on the foregoing description, different hardware types should correspondingly have different preset frequency thresholds.

[0045] For example, suppose an electronic device includes a central processing unit (CPU), a graphics processing unit (GPU), and a non-volatile memory (NVMe) storage device. The preset frequency threshold set for the NVMe storage device is "average I / O operations per hour less than 1000". After obtaining usage information from the past two months, analysis shows that the NVMe storage device averaged 650 I / O operations per hour over the past two months. Since 650 is less than 1000, the NVMe storage device is identified as the target hardware. Meanwhile, analysis shows that the GPU's activity time accounted for 70% of the time frame, while the set frequency threshold for it is "activity time percentage less than 20%". Since 70% is greater than 20%, the GPU is not identified as the target hardware.

[0046] Some power reduction strategies, such as Dynamic Voltage Frequency Scaling (DVFS) or Operating Average Power Limiting (RAPL), can reduce CPU power consumption, but they do not address the question of "which component should be limited" among the many components of the entire server. This embodiment introduces a comparison between "usage frequency" and "frequency threshold," which makes power adjustment no longer blindly target a preset component (such as the CPU), but can accurately identify the hardware that is truly under low load in specific business scenarios.

[0047] Figure 3 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0048] like Figure 3 As shown, based on the foregoing embodiments, operation S130 may include operation S310.

[0049] In operation S310, in response to the target hardware having corresponding power consumption control parameters, the power consumption of the target hardware is reduced based on these parameters. These power consumption control parameters include at least one of the following: the target hardware's operating frequency and / or operating voltage, the target hardware's I / O bandwidth, and the target hardware's operating mode. Power consumption control parameters refer to a set of adjustable attributes provided by the hardware component, configurable via software or firmware instructions, and directly or indirectly affecting the energy consumption of that hardware component, used to achieve fine-grained management of hardware power consumption.

[0050] In one implementation, the process of reducing the power consumption of the target hardware includes directly adjusting one or more specific operating parameters of the hardware. Specifically, it involves identifying specific power control parameters supported by the target hardware, such as operating frequency or bus speed, and then sending a command to modify the value of that parameter to a lower setting, thereby directly reducing power consumption.

[0051] In another implementation, the process of reducing the power consumption of the target hardware includes setting an overall power consumption limit. Instead of directly interfering with any specific operating parameter of the hardware, a high-level power management interface is used to set a maximum allowable power consumption value for the target hardware, i.e., the power limit. Upon receiving this limit, the hardware's own power management unit automatically and dynamically adjusts one or more of its internal operating parameters (such as operating frequency, voltage, clock gating, etc.) to ensure that its actual power consumption does not exceed the set limit.

[0052] For example, in the previous implementation: if the target hardware is a central processing unit (CPU) with operating frequency and operating voltage as power consumption control parameters, its clock speed can be directly reduced from 3.0 GHz to 1.8 GHz using Dynamic Voltage and Frequency Scaling (DVFS) technology, and its core operating voltage can be reduced accordingly. If the target hardware is a Non-Volatile Memory Host Controller Interface Specification (NVMe) solid-state drive with I / O bandwidth as a power consumption control parameter, the operating mode of its connected bus interface (e.g., PCIe) can be downgraded from a Gen4 x4 channel to a Gen3 x4 channel.

[0053] For example, in a later implementation: if the target hardware is an Intel CPU that supports the Average Power Limit (RAPL) interface, a lower power cap can be written to its RAPL interface, such as adjusting the CPU package's Long Term Power Limit (PL1) from 95 watts to 65 watts. If the target hardware is a memory module (DIMM), a lower power cap can also be set for it via the RAPL interface.

[0054] Figure 4Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0055] like Figure 4 As shown, based on the foregoing embodiments, operation S130 may include operation S410.

[0056] When operating S410, in response to the target hardware lacking corresponding power consumption control parameters, the frequency of use of the target hardware is reduced, or the function of the target hardware is turned off, based on the current operating requirements of the electronic device. "Lacking corresponding power consumption control parameters" means that the target hardware itself does not provide an interface or attribute that can be directly configured via software or firmware for fine-tuning its power consumption; for example, it may lack Dynamic Voltage Frequency Scaling (DVFS) or Operating Average Power Limiting (RAPL) functions. "Current operating requirements of the electronic device" refers to the specific tasks or operating states of the electronic device at the time of making the power consumption adjustment decision, used to determine whether the hardware usage frequency can be reduced or its function turned off.

[0057] In one implementation, it is first determined whether the target hardware belongs to the type without power control parameters. If so, the current operational requirements are further analyzed. If the operational requirements permit, the hardware is indirectly controlled through upper-layer software (such as the operating system, virtual machine manager, or specific applications). One approach is to reduce the number of work requests or data flows sent to the hardware through a scheduler or flow control mechanism, thereby reducing its frequency of invocation and activity level. Another approach is to completely disable or logically shut down the hardware through system instructions when it is determined that the hardware's function is not essential to the current operational requirements, in order to achieve maximum energy savings.

[0058] According to embodiments of this disclosure, before making a power reduction decision, it is assessed whether the current operational requirements meet specific applicable conditions, and it is also checked whether there are any exceptions that must be avoided. Applicable conditions may refer to the current operating state or task of the electronic device, which makes the function of the target hardware temporarily unnecessary or degradeable, and limiting its power consumption will not affect the correct execution of core services.

[0059] For example, the operation of "disabling the function of target hardware" is applicable when the current operational requirements are completely decoupled from the function of the target hardware. A typical example is when the target hardware is a network adapter, and the electronic device is performing a purely local computing task (such as software compilation without network dependencies, local system testing, or offline machine learning model training with the entire dataset downloaded). In this case, the network function is not necessary to complete the core task, and the condition for disabling the hardware is met, and it can be disabled through operating system commands.

[0060] For example, the operation of "reducing the frequency of target hardware usage" is applicable when the current operational requirements place low demands on the performance of the target hardware, or when multiple tasks with distinguishable priorities are using the hardware. For instance, when the target hardware is a RAID adapter, and the system is running a main application with low I / O performance requirements while simultaneously running a low-priority background disk inspection task, the condition for reducing its usage frequency is met. This can be achieved by throttling the background task through the I / O scheduler, thereby indirectly reducing the workload of the RAID adapter.

[0061] Secondly, to ensure system stability and security, the method also includes the identification and handling of exceptional cases. One exceptional case is when a piece of hardware is identified as the target hardware based on historical usage information, but its function is critical to ensuring the integrity, consistency, or manageability of the currently running tasks. In this case, power limiting operations must be prohibited. A key example of an exceptional case is when the target hardware is a RAID adapter. Even if its average I / O load is low, if the system detects that it is performing critical management operations such as RAID array rebuilding or full-disk data consistency verification, the power adjustment method will recognize this state and skip any limits on the adapter to prevent data corruption. Another example of an exceptional case involves network adapters. When an electronic device acts as an edge node and needs to maintain a long-lived connection through the network adapter to receive remote management commands or send heartbeats, even if its data throughput is low, this connection is necessary to maintain system manageability, and therefore the hardware cannot be shut down.

[0062] Conventional power management methods typically rely on the power control interface provided by the hardware itself. For components lacking such interfaces (such as some network adapters or RAID adapters), these methods are ineffective, leaving this portion of power consumption an unmanageable "blind spot." This disclosure extends power management to these previously "uncontrolled" hardware components by introducing an analysis of "current operational requirements" and clearly distinguishing between "applicable conditions" and "exceptions." This avoids improper power consumption limits during critical operations (such as RAID rebuilds), preventing catastrophic consequences such as data loss or system downtime.

[0063] Based on the aforementioned embodiments, a second power source supplies power to the electronic device after the first power source fails, and the power supply of the second power source is less than or equal to a preset power threshold. The execution of the power consumption adjustment method is predicated on a series of monitored power supply state transitions. The power management system of the electronic device continuously monitors its power supply state. Upon detecting a failure or disconnection of the first power source, the power management system performs a power supply switch, with the second power source starting to supply power to the device to maintain operation. After this switch, the electronic device enters a state relying on backup power and continuously monitors the "power supply" of the second power source. "Power supply" is a quantifiable indicator representing remaining energy reserves, such as a percentage of total capacity (state of charge, SoC), remaining watt-hours (Wh), or estimated remaining operating time. Subsequent power consumption adjustment operations (such as S110 to S130) are only triggered when the monitored power supply decreases to a level less than or equal to the "preset power threshold."

[0064] For example, an electronic device is a server deployed in an edge data center. Its primary power source is mains power, and its secondary power source is an uninterruptible power supply (UPS) with a total capacity of 2000 watt-hours (Wh). The preset power threshold is set to 30% of the total capacity (i.e., 600 Wh). When a regional power outage causes a mains power interruption, the server automatically switches to UPS power. After running continuously for a period of time, the UPS management system reports that its remaining power supply is 580 Wh. Since 580 Wh is less than the preset 600 Wh threshold, the trigger condition is determined to be met, and the power consumption adjustment process is initiated.

[0065] In another embodiment, the electronic device is a portable workstation for field data acquisition, with an external solar panel as its primary power source and a built-in lithium battery pack as its secondary power source. A preset power threshold is set to 25% remaining power. When night falls or during consecutive cloudy days, the solar panel stops supplying power, and the workstation switches to power from the built-in battery. The battery management system (BMS) monitors the battery's state of charge (SoC) in real time. When it reports that the remaining power has dropped to 20%, since 20% is less than the preset 25% threshold, the trigger condition is determined to be met, thereby initiating a power consumption adjustment method to ensure that data recording time can be extended as much as possible without external power supplementation.

[0066] According to embodiments of this disclosure, by clearly defining the triggering timing of the power consumption adjustment operation—that is, after the first power source fails and the power supply from the second power source drops to a preset critical threshold—the power consumption optimization strategy is ensured to be activated at the most critical and appropriate moment. This avoids prematurely and unnecessarily sacrificing system performance when backup power is still sufficient, and also prevents the risk of data loss or service interruption due to the sudden depletion of backup power or abnormal system shutdown caused by failure to intervene in a timely manner.

[0067] Figure 5 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0068] like Figure 5 As shown, based on the foregoing embodiments, the information also includes historical power consumption information of the hardware within a preset time range when the electronic device is powered by a second power source. "Historical power consumption information" can refer to the actual power consumption data of each hardware component recorded when the electronic device is independently powered by a second power source (e.g., an uninterruptible power supply UPS).

[0069] In one implementation, this historical power consumption information is acquired in real time. Specifically, from the moment the first power source is interrupted and the electronic devices switch to a second power source, the monitoring module (such as the out-of-band management controller, BMC) begins to collect and record the power consumption data of each hardware component continuously and in real time. For example, when the mains power is interrupted and the server switches to UPS power, the BMC immediately begins to record the power consumption of components such as the CPU and GPU at a frequency of once per second. After running for 30 minutes, the UPS power drops to a threshold. At this point, the analysis focuses on the data generated within those past 30 minutes, representing the actual power consumption performance of each hardware component under actual load during this power outage event.

[0070] In another implementation, this historical power consumption information is acquired in advance. Specifically, electronic devices have operated under a secondary power source during certain time periods in the past (e.g., during planned UPS periodic self-tests, equipment maintenance, or the last unexpected power outage), and power consumption data for each piece of hardware is recorded and stored during these periods. For example, a server's UPS is configured to perform a monthly discharge self-test, during which the server operates powered by the UPS for 10 minutes. The BMC records and archives the power consumption of each piece of hardware during these 10 minutes. When a real mains power outage occurs and the UPS power drops to a threshold, the power consumption data recorded during the most recent self-tests is retrieved.

[0071] Operation S130 may include operation S510.

[0072] When operating the S510, power consumption of target hardware is reduced based on historical power consumption information. This historical information guides the priority and magnitude of power reduction, prioritizing target hardware with higher power consumption in backup power mode to achieve maximum energy efficiency.

[0073] The source of historical power consumption information varies, leading to different methods for reducing the power consumption of target hardware. In one implementation (corresponding to real-time acquired historical power consumption information), a previously determined list of target hardware is combined with real-time power consumption data acquired during the power outage. For each target hardware in the list, its actual power consumption during the event is evaluated. Target hardware with higher power consumption values ​​will be assigned a higher priority for power consumption limiting. Continuing with the previous real-time acquisition example, suppose GPU-1 and GPU-2 were both selected as target hardware due to their consistently low usage. However, the power consumption log for the 30-minute power outage shows that GPU-1 is processing a burst task with an average power consumption of 70W, while GPU-2 has been idle with a power consumption of 15W. Based on this real-time information, it is determined that limiting GPU-1 at this moment would severely impact business operations. Therefore, GPU-2, which is in a high idle power consumption state, will be prioritized and placed into a deep sleep state, achieving a precise response to the current situation.

[0074] In another implementation (corresponding to pre-acquired historical power consumption information), historical power consumption profiles are used to perform power limiting. When power adjustment is triggered, the pre-acquired power consumption baseline data is used to evaluate the target hardware. Continuing with the aforementioned pre-acquired example, when a real power outage occurs, the GPU-2 and NVMe SSD are identified as the target hardware. Historical self-test data is queried, revealing that the idle power consumption of the GPU-2 in UPS mode (15W) is significantly higher than the average power consumption of the NVMe SSD (8W). Based on this known and stable power consumption characteristic, the GPU-2 can be prioritized for limiting without waiting for a re-evaluation of power consumption during the current power outage, thus accelerating the response time.

[0075] Methods that rely solely on long-term usage frequency or activity levels to select target hardware may fail to accurately identify the actual power consumption "giants" during a specific period of emergency power supply. A piece of hardware may be inactive for extended periods, but its inherent idle power consumption may still be considerable during backup power supply. This disclosure, by introducing and utilizing "historical power consumption information during backup power supply," moves power adjustment from inference based solely on historical load patterns to real power consumption data measured in current or historical emergency states. This allows for more precise identification and suppression of components that consume the most backup power, thereby achieving more significant energy savings under the same conditions.

[0076] Figure 6 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0077] like Figure 6 As shown, based on the aforementioned embodiments, operation S130 may include operations S610 to S620.

[0078] When operating the S610, the power consumption reduction target for the target hardware is determined based on the remaining power from the second power source. "Power consumption reduction target" is a quantitative indicator used to guide the intensity of power reduction operations. It can be a specific power reduction value, a percentage, a preset energy-saving level, or a target power consumption limit.

[0079] In one implementation, the determination process is based on a tiered strategy, predefining multiple power ranges and associating each range with a "power reduction level". When power adjustment is triggered, the current remaining power of the second power source is obtained, and its range is determined to identify the appropriate reduction level.

[0080] For example, the server's energy-saving strategy defines three power ranges: 1. When the remaining power is 15%~30%, the power reduction is "medium"; 2. When the remaining power is 5%~15%, the reduction is "deep"; 3. When the remaining power is less than 5%, the reduction is "extreme". When the UPS reports that its remaining power is 12%, it is determined to be in the second range, and therefore the power reduction for this operation is determined to be "deep".

[0081] In another implementation, this determination process is based on continuous function calculations. Using a preset mathematical formula, the remaining battery power is used as an input variable to directly calculate the specific power reduction. For example, for a target CPU, its power reduction (represented as the target power limit) is calculated using the formula: Target Limit = Base Limit * (Current Remaining Battery Percentage / Initial Trigger Threshold Percentage). Assuming the base limit is 65W and the initial trigger threshold is 30%, if the current UPS remaining battery power is 18%, then the target limit is calculated to be 65W * (18% / 30%) = 39W. This 39W is the specific power reduction determined in this operation.

[0082] When operating the S620, the power consumption of the target hardware is reduced according to the degree of power reduction. For example, continuing the example of the aforementioned tiered strategy, when the determined degree is "depth", the strategy library is consulted to find that the "depth" degree corresponds to the following operations: setting the upper limit of the target CPU's operating frequency to 1.5GHz; placing the target GPU in its second lowest power-saving state; and reducing the PCIe link width of the target NVMe hard drive from x4 to x2. Then, corresponding control commands are sent to each piece of hardware to complete these settings.

[0083] In another implementation (corresponding to continuous function calculation), the calculated values ​​are directly used to set the hardware's power consumption parameters. Continuing with the previous example of continuous function calculation, when the calculated CPU target upper limit is 39W, the average power limit interface is used to directly set the CPU's package power limit to 39W. Upon receiving this instruction, the CPU's internal power management unit automatically adjusts its operating frequency and voltage to ensure that its average power consumption does not exceed this newly set upper limit.

[0084] Some power adjustment methods apply a fixed-intensity power limit only once after triggering, and this "one-size-fits-all" approach cannot adapt to continuously changing backup power levels. When backup power is sufficient, this fixed limit may excessively sacrifice performance; while when the power is about to run out, the limiting force may not be sufficient to maximize battery life. This embodiment dynamically correlates the degree of power reduction with the remaining power of the second power source. When the backup power just reaches a threshold, a milder power limit can be applied to preserve more performance; as the power further depletes, the limiting force is automatically increased to prioritize operating time.

[0085] Figure 7 Another flowchart illustrating a power consumption adjustment method according to an embodiment of the present disclosure is shown schematically.

[0086] like Figure 7 As shown, based on the aforementioned embodiments, operation S130 may include operation S710.

[0087] In operation S710, in response to the existence of multiple functionally identical first hardware components among the various target hardware components, the power consumption of at least some of the first hardware components is reduced, or at least some of the first hardware components are turned off. Functionally identical first hardware components can refer to hardware components configured in an electronic device that have equivalent or similar functional specifications, are capable of performing the same type of task, and are interchangeable, such as multiple graphics processing units (GPUs) configured in a server, multiple network interface controllers (NICs), or multiple central processing units (CPUs) in a multiprocessor system.

[0088] In one implementation, the operation includes: first, identifying functionally identical hardware groups belonging to a target hardware set; then, assessing the total load currently running on the hardware group; if the total load can be carried by one or a portion of the hardware in the group, migrating or consolidating all loads to that portion of the hardware; and finally, performing power reduction operations on the remaining hardware that is completely idle after the migration is completed, such as putting it into deep sleep or disconnecting its power supply directly through the power controller.

[0089] For example, the electronic device is an AI server configured with four identical GPUs (GPU-0, GPU-1, GPU-2, GPU-3). These four GPUs were identified as target hardware due to their historically low utilization. When the power adjustment method is triggered, it is detected that these four GPUs belong to the "first hardware with the same function". Further analysis reveals that only a small number of lightweight inference tasks are currently running on GPU-1 and GPU-2, while the total load can be handled by a single GPU. Therefore, the following steps are performed: all tasks on GPU-1 and GPU-2 are migrated to GPU-0; after confirming that GPU-1, GPU-2, and GPU-3 have no active tasks, instructions are sent to these three GPUs through the corresponding management interface to put them into a deep sleep state, or, if the motherboard supports it, their power supply slots are directly shut down. Through this operation, three GPUs are "shut down", leaving only one GPU active, thereby saving the static standby power consumption of the three GPUs.

[0090] For example, consider a server equipped with two 10GbE network interface controllers (NIC-0 and NIC-1). These two NICs are configured in link aggregation mode to provide redundancy and high bandwidth. In emergency power mode, these two NICs are identified as the target hardware, and they are considered "the first hardware with identical functionality." Analysis reveals that the current actual network traffic is far lower than the processing capacity of a single NIC. Instead of directly shutting down one NIC (as this could trigger network protocol renegotiation, leading to a brief interruption), the approach is to "reduce the power consumption of at least some hardware." Specifically, the logical relationship of the links can be maintained, but one NIC (e.g., NIC-1) can be placed in a low-power standby state via the driver, with all network traffic preferentially transmitted through NIC-0. NIC-1 will only be activated if NIC-0 fails or there is a sudden surge in traffic.

[0091] According to embodiments of this disclosure, in systems equipped with multiple parallel or redundant hardware, conventional power management methods may independently limit the power consumption of each hardware component. However, even in low-power modes, each powered-on hardware component still generates a significant static power consumption. This accumulated static power consumption represents a huge waste of energy in emergency power supply scenarios. By identifying these functionally identical hardware groups and implementing a load consolidation strategy, embodiments of this disclosure can concentrate tasks from multiple hardware units onto a few or even a single unit. Then, redundant hardware that is no longer performing tasks can be completely shut down or placed in a near-zero-power deep sleep state, achieving a more substantial and thorough energy-saving effect.

[0092] Figure 8 A block diagram of a power consumption adjustment device according to an embodiment of the present disclosure is shown schematically.

[0093] like Figure 8 As shown, the power consumption adjustment device 800 may include a first acquisition module 810, a first determination module 820, and a first reduction module 830.

[0094] The first acquisition module 810 is used to acquire usage information of multiple hardware components in the electronic device in response to the first power source of the electronic device stopping power supply and the power of the second power source of the electronic device being less than or equal to a preset power threshold; the usage information at least characterizes the degree of use of the hardware within a preset time range. In some embodiments, the first acquisition module 810 can be used to perform operation S110 in the power consumption adjustment method described above, which will not be elaborated here.

[0095] The first determining module 820 is used to determine at least one target hardware based on various usage information. In some embodiments, the first determining module 820 may be used to perform operation S120 in the power consumption adjustment method described above, which will not be elaborated here.

[0096] The first reduction module 830 is used to reduce the power consumption of each target hardware component. In some embodiments, the first reduction module 830 may be used to perform operation S130 in the power consumption adjustment method described above, which will not be elaborated here.

[0097] According to embodiments of this disclosure, the first determining module may include a second determining module.

[0098] The second determining module is used to select hardware that is used at a frequency less than or equal to a preset frequency threshold within a preset time range as target hardware. In some embodiments, the second determining module may be used to perform operation S210 in the power consumption adjustment method described above, which will not be elaborated here.

[0099] According to embodiments of this disclosure, the first reduction module may include a second reduction module.

[0100] The second power reduction module is used to reduce the power consumption of the target hardware in response to the target hardware having corresponding power consumption control parameters, based on the power consumption control parameters. The power consumption control parameters include at least one of the target hardware's operating frequency and / or operating voltage, the target hardware's I / O bandwidth, and the target hardware's operating mode. In some embodiments, the second power reduction module can be used to perform operation S310 in the power consumption adjustment method described above, which will not be elaborated upon here.

[0101] According to embodiments of this disclosure, the first reduction module may include a third reduction module.

[0102] The third reduction module is used in response to the target hardware not having corresponding power consumption control parameters, based on the current operating requirements of the electronic device, to reduce the frequency of use of the target hardware, or to disable the function of the target hardware. In some embodiments, the third reduction module can be used to perform operation S410 in the power consumption adjustment method described above, which will not be elaborated here.

[0103] According to embodiments of this disclosure, the information used also includes historical power consumption information of the hardware within a preset time range when the electronic device is powered by a second power source; the first reduction module may include a fourth reduction module.

[0104] The fourth reduction module is used to reduce the power consumption of the target hardware based on historical power consumption information. In some embodiments, the fourth reduction module can be used to perform operation S510 in the power consumption adjustment method described above, which will not be elaborated here.

[0105] According to embodiments of this disclosure, the first reduction module may include a second determining module and a fifth reduction module.

[0106] The second determining module is used to determine the power consumption reduction rate of the target hardware based on the remaining power of the second power source. In some embodiments, the second determining module may be used to perform operation S610 in the power consumption adjustment method described above, which will not be elaborated here.

[0107] The fifth reduction module is used to reduce the power consumption of the target hardware according to the power consumption reduction amount. In some embodiments, the fifth reduction module can be used to perform operation S620 in the power consumption adjustment method described above, which will not be elaborated here.

[0108] According to an embodiment of this disclosure, the first reduction module may include a sixth reduction module.

[0109] The sixth reduction module is used to reduce the power consumption of at least some of the first hardware components among the multiple first hardware components with the same function, or to turn off at least some of the first hardware components, in response to the existence of multiple first hardware components with the same function among the target hardware components. In some embodiments, the sixth reduction module can be used to perform operation S710 in the power consumption adjustment method described above, which will not be elaborated here.

[0110] Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as hardware circuitry, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a System-on-Chip, a System-on-a-Substrate, a System-on-Package, an Application-Specific Integrated Circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.

[0111] For example, any plurality of the first acquisition module 810, the first determination module 820, and the first reduction module 830 can be combined into one module / unit / subunit, or any one of these modules / units / subunits can be split into multiple modules / units / subunits. Alternatively, at least part of the functionality of one or more of these modules / units / subunits can be combined with at least part of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this disclosure, at least one of the first acquisition module 810, the first determination module 820, and the first reduction module 830 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in a suitable combination of any of these. Alternatively, at least one of the first acquisition module 810, the first determination module 820, and the first reduction module 830 may be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0112] It should be noted that the data processing system part in the embodiments of this disclosure corresponds to the data processing method part in the embodiments of this disclosure. The specific description of the data processing system part is referred to in the data processing method part, and will not be repeated here.

[0113] Figure 9A block diagram of an electronic device suitable for implementing the methods described above, according to embodiments of the present disclosure, is illustrated schematically. Figure 9 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0114] like Figure 9 As shown, an electronic device 900 according to an embodiment of the present disclosure includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage portion 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0115] RAM 903 stores various programs and data required for the operation of electronic device 900. Processor 901, ROM 902, and RAM 903 are interconnected via bus 904. Processor 901 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 902 and / or RAM 903. It should be noted that the programs may also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0116] According to embodiments of this disclosure, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to a bus 904. The electronic device 900 may also include one or more of the following components connected to the input / output (I / O) interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 910 as needed so that computer programs read from it can be installed into the storage section 908 as needed.

[0117] According to embodiments of this disclosure, the method flow according to embodiments of this disclosure can be implemented as a computer software program. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 909, and / or installed from removable medium 911. When the computer program is executed by processor 901, it performs the functions defined in the system of embodiments of this disclosure. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0118] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0119] According to embodiments of this disclosure, the computer-readable storage medium can be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0120] For example, according to embodiments of this disclosure, a computer-readable storage medium may include the ROM 902 and / or RAM 903 described above and / or one or more memories other than ROM 902 and RAM 903.

[0121] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this disclosure. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the power consumption adjustment methods provided in the embodiments of this disclosure.

[0122] When the computer program is executed by the processor 901, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0123] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices or magnetic storage devices. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via communication section 909, and / or installed from removable medium 911. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof. According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0124] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features described in the various embodiments of the present disclosure can be combined and / or combined in various ways, even if such combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments of this disclosure may be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0125] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1.A method for adjusting power consumption, comprising: in response to a first power supply source of an electronic device stopping supplying power and a second power supply source of the electronic device having an amount of power less than or equal to a preset amount of power threshold, obtaining usage information of a plurality of hardware in the electronic device; the usage information at least representing a usage degree of the hardware in a preset time range; determining at least one target hardware according to each of the usage information; reducing power consumption of each of the target hardware. 2.The method of claim 1, wherein the determining at least one target hardware comprises: regarding the hardware whose usage frequency in the preset time range is less than or equal to a preset frequency threshold as the target hardware. 3.The method of claim 1, wherein the reducing power consumption of each of the target hardware comprises: in response to the target hardware having a corresponding power consumption control parameter, reducing power consumption of the target hardware based on the power consumption control parameter; wherein the power consumption control parameter comprises at least one of a working frequency and / or a working voltage of the target hardware, an I / O bandwidth of the target hardware, and a working mode of the target hardware. 4.The method of claim 3, wherein the reducing power consumption of each of the target hardware comprises: in response to the target hardware not having a corresponding power consumption control parameter, reducing a frequency of usage of the target hardware or shutting down a function of the target hardware based on a current operation requirement of the electronic device. 5.The method of claim 1, wherein the second power supply source supplies power to the electronic device after the first power supply source stops supplying power, and the second power supply source has an amount of power supply less than or equal to a preset amount of power threshold. 6.The method of claim 1, wherein the usage information further comprises historical power consumption information of the hardware in a preset time range when the electronic device is powered by the second power supply source, and the reducing power consumption of each of the target hardware comprises: reducing power consumption of the target hardware according to the historical power consumption information. 7.The method of claim 1, wherein the reducing power consumption of each of the target hardware comprises: determining a power consumption reduction amplitude of the target hardware according to a remaining amount of power of the second power supply source; reducing power consumption of the target hardware according to the power consumption reduction amplitude. 8.The method of claim 1, wherein the reducing power consumption of each of the target hardware comprises: in response to a plurality of first hardware having the same function existing in each of the target hardware, reducing power consumption of at least part of the first hardware or shutting down at least part of the first hardware. 9.An apparatus for adjusting power consumption, comprising: a first obtaining module configured to, in response to a first power supply source of an electronic device stopping supplying power and a second power supply source of the electronic device having an amount of power less than or equal to a preset amount of power threshold, obtain usage information of a plurality of hardware in the electronic device, the usage information at least representing a usage degree of the hardware in a preset time range; a first determining module configured to determine at least one target hardware according to each of the usage information; and a first reducing module configured to reduce power consumption of each of the target hardware. ​ 10. An electronic device comprising: at least one processor; and a memory connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: in response to a first power supply source of the electronic device stopping power supply, and a second power supply source of the electronic device having an amount of power less than or equal to a preset amount of power threshold, acquire usage information of a plurality of hardware in the electronic device; the usage information at least representing a usage degree of the hardware within a preset time range; determine at least one target hardware according to each of the usage information; and reduce power consumption of each of the target hardware.

Images