CN115712333A - Power management method and device - Google Patents
Power management method and device Download PDFInfo
- Publication number
- CN115712333A CN115712333A CN202211468146.XA CN202211468146A CN115712333A CN 115712333 A CN115712333 A CN 115712333A CN 202211468146 A CN202211468146 A CN 202211468146A CN 115712333 A CN115712333 A CN 115712333A
- Authority
- CN
- China
- Prior art keywords
- power
- server
- residual
- power supply
- functional node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Power Sources (AREA)
Abstract
The present specification provides a power management method and apparatus, and relates to the field of communication technology. A method of power management, comprising: determining the residual power of the server according to the total power provided by a power supply in the server and the occupied power of the current component, wherein the current residual power is the difference value of the total power and the occupied power of the power supply; if the functional node is detected to be inserted into the server, acquiring the required power of the functional node; if the required power is determined to be not less than the residual power, switching at least part of standby power in the server into a power supply according to a preset power supply strategy, and recalculating the residual power until the residual power is greater than the required power; and starting the functional node inserted into the server. By the method, the reliability of the server can be improved.
Description
Technical Field
The present disclosure relates to the field of communications technologies, and in particular, to a power management method and apparatus.
Background
With the development of network technology, the demand of servers carrying data and applications is gradually increased, and the requirements for the management and maintenance of the servers are also higher and higher.
In the knife box server, the OM (on board Management) can implement power Management for other components in the server (including the web board, the fan, the display, each node, the power supply, and the like). When the hot plug of the components is carried out, the OM judges according to the current residual power of the knife box server and the required power of the inserted components, if the required power of the inserted components is larger than the current residual power, the OM refuses to power on the inserted components or forces the inserted components to power off, so that the inserted components which are in urgent need of power on cannot be powered on and started quickly, the functions of the server cannot be realized, even the server fails, and the reliability of the server is reduced.
Disclosure of Invention
To overcome the problems in the related art, the present specification provides a power management method and apparatus.
In combination with the first aspect of the embodiments of the present specification, the present application provides a power management method, including:
determining the residual power of the server according to the total power provided by a power supply in the server and the occupied power of the current component, wherein the current residual power is the difference value of the total power and the occupied power of the power supply;
if the functional node is detected to be inserted into the server, acquiring the required power of the functional node;
if the required power is determined to be not less than the residual power, switching at least part of standby power in the server into a power supply according to a preset power supply strategy, and recalculating the residual power until the residual power is greater than the required power;
and starting the functional node inserted into the server.
Optionally, the power supply strategy at least comprises N + N, N +1 and no redundancy.
Optionally, after at least part of the standby power in the server is switched to the power supply, the method further includes:
if the power supply strategy is set to be non-redundant and the required power is greater than the residual power, the distributed power of the powered functional nodes in the server is sequentially recovered from low to high according to the priority of the components, and the residual power is recalculated until the residual power is greater than the required power.
Optionally, after determining that the required power is not less than the remaining power, the method further includes:
and reporting alarm information and recording an abnormal log.
In combination with the second aspect of the embodiments of the present specification, the present application provides a power management apparatus, including:
the operation unit is used for determining the residual power of the server according to the total power provided by the power supply in the server and the occupied power of the current component, wherein the current residual power is the difference value of the total power and the occupied power of the power supply;
the acquisition unit is used for acquiring the required power of the functional node if the functional node is detected to be inserted into the server;
the switching unit is used for switching at least part of standby power supplies in the server into power supply supplies according to a preset power supply strategy if the required power is determined to be not less than the residual power, and recalculating the residual power until the residual power is greater than the required power;
and the starting unit is used for starting the functional node inserted into the server.
Optionally, the power supply strategy at least comprises N + N, N +1 and no redundancy.
Optionally, the apparatus further includes:
and the recovery unit is used for sequentially recovering the distributed power of the electrified functional nodes in the server from low to high according to the priority of the components if the power supply strategy is set to be non-redundant and the required power is greater than the residual power, and recalculating the residual power until the residual power is greater than the required power.
Optionally, the apparatus further includes:
and the alarm unit is used for reporting alarm information and recording an abnormal log.
In combination with a third aspect of embodiments herein, there is provided a management device comprising a transceiver, a processor, and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: implementing the method steps of any of the above.
In connection with a fourth aspect of embodiments herein, there is provided a machine-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, cause the processor to: implementing the method steps of any of the above.
The technical scheme provided by the implementation mode of the specification can have the following beneficial effects:
in the embodiment of the specification, a power supply strategy is preset in the management device of the server, and when the required power of the newly inserted functional node is greater than the residual power of the server, the power supply strategy of the server is switched to enable the standby power supply to be switched to the power supply to provide larger residual power, so that the power-on requirement of the newly inserted functional node is met, the power-on time of the new functional node is shortened, and the reliability of the server is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the specification.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present specification and together with the description, serve to explain the principles of the specification.
FIG. 1 is a flow chart of a method of power management to which the present application relates;
fig. 2 is a schematic structural diagram of a server according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a power management device according to the present application;
fig. 4 is a schematic structural diagram of a management device according to the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present specification.
The present application provides a power management method, as shown in fig. 1, including:
s100, determining the residual power of the server according to the total power of the power supply provided by the power supply in the server and the occupied power of the current component.
As shown in fig. 2, the server may include an OM, a plurality of functional nodes, a power supply, a fan, and other components, where Management devices such as a BMC (Baseboard Management Controller) and a BIOS (Basic Input Output System) chip are disposed on the functional nodes, the BMC is configured to detect states of the devices on the functional nodes and control operations such as powering on and powering off of the functional nodes, and the BIOS chip is configured to implement a self-check and a start process of the functional nodes. The functional nodes can be computing nodes, storage nodes, interconnection nodes and the like plugged into the server. And the interconnection node is used for realizing data interaction between the storage node and the computing node in the server and interaction between the computing node and external equipment (such as a switch, other servers and the like).
In a server, total power = the number of power supplies × the power of a single power supply × X%, where X denotes a redundancy coefficient, and is set to range from 0 to 100, which may be set according to actual requirements, for example, may be set to 90%, that is, 10% of power is reserved.
The occupied power of the current component = fixed power + power-on power;
the fixed power = OM maximum power + fan maximum power + functional node BMC power + basic IO power;
the power-on power = processor + display card + memory + single-board base power.
Based on the formula, the numbers of the OM, the functional nodes, the fans and the power supply are not unique, and calculation can be carried out according to actual deployment.
Because the server has a power supply strategy, part of the power supplies can be in a power supply state and part of the power supplies can be in a backup state according to the power supply strategy. For example, the power supply strategy at least comprises N + N, N +1 and no redundancy. The form of N +2, N +3 … N + (N-1) and the like can be set according to actual needs, and the form is not limited.
The OM can determine the number of power supplies in the power supply state and the power that can be provided by collecting information of the power supplies, thereby determining the total power of the power supply.
When the server is in a working state, the OM performs interaction of heartbeat messages with the functional nodes in the server. Through the interaction heartbeat message, on one hand, whether the functional node has a fault can be determined, on the other hand, the heartbeat message can carry the current power of the functional node, and the OM can analyze the power-on power of one functional node from the load by receiving the heartbeat message. And then, the occupied power of the current component can be calculated through the fixed power and the power-on power of the plurality of functional nodes. Then, OM can determine the remaining power of the server, and the current remaining power is the difference between the total power of the power supply and the occupied power.
S101, if the functional node is detected to be inserted into the server, the required power of the functional node is obtained.
In the server, as shown in fig. 2, a slot into which a functional node is not inserted may be included. In order to improve the working efficiency of the server, a new functional node can be inserted into the slot.
After the functional node is inserted into the slot, the BMC of the functional node is powered on to manage the power-on process of the functional node. And activating a BIOS chip through the BMC to perform self-checking, collecting the power of each component of the functional node, and summarizing to form the required power of the functional node.
And after the OM determines the required power of the functional node and the residual power of the server, determining subsequent execution according to the magnitude relation of the required power and the residual power of the server. If the required power is less than the residual power, the OM can directly issue a power-on instruction to the BMC, so that the BMC controls the functional node to be powered on. If the required power is not less than the remaining power, it indicates that the newly inserted functional node may not be powered up normally, and step S102 is executed.
And S102, if the required power is determined to be not less than the residual power, switching at least part of standby power in the server into a power supply according to a preset power supply strategy, and recalculating the residual power until the residual power is greater than the required power.
Since a plurality of power supply strategies can be configured at the server, and the power supply strategies can be divided into certain levels, namely, the power supply strategies are arranged in descending order from the top to the bottom according to the number of the adopted standby power supplies, for example, when the power supply strategies comprise N + N, N +1 and no redundancy, the number of the adopted standby power supplies which can be N + N is N, the level is the highest, and the level of the non-redundancy is the lowest after N + 1. When the server works, the power supply strategy is selected from high to low according to the level, namely, the power supply strategy is preferentially set to be N + N.
When the required power is determined to be not less than the residual power, the OM may determine that the server cannot power up the newly inserted functional node. At this time, because the standby power supply still exists in the server, the OM may change the power supply policy currently selected by the server, and select the N +1 power supply policy of the first-level priority, that is, switch more standby power supplies to the power supply, so as to provide more power.
And after the OM switches the standby power supply to the power supply, the OM detects the increase of the power supply and recalculates the residual power.
At this time, if OM determines that the remaining power is greater than the required power, step S103 may be skipped to, and the functional node is started to complete the power-on process. If the OM determines that the remaining power is still not greater than the required power, the OM continuously adjusts the power supply strategy under the condition that the standby power still exists in the server, and switches more standby power supplies into the power supply so as to increase the amount of the remaining power to meet the requirement of the newly inserted functional node.
S103, starting the functional node inserted into the server.
By adjusting the power supply strategy, the standby power supply is switched to the power supply to increase the residual power of the server, so that the situation that the residual power cannot meet the required power at the time point when the functional node is inserted into the server can be avoided, the newly inserted functional node is directly refused to be powered on, the power control of the server is more flexible, and the reliability of the server is improved.
Optionally, after the step S102, switching at least part of the standby power supply in the server to the power supply, the method further includes:
and S104, if the power supply strategy is set to be non-redundant and the required power is greater than the residual power, sequentially recovering the distributed power of the electrified functional nodes in the server from low to high according to the priority of the components, and recalculating the residual power until the residual power is greater than the required power.
If the power supply strategy of the server is already adjusted to the lowest level (i.e. no redundancy), it indicates that there is no backup power source that can be switched to the power supply source in the server, and the remaining power in the server cannot be continuously increased. If the residual power of the server still cannot meet the required power of the functional node, the power-on process of the functional node can be terminated only, and the operation of a worker is waited.
Because the newly inserted functional node needs to complete the power-on process, when the power supply and the standby power supply cannot meet the requirements, the OM can recover partial power from the currently running component to meet the power requirement of the newly inserted functional node.
The priority of the components may be configured in the OM in advance, or may be configured by the system guiding the staff after the server operates.
For example, in the following, taking an interconnection node included in a server as an example, an OM may configure a priority order according to a serial number of the interconnection node, and take 6 interconnection nodes as an example, configure slot numbers 1 to 3 as a high priority, and slot numbers 4 to 6 as a low priority, where the reason for this is that the interconnection node is set as a master, that is, 1 to 3 are master, and 4 to 6 are slave, and therefore, the OM may recover power of a part of the slave interconnection nodes.
In another way, taking the compute node and the storage node as examples, the priority order may also be set, and taking 16 nodes as examples, slot numbers 1-8 may be set as high priority, and slot numbers 9-16 may be set as low priority.
Of course, the setting of the component priority is not limited to this, and other components may be set for power recovery.
Optionally, after determining that the required power is not less than the remaining power in step S102, the method further includes:
and S105, reporting the alarm information and recording an abnormal log.
Since there is still a certain abnormality in practice by switching the standby power supply to the power supply, the OM still notifies the worker to perform the repair in response to the above-mentioned warning message.
In addition, by recording the abnormal log, the worker can inquire the time and reason for discovering the abnormality.
Correspondingly, the present application provides a power management apparatus, as shown in fig. 3, including:
the operation unit is used for determining the residual power of the server according to the total power of the power supply provided by the power supply in the server and the occupied power of the current component, wherein the current residual power is the difference value of the total power of the power supply and the occupied power;
the acquisition unit is used for acquiring the required power of the functional node if the functional node is detected to be inserted into the server;
the switching unit is used for switching at least part of standby power supplies in the server into a power supply according to a preset power supply strategy if the required power is determined to be not less than the residual power, and recalculating the residual power until the residual power is greater than the required power;
and the starting unit is used for starting the functional node inserted into the server.
Optionally, the power supply strategy at least comprises N + N, N +1 and no redundancy.
Optionally, the apparatus further includes:
and the recovery unit is used for sequentially recovering the distributed power of the electrified functional nodes in the server from low to high according to the priority of the components if the power supply strategy is set to be non-redundant and the required power is greater than the residual power, and recalculating the residual power until the residual power is greater than the required power.
Optionally, the apparatus further includes:
and the alarm unit is used for reporting alarm information and recording an abnormal log.
Correspondingly, the present application provides a management device, as shown in fig. 4, comprising a transceiver, a processor, and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: implementing the method steps of any of the above.
Accordingly, the present application provides a machine-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, cause the processor to: implementing the method steps of any of the above.
The technical scheme provided by the implementation mode of the specification can have the following beneficial effects:
in the embodiment of the specification, a power supply strategy is preset in the management device of the server, and when the required power of the newly inserted functional node is greater than the residual power of the server, the power supply strategy of the server is switched, so that the standby power supply is switched to the power supply to provide larger residual power, and the power-on requirement of the newly inserted functional node is met, thereby reducing the power-on time of the new functional node and improving the reliability of the server.
It will be understood that the present description is not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.
The above description is only for the purpose of illustrating the preferred embodiments of the present disclosure and is not to be construed as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (10)
1. A method of power management, comprising:
determining the residual power of a server according to the total power provided by a power supply in the server and the occupied power of a current component, wherein the current residual power is the difference value between the total power of the power supply and the occupied power;
if the functional node is detected to be inserted into the server, acquiring the required power of the functional node;
if the required power is determined to be not less than the residual power, switching at least part of standby power in the server into a power supply according to a preset power supply strategy, and recalculating the residual power until the residual power is greater than the required power;
and starting the functional node inserted into the server.
2. The method of claim 1, wherein the power strategy comprises at least N + N, N +1 and no redundancy.
3. The method of claim 1, further comprising, after switching at least a portion of the backup power source in the server to a power supply source:
if the power supply strategy is set to be non-redundant and the required power is greater than the residual power, sequentially recovering the distributed power of the powered functional nodes in the server according to the priority of the components from low to high, and recalculating the residual power until the residual power is greater than the required power.
4. The method of claim 1, after determining that the required power is not less than the remaining power, further comprising:
and reporting alarm information and recording an abnormal log.
5. A power management apparatus, comprising:
the operation unit is used for determining the residual power of the server according to the total power provided by a power supply in the server and the occupied power of a current component, wherein the current residual power is the difference value of the total power and the occupied power of the power supply;
the acquisition unit is used for acquiring the required power of the functional node if the functional node is detected to be inserted into the server;
the switching unit is used for switching at least part of standby power supplies in the server into power supply supplies according to a preset power supply strategy if the required power is determined to be not less than the residual power, and recalculating the residual power until the residual power is greater than the required power;
and the starting unit is used for starting the functional node inserted into the server.
6. The apparatus of claim 5, wherein the power policy comprises at least N + N, N +1 and no redundancy.
7. The apparatus of claim 5, further comprising:
and the recovery unit is used for sequentially recovering the distributed power of the powered functional nodes in the server from low to high according to the component priority if the power supply strategy is set to be non-redundant and the required power is greater than the residual power, and recalculating the residual power until the residual power is greater than the required power.
8. The apparatus of claim 5, further comprising:
and the alarm unit is used for reporting alarm information and recording an abnormal log.
9. A management device comprising a transceiver, a processor, and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: -carrying out the method steps of any one of claims 1 to 4.
10. A machine-readable storage medium having stored thereon machine-executable instructions that, when invoked and executed by a processor, cause the processor to: -carrying out the method steps of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211468146.XA CN115712333A (en) | 2022-11-22 | 2022-11-22 | Power management method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211468146.XA CN115712333A (en) | 2022-11-22 | 2022-11-22 | Power management method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115712333A true CN115712333A (en) | 2023-02-24 |
Family
ID=85234186
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211468146.XA Pending CN115712333A (en) | 2022-11-22 | 2022-11-22 | Power management method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115712333A (en) |
-
2022
- 2022-11-22 CN CN202211468146.XA patent/CN115712333A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5575829B2 (en) | Rack server system and control method thereof | |
| CN100342333C (en) | System management controller negotiation protocol | |
| US7831860B2 (en) | System and method for testing redundancy and hot-swapping capability of a redundant power supply | |
| CN103324495A (en) | Method and system for data center server boot management | |
| CN103197748A (en) | Server cabinet system and power management method thereof | |
| CN101996111A (en) | Switching method, device and distributed blade server system | |
| CN103135732B (en) | Server cabinet system | |
| EP2835716A1 (en) | Information processing device and virtual machine control method | |
| US20230229223A1 (en) | Method to allow for higher usable power capacity in a redundant power configuration | |
| CN111984471B (en) | Cabinet power BMC redundancy management system and method | |
| US20130061082A1 (en) | Balancing power consumption and high availability in an information technology system | |
| JP2009005544A (en) | Control method and control circuit of storage battery system | |
| JP4401973B2 (en) | Information processing apparatus, failure monitoring method, failure monitoring program | |
| WO2016082433A1 (en) | Mainframe power management method, apparatus and mainframe system | |
| CN102103712A (en) | Power supply configuration method and device | |
| JP2008015803A (en) | Blade type computer management system | |
| CN113741668B (en) | Power supply assembly for delivering power at multiple voltages | |
| JP5626884B2 (en) | Power supply management system and power supply management method | |
| CN115712333A (en) | Power management method and device | |
| JP2005250840A (en) | Information processing apparatus for fault-tolerant system | |
| CN116641907A (en) | Server fan control method, system, terminal and storage medium | |
| CN103718536A (en) | System and method for powering redundant components | |
| CN115225464B (en) | Network switching method, device, equipment and storage medium | |
| CN114124803B (en) | Device management method and device, electronic device and storage medium | |
| CN111078454A (en) | Cloud platform configuration recovery method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |