CN111414070B - Case power consumption management method and system, electronic device and storage medium - Google Patents
Case power consumption management method and system, electronic device and storage medium Download PDFInfo
- Publication number
- CN111414070B CN111414070B CN202010152709.9A CN202010152709A CN111414070B CN 111414070 B CN111414070 B CN 111414070B CN 202010152709 A CN202010152709 A CN 202010152709A CN 111414070 B CN111414070 B CN 111414070B
- Authority
- CN
- China
- Prior art keywords
- power consumption
- node
- value
- current
- predicted
- 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.)
- Active
Links
- 238000007726 management method Methods 0.000 title claims abstract description 43
- 238000004458 analytical method Methods 0.000 claims abstract description 32
- 238000000034 method Methods 0.000 claims abstract description 18
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 13
- 230000035772 mutation Effects 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000011084 recovery Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 229920002492 poly(sulfone) Polymers 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3206—Monitoring of events, devices or parameters that trigger a change in power modality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
- G06F11/3062—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations where the monitored property is the power consumption
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application discloses a case power consumption management method, a case power consumption management system, an electronic device and a readable storage medium, wherein the method comprises the following steps: acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods; analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result; predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value; and adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node. Therefore, the power consumption value of the next time period can be predicted based on the historical data of each node in the historical time period, so that the power consumption value of each node is adjusted and distributed in advance, the power consumption adjustment of each node in the server can be realized more timely and rapidly, the performance of the service mutation node is prevented from being influenced, and the service processing capacity is improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and a system for managing power consumption of a chassis, an electronic device, and a computer-readable storage medium.
Background
With the gradual maturity of services such as cloud computing, AI, unmanned driving, edge computing, assisted accelerated computing and the like, more and more high-power-consumption devices such as general servers, AI servers, GPU accelerator cards and FPGA accelerator cards are rapidly put into use, and the density of a machine room is higher and higher.
Due to the increase of the density of the machine room, the number of nodes is greatly increased, and meanwhile, the rapid increase of power consumption is inevitably brought. The existing machine room power transformation is limited by various external factors and cannot follow up timely, so that certain limitation on the power consumption of equipment is needed when new equipment is replaced in an old machine room. In addition, due to the centralization and clustering of modern computer rooms, the number of server nodes and cluster nodes is rapidly increasing, but the use efficiency is not high before the low-level loitering. Taking the X86 server cluster which is the most popular application as an example, the utilization rate is generally considered to be lower than 30%, while the average utilization rate of the Intel servers is considered to be only 10%, which causes waste of a large amount of power resources.
In a conventional power consumption management scheme, for equipment in a whole chassis, an RMC/CMC management module is provided on the whole chassis, total power consumption of the whole chassis is set through the RMC/CMC management module, and then the total power consumption is evenly distributed to each node. The scheme has the advantages that a customer does not need to be connected with each node to be respectively arranged, only the total power consumption value of the whole chassis needs to be arranged for the RMC/CMC, and the maintenance workload can be reduced. However, under this scheme, there is a high possibility that there exists a node with heavy traffic, and the processing speed and efficiency of the node will be limited by the evenly distributed power consumption; for nodes with idle services, power consumption is wasted, so that power consumption is idle and cannot be fully utilized. For example, the total power consumption of a certain chassis can be allocated to be 2000W, the chassis consists of 5 nodes, A, B nodes are busy in traffic, the down-conversion processing is caused due to the power consumption limit of 400W, and the actual power consumption is 200W when C, D, E three nodes are generally in traffic. Thus, the power consumption of the whole chassis is 1400W, 600W is wasted, and the traffic of the A, B node is limited.
In another conventional power consumption management scheme, dynamic adjustment and dynamic allocation of the power of the whole power system are performed according to the current actual power consumption of each node, and although dynamic allocation can be performed well, there is a certain delay and hysteresis in the allocation adjustment, especially when the traffic of a certain node is suddenly increased, the required power may exceed the originally allocated power, and the traffic may receive a certain influence during the reallocation adjustment.
Therefore, how to solve the above problems is a great concern for those skilled in the art.
Disclosure of Invention
The present application aims to provide a chassis power consumption management method, a chassis power consumption management system, an electronic device, and a computer-readable storage medium, which can adjust and allocate power consumption values of nodes in advance, avoid affecting the performance of service mutation nodes, and improve service processing capability.
In order to achieve the above object, the present application provides a chassis power consumption management method, including:
acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods;
analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result;
predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value;
and adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node.
Optionally, the analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result includes:
generating a power consumption historical time curve corresponding to each node according to all the power consumption historical data;
and training by using a time sequence prediction algorithm based on the power consumption historical time curve to obtain a power consumption prediction model.
Optionally, the predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value includes:
and taking the power consumption value corresponding to each node in the current time period and the time of the next time period as the input of the power consumption prediction model to obtain the predicted power consumption value corresponding to each node output by the power consumption prediction model.
Optionally, after the power consumption of each node is adjusted or allocated according to the predicted power consumption value, the current total power consumption of the chassis, and the current power consumption allocation value of the node, the method further includes:
after the next time period is finished, acquiring the actual power consumption value of each node;
and correcting and adjusting the probability distribution weight in the power consumption prediction model by combining the actual power consumption value and the predicted power consumption value.
Optionally, the adjusting or allocating the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis, and the current power consumption allocation value of the node includes:
if the predicted power consumption value is higher than the current power consumption distribution value, distributing power consumption for the nodes in advance based on the predicted power consumption value;
if the predicted power consumption value is lower than the current power consumption distribution value, determining the power consumption utilization rate of the current node;
and if the power consumption utilization rate of the current node is lower than a first threshold value, recovering the power consumption distributed to the current node in advance.
Optionally, the method further includes:
if the predicted power consumption values corresponding to all the nodes are higher than the corresponding current power consumption distribution values, judging whether the total power consumption of the current chassis is larger than a first power consumption increase total amount or not;
if yes, distributing required power consumption for each node directly according to the predicted power consumption value;
and if not, distributing corresponding power consumption for each node according to a preset distribution proportion.
Optionally, the method further includes:
if the predicted power consumption value corresponding to part of the nodes is higher than the current power consumption distribution value and the predicted power consumption value corresponding to part of the nodes is lower than the current power consumption distribution value, judging whether the total power consumption of the current chassis is larger than a second power consumption increase total amount or not;
if the total power consumption of the current chassis is larger than or equal to the second power consumption increase total amount, directly distributing power consumption for each node according to the predicted power consumption value;
if the total power consumption of the current chassis is smaller than a second power consumption increase total amount, determining a target node of which the predicted power consumption value is lower than the current power consumption distribution value, and determining a power consumption utilization rate corresponding to the target node;
judging whether the power consumption utilization rate corresponding to the target node is greater than a second threshold value;
if not, recovering the pre-distribution power consumption of the target node in advance, and redistributing the nodes of which the predicted power consumption values are higher than the current power consumption distribution values after the power consumption is recovered.
In order to achieve the above object, the present application provides a chassis power consumption management system, including:
the history acquisition module is used for acquiring power consumption history data corresponding to all nodes in the case in a preset number of history time periods;
the historical analysis module is used for analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result;
the power consumption prediction module is used for predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value;
and the adjusting and distributing module is used for adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node.
To achieve the above object, the present application provides an electronic device including:
a memory for storing a computer program;
a processor for implementing the steps of any of the aforementioned disclosed chassis power consumption management methods when executing the computer program.
To achieve the above object, the present application provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the chassis power consumption management methods disclosed in the foregoing.
According to the scheme, the chassis power consumption management method provided by the application comprises the following steps: acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods; analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result; predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value; and adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node. Therefore, the power consumption value of the next time period can be predicted based on the historical data of each node in the historical time period, so that the power consumption value of each node is adjusted and distributed in advance, the power consumption adjustment of each node in the server can be realized more timely and rapidly, the performance of the service mutation node is prevented from being influenced, and the service processing capacity is improved.
The application also discloses a chassis power consumption management system, an electronic device and a computer readable storage medium, which can also achieve the technical effects.
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 application.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a chassis power consumption management method disclosed in an embodiment of the present application;
fig. 2 is a flowchart of another chassis power consumption management method disclosed in the embodiment of the present application;
fig. 3 is a topology diagram of a whole server structure disclosed in the embodiment of the present application;
fig. 4 is a structural diagram of a chassis power consumption management system according to an embodiment of the present application;
fig. 5 is a block diagram of an electronic device disclosed in an embodiment of the present application;
fig. 6 is a block diagram of another electronic device disclosed in the embodiments of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the conventional technology, in a power consumption management scheme, the total power consumption of the whole chassis is set by using an RMC/CMC management module on the whole chassis, and then the total power consumption is evenly distributed to each node. Under the scheme, the processing speed and efficiency of the nodes with heavy traffic are limited by the evenly distributed power consumption; for nodes with idle services, power consumption is wasted, and the nodes cannot be fully utilized. In another power consumption management scheme, dynamic adjustment and dynamic allocation of the power of the whole power system are performed according to the current actual power consumption of each node, and certain delay and hysteresis exist in allocation adjustment, and particularly when the service of a certain node is suddenly increased, the required power may exceed the originally allocated power, and the service may receive certain influence during the period of reallocation adjustment.
Therefore, the embodiment of the application discloses a chassis power consumption management method, which can more timely and rapidly realize power consumption adjustment of each node in a server, avoid influencing the performance of a service mutation node, and improve service processing capacity.
Referring to fig. 1, a flowchart of a chassis power consumption management method disclosed in an embodiment of the present application is shown in fig. 1, and includes:
s101: acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods;
in the embodiment of the application, historical power consumption data of all nodes in a chassis are obtained first. In specific implementation, the power consumption history data corresponding to each node in a preset number of history time periods may be recorded. For example, the average power in a time period of 10 minutes may be calculated starting at point 0, and 144 data records may be recorded each day, and 90 days of historical data may be recorded each node. Therefore, a group of corresponding historical data values can be recorded for each node, and after a new power consumption data value is generated, the historical data is written into the data record, the oldest historical data is deleted, and the data are sequentially and circularly recorded.
S102: analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result;
in this step, the power consumption history data corresponding to each node is analyzed to obtain a corresponding analysis result. Specifically, the process of analyzing all the power consumption history data of each node to obtain a historical power consumption analysis result may include: generating a power consumption historical time curve corresponding to each node according to all the power consumption historical data; and training by using a time sequence prediction algorithm based on the power consumption historical time curve to obtain a power consumption prediction model.
It can be understood that the embodiment of the application can count the power consumption historical time curve of each node, count the power consumption distribution and change curve situation in each time period, fit and analyze the power curve in each historical time point, analyze the power value with the maximum probability of occurrence in the point, correct the power value through the true value, abandon the discrete point and the point with lower probability, and continuously learn and evolve by self.
It should be noted that, because the power consumption cannot be adjusted and calculated too frequently, the embodiment of the present application may perform analysis for a time period of ten minutes, and statistically analyze the probability distribution points and the variation fitting curve of the average power value in the time period.
In specific implementation, a power consumption curve of the node may be counted and analyzed according to historical data of a past period of time, the probability of the power value distribution point at each time point is counted, a time prediction sequence algorithm is further adopted to train a model, for example, a Facebook time sequence prediction algorithm, and finally a power consumption prediction model is obtained.
S103: predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value;
in this step, the power consumption value that may be required by each node in the next time period may be predicted according to the historical power consumption analysis result of the above step.
When the power consumption value of each node in the next time period is predicted based on the historical power consumption analysis result, the power consumption value corresponding to each node in the current time period and the time of the next time period can be used as the input of a power consumption prediction model, the power consumption prediction model is used for predicting to obtain a power consumption value with the maximum probability, and the power consumption value is output as the predicted power consumption value.
S104: and adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node.
Specifically, the power consumption difference value can be determined according to the predicted power consumption value and the current power consumption distribution value of the node, that is, how much power consumption or redundant power consumption needs to be distributed, and then each node is distributed and adjusted according to the specific condition whether the total power consumption of the current chassis meets the power consumption to be distributed or not.
It can be understood that after the power consumption of each node is adjusted or distributed according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node, the actual power consumption value of each node in the time period can be obtained after the next time period is finished, so that the error value can be determined by combining the actual power consumption value and the predicted power consumption value, and the probability distribution weight in the prediction model is corrected and adjusted based on the error value, so as to further improve the accuracy of the prediction model.
According to the scheme, the chassis power consumption management method provided by the application comprises the following steps: acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods; analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result; predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value; and adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node. Therefore, the power consumption value of the next time period can be predicted based on the historical data of each node in the historical time period, so that the power consumption value of each node is adjusted and distributed in advance, the power consumption adjustment of each node in the server can be realized more timely and rapidly, the performance of the service mutation node is prevented from being influenced, and the service processing capacity is improved.
The embodiment of the application discloses a chassis power consumption management method, and compared with the previous embodiment, the embodiment further explains and optimizes the technical scheme. Specifically, the method comprises the following steps:
referring to fig. 2, a flowchart of another chassis power consumption management method provided in the embodiment of the present application is shown in fig. 2, and includes:
s201: acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods;
s202: analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result;
s203: predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value;
s204: if the predicted power consumption value is higher than the current power consumption distribution value of the node, distributing power consumption for the node in advance based on the predicted power consumption value;
s205: if the predicted power consumption value is lower than the current power consumption distribution value, determining the power consumption utilization rate of the current node;
s206: and if the power consumption utilization rate of the current node is lower than a first threshold value, recovering the power consumption distributed to the current node in advance.
In the embodiment of the application, after the predicted power consumption value of the node is obtained, if the predicted power consumption value of the node is higher than the current power consumption distribution value, it indicates that the power consumption may be insufficient in the next time period, so that the power consumption needs to be distributed to the node in advance to avoid affecting the node performance; if the predicted power consumption value is lower than the current power consumption distribution value, the node still has redundant power consumption. At this time, whether the power consumption utilization rate of the node is lower than the first threshold value can be further judged, and if so, the power consumption pre-allocated to the node can be recycled in advance to avoid power consumption waste.
Further, if the predicted power consumption values corresponding to all the nodes are higher than the current power consumption distribution value, accumulating the difference between the predicted values of all the nodes and the current distribution value to obtain a first power consumption increase total amount, and judging whether the total power consumption of the current chassis is greater than the first power consumption increase total amount. If so, representing that the residual total power consumption can meet the requirement of power consumption increase, and directly distributing required power consumption for each node according to the predicted power consumption value; if not, the representation that the residual total power consumption cannot meet the power consumption increase is carried out, and at the moment, corresponding power consumption can be distributed to each node according to a preset distribution proportion. The preset allocation proportion can be determined according to the power consumption increase value of each node.
If the predicted power consumption value corresponding to part of the nodes is higher than the current power consumption distribution value and the predicted power consumption value corresponding to part of the nodes is lower than the current power consumption distribution value, the power consumption increment of the nodes with the predicted power consumption values higher than the current power consumption distribution value can be accumulated to obtain a second power consumption increment total amount, and whether the total power consumption of the current chassis is larger than or equal to the second power consumption increment total amount or not is judged; if yes, power consumption can be directly distributed to each node according to the predicted power consumption value; if not, the power consumption of the node with reduced power consumption can be recycled in advance. Specifically, a target node with a predicted power consumption value lower than the current power consumption distribution value can be selected, and the power consumption utilization rate corresponding to the target node is determined. Judging whether the power consumption utilization rate corresponding to the target node is greater than a second threshold value or not; if not, the pre-distribution power consumption of the target node can be recycled in advance, and the power consumption of the rest nodes is distributed after the pre-distribution power consumption is recycled.
The following describes a chassis power consumption management method provided in the embodiments of the present application by using a specific example. Fig. 3 is a topological schematic diagram of a whole server structure, and as shown in fig. 3, specifically, a user communicates with the RMC/CMC through a network, and can set the total power consumption of the chassis, set whether to open the total power consumption limit, set an inquiry interval, and the like. The RMC/CMC is used for communicating with the BMC of each node, and is used for inquiring the current power consumption value of each node and configuring the distributed power of each node. The BMC of each node communicates with the PSU for inquiring the real-time power consumption of the node, one node may have a plurality of PSUs, and the BMC can perform accumulation processing after inquiry for RMC/CMC inquiry. In addition, a user can also set whether to start the power consumption adjusting function, if the user closes the power consumption adjusting function, the RMC/CMC informs each node BMC of closing the power consumption adjustment, and the BMC configures the system to run at full power; if the user starts the power consumption adjusting function, the RMC/CMC calculates the distributed power of each node according to the current condition of each node, configures the distributed power to each node BMC, and informs each node BMC to start power consumption adjustment. Further, the RMC/CMC may also poll the actual power consumption of each node, and the specific polling time may be set by the user. If the power consumption adjusting function is closed, no polling is needed; and if the power consumption adjusting function is started, the RMC/CMC performs polling operation on the power consumption value of each node at the polling time interval configured last time.
When power consumption is adjusted, a value which can be controlled arbitrarily can be calculated according to a total power consumption value set by a user and a minimum power consumption value of each node. In specific implementation, when the power consumption value of a certain node reaches more than 90% of the distribution value, the RMC/CMC may analyze the power consumption increase rate according to the data collected in the previous two times, and pre-determine the power consumption increment of the subsequent 3 intervals, and if the power consumption pool is sufficient, configure the value to the node, and increase the power value of the node; when the power consumption value of a certain node is reduced to 50% of the distribution value, the descending trend can be analyzed according to the data collected in the previous two times, the power consumption distribution is reduced, and redundant power consumption is recovered to a power consumption pool. When the power consumption pool is exhausted, the nodes with lower utilization rate can be recovered.
In the embodiment of the application, historical power consumption values of all time periods of all nodes can be recorded, a power consumption historical time curve is obtained by counting power consumption distribution and change curve conditions of all nodes in each time period, model training is performed by adopting a time prediction sequence algorithm, the power consumption value of the current time period and the time of the next time period are used as input of a model to perform power consumption prediction, a power value with the maximum probability is obtained, and the power value is used as a predicted value of the next time period. Preferably, after the time period is over, a predicted error value may be calculated according to the actual power value of the time period and the previous predicted value, so as to correct the probability distribution weight in the prediction model, and correct and reduce the next error value.
In particular implementations, the policies for automatic adjustment and allocation of power consumption may include, but are not limited to, the following: and for the condition that the predicted value is smaller than the pre-allocated value, if the power consumption utilization rate of the node is low, carrying out a pre-recovery strategy, and if the utilization rate is high, not carrying out adjustment temporarily. When the predicted value is larger than the pre-allocated value, that is, the service may suddenly rise in the next time period, the predicted value is used to perform power allocation and adjustment in advance to prevent the sudden rise of the service, so that the running speed of the CPU is limited and the service processing performance is affected. When the predicted value and the pre-assigned value are kept unchanged or float less, the adjustment can be temporarily not carried out, so that frequent adjustment is avoided.
For the condition that the predicted values of all nodes are reduced, the predicted values can be kept unchanged temporarily, and the original automatic adjustment mechanism is utilized for recovery; for the condition that all the node predicted values are increased, if the residual power consumption in the power consumption pool is more than or equal to all the increased amounts, directly adjusting according to the predicted values; if the power consumption of the power consumption pool is set to be 0, no adjustment is carried out; if the residual power consumption in the power consumption pool is less than the increased power consumption, the residual nodes can be distributed and adjusted according to the proportion on the basis of preferentially ensuring the power consumption of the heat dissipation nodes and the network nodes.
For the situation that the power consumption of part of nodes is increased and the power consumption of part of nodes is reduced, the nodes can be adjusted and added firstly to avoid influencing services due to uncertainty of prediction. If the surplus of the power consumption pool is larger than or equal to the increased power consumption of the node, the power consumption can be directly distributed to the node according to the predicted value. If the surplus of the power consumption pool is smaller than the increment, the utilization rate of the nodes is predicted to be reduced, if the idle power consumption of the nodes is larger than the power consumption difference value, a recovery mechanism can be started, a part of power consumption is recovered to the power consumption pool, and then the nodes which are predicted to be increased are distributed. And if the distribution according to the pre-measured quantity cannot be met after the recovery, performing power consumption distribution according to the proportion. If the node with reduced prediction is accurately predicted, the corresponding power consumption can be automatically recovered after the power consumption of the node is reduced.
In the following, a chassis power consumption management system provided by an embodiment of the present application is introduced, and a chassis power consumption management system described below and a chassis power consumption management method described above may refer to each other.
Referring to fig. 4, a structure diagram of a chassis power consumption management system according to an embodiment of the present application is shown in fig. 4, and includes:
a history obtaining module 301, configured to obtain power consumption history data corresponding to all nodes in the chassis in a preset number of history time periods;
a history analysis module 302, configured to analyze all the power consumption history data of each node to obtain a history power consumption analysis result;
a power consumption prediction module 303, configured to predict a power consumption value of each node in a next time period based on the historical power consumption analysis result, so as to obtain a predicted power consumption value;
and an adjusting and allocating module 304, configured to adjust or allocate power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis, and the current power consumption allocation value of the node.
For the specific implementation process of the modules 301 to 304, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not repeated here.
The present application further provides an electronic device, referring to fig. 5, a structure diagram of an electronic device provided in an embodiment of the present application, as shown in fig. 5, includes:
a memory 100 for storing a computer program;
the processor 200, when executing the computer program, may implement the steps provided by any of the foregoing embodiments.
Specifically, the memory 100 includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer-readable instructions, and the internal memory provides an environment for the operating system and the computer-readable instructions in the non-volatile storage medium to run. The processor 200 may be a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor or other data Processing chip in some embodiments, and provides computing and controlling capabilities for an electronic device, and when executing the computer program stored in the memory 100, the steps of the chassis power consumption management method provided in any of the foregoing embodiments may be implemented.
On the basis of the above embodiment, as a preferred implementation, referring to fig. 6, the electronic device further includes:
and an input interface 300 connected to the processor 200, for acquiring computer programs, parameters and instructions imported from the outside, and storing the computer programs, parameters and instructions into the memory 100 under the control of the processor 200. The input interface 300 may be connected to an input device for receiving parameters or instructions manually input by a user. The input device may be a touch layer covered on a display screen, or a button, a track ball or a touch pad arranged on a terminal shell, or a keyboard, a touch pad or a mouse, etc.
And a display unit 400 connected to the processor 200 for displaying data processed by the processor 200 and for displaying a visualized user interface. The display unit 400 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch panel, or the like.
And a network port 500 connected to the processor 200 for performing communication connection with each external terminal device. The communication technology adopted by the communication connection can be a wired communication technology or a wireless communication technology, such as a mobile high definition link (MHL) technology, a Universal Serial Bus (USB), a High Definition Multimedia Interface (HDMI), a wireless fidelity (WiFi), a bluetooth communication technology, a low power consumption bluetooth communication technology, an ieee802.11 s-based communication technology, and the like.
While FIG. 6 shows only an electronic device having the assembly 100 and 500, those skilled in the art will appreciate that the configuration shown in FIG. 6 is not intended to be limiting of electronic devices and may include fewer or more components than those shown, or some components may be combined, or a different arrangement of components.
The present application also provides a computer-readable storage medium, which may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk. The storage medium stores thereon a computer program, which when executed by a processor implements the steps of the chassis power consumption management method provided by any of the foregoing embodiments.
According to the method and the device, the power consumption value of the next time period can be predicted based on the historical data of each node in the historical time period, so that the power consumption value of each node is adjusted and distributed in advance, the power consumption adjustment of each node in the server can be realized more timely and rapidly, the performance of a service mutation node is prevented from being influenced, and the service processing capacity is improved.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (9)
1. A chassis power consumption management method is characterized by comprising the following steps:
acquiring power consumption historical data corresponding to all nodes in a case in a preset number of historical time periods;
analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result;
predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value;
adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node;
if the predicted power consumption values corresponding to all the nodes are higher than the corresponding current power consumption distribution values, judging whether the total power consumption of the current chassis is larger than a first power consumption increase total amount or not;
if yes, distributing required power consumption for each node directly according to the predicted power consumption value;
and if not, distributing corresponding power consumption for each node according to a preset distribution proportion.
2. The chassis power consumption management method according to claim 1, wherein the analyzing all the power consumption history data of each node to obtain a historical power consumption analysis result includes:
generating a power consumption historical time curve corresponding to each node according to all the power consumption historical data;
and training by using a time sequence prediction algorithm based on the power consumption historical time curve to obtain a power consumption prediction model.
3. The chassis power consumption management method according to claim 2, wherein the predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value includes:
and taking the power consumption value corresponding to each node in the current time period and the time of the next time period as the input of the power consumption prediction model to obtain the predicted power consumption value corresponding to each node output by the power consumption prediction model.
4. The chassis power consumption management method according to claim 2, wherein after the adjusting or allocating the power consumption of each node according to the predicted power consumption value, the current total chassis power consumption, and the current power consumption allocation value of the node, the method further comprises:
after the next time period is finished, acquiring the actual power consumption value of each node;
and correcting and adjusting the probability distribution weight in the power consumption prediction model by combining the actual power consumption value and the predicted power consumption value.
5. The chassis power consumption management method according to any one of claims 1 to 4, wherein the adjusting or allocating the power consumption of each node according to the predicted power consumption value, the current chassis total power consumption, and the current power consumption allocation value of the node includes:
if the predicted power consumption value is higher than the current power consumption distribution value, distributing power consumption for the nodes in advance based on the predicted power consumption value;
if the predicted power consumption value is lower than the current power consumption distribution value, determining the power consumption utilization rate of the current node;
and if the power consumption utilization rate of the current node is lower than a first threshold value, recovering the power consumption distributed to the current node in advance.
6. The chassis power consumption management method of claim 1, further comprising:
if the predicted power consumption value corresponding to part of the nodes is higher than the current power consumption distribution value and the predicted power consumption value corresponding to part of the nodes is lower than the current power consumption distribution value, judging whether the total power consumption of the current chassis is larger than a second power consumption increase total amount or not;
if the total power consumption of the current chassis is larger than or equal to the second power consumption increase total amount, directly distributing power consumption for each node according to the predicted power consumption value;
if the total power consumption of the current chassis is smaller than a second power consumption increase total amount, determining a target node of which the predicted power consumption value is lower than the current power consumption distribution value, and determining a power consumption utilization rate corresponding to the target node;
judging whether the power consumption utilization rate corresponding to the target node is greater than a second threshold value;
if not, recovering the pre-distribution power consumption of the target node in advance, and redistributing the nodes of which the predicted power consumption values are higher than the current power consumption distribution values after the power consumption is recovered.
7. A chassis power consumption management system, comprising:
the history acquisition module is used for acquiring power consumption history data corresponding to all nodes in the case in a preset number of history time periods;
the historical analysis module is used for analyzing all the power consumption historical data of each node to obtain a historical power consumption analysis result;
the power consumption prediction module is used for predicting the power consumption value of each node in the next time period based on the historical power consumption analysis result to obtain a predicted power consumption value;
the adjusting and distributing module is used for adjusting or distributing the power consumption of each node according to the predicted power consumption value, the current total power consumption of the chassis and the current power consumption distribution value of the node; if the predicted power consumption values corresponding to all the nodes are higher than the corresponding current power consumption distribution values, judging whether the total power consumption of the current chassis is larger than a first power consumption increase total amount or not; if yes, distributing required power consumption for each node directly according to the predicted power consumption value; and if not, distributing corresponding power consumption for each node according to a preset distribution proportion.
8. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the chassis power consumption management method according to any one of claims 1 to 6 when executing the computer program.
9. A computer-readable storage medium, having a computer program stored thereon, which, when being executed by a processor, carries out the steps of the chassis power consumption management method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010152709.9A CN111414070B (en) | 2020-03-06 | 2020-03-06 | Case power consumption management method and system, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010152709.9A CN111414070B (en) | 2020-03-06 | 2020-03-06 | Case power consumption management method and system, electronic device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111414070A CN111414070A (en) | 2020-07-14 |
| CN111414070B true CN111414070B (en) | 2021-10-15 |
Family
ID=71494327
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010152709.9A Active CN111414070B (en) | 2020-03-06 | 2020-03-06 | Case power consumption management method and system, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111414070B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111929496A (en) * | 2020-09-05 | 2020-11-13 | 广州天萌建筑设计有限公司 | Energy consumption acquisition, monitoring, analysis and alarm system for independent rooms of office building |
| CN112114650B (en) | 2020-09-11 | 2022-11-15 | 苏州浪潮智能科技有限公司 | Power consumption regulation and control method, device, equipment and readable storage medium |
| CN113567722B (en) * | 2021-07-08 | 2023-05-26 | 浙江万胜智能科技股份有限公司 | Power control method and device for electric appliance |
| CN113742167B (en) * | 2021-07-30 | 2023-08-08 | 苏州浪潮智能科技有限公司 | Control method, control device and control equipment for equipment power limitation |
| CN114840072A (en) * | 2022-04-22 | 2022-08-02 | Oppo广东移动通信有限公司 | Image quality adjusting method and device, storage medium and electronic equipment |
| CN115049083B (en) * | 2022-08-15 | 2022-11-22 | 天津理工大学 | Electromechanical device operation management method, device and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442807A (en) * | 2008-12-30 | 2009-05-27 | 北京邮电大学 | Method and system for distribution of communication system resource |
| CN102126496A (en) * | 2011-01-24 | 2011-07-20 | 浙江大学 | Parallel hybrid management control system and management control method thereof |
| CN109324679A (en) * | 2018-09-21 | 2019-02-12 | 郑州云海信息技术有限公司 | A kind of server energy consumption control method and device |
| CN109388488A (en) * | 2017-08-02 | 2019-02-26 | 联想企业解决方案(新加坡)有限公司 | Power allocation in computer system |
| CN109791532A (en) * | 2016-10-10 | 2019-05-21 | 国际商业机器公司 | Power Management in Decentralized Computing Systems |
| CN110298456A (en) * | 2019-07-05 | 2019-10-01 | 北京天泽智云科技有限公司 | Plant maintenance scheduling method and device in group system |
| CN110399216A (en) * | 2019-06-27 | 2019-11-01 | 苏州浪潮智能科技有限公司 | A kind of distribution method, system, device and the readable storage medium storing program for executing of complete machine case power consumption |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9286646B1 (en) * | 2013-07-05 | 2016-03-15 | Clean Power Research, L.L.C. | Method for managing centralized power generation with the aid of a digital computer |
| US10700523B2 (en) * | 2017-08-28 | 2020-06-30 | General Electric Company | System and method for distribution load forecasting in a power grid |
| TWI659297B (en) * | 2017-12-07 | 2019-05-11 | 技嘉科技股份有限公司 | Method for system power management and computing system thereof |
| US11379708B2 (en) * | 2018-08-09 | 2022-07-05 | Nvidia Corporation | Techniques for efficiently operating a processing system based on energy characteristics of instructions and machine learning |
-
2020
- 2020-03-06 CN CN202010152709.9A patent/CN111414070B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442807A (en) * | 2008-12-30 | 2009-05-27 | 北京邮电大学 | Method and system for distribution of communication system resource |
| CN102126496A (en) * | 2011-01-24 | 2011-07-20 | 浙江大学 | Parallel hybrid management control system and management control method thereof |
| CN109791532A (en) * | 2016-10-10 | 2019-05-21 | 国际商业机器公司 | Power Management in Decentralized Computing Systems |
| CN109388488A (en) * | 2017-08-02 | 2019-02-26 | 联想企业解决方案(新加坡)有限公司 | Power allocation in computer system |
| CN109324679A (en) * | 2018-09-21 | 2019-02-12 | 郑州云海信息技术有限公司 | A kind of server energy consumption control method and device |
| CN110399216A (en) * | 2019-06-27 | 2019-11-01 | 苏州浪潮智能科技有限公司 | A kind of distribution method, system, device and the readable storage medium storing program for executing of complete machine case power consumption |
| CN110298456A (en) * | 2019-07-05 | 2019-10-01 | 北京天泽智云科技有限公司 | Plant maintenance scheduling method and device in group system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111414070A (en) | 2020-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111414070B (en) | Case power consumption management method and system, electronic device and storage medium | |
| CN110858161B (en) | Resource allocation method, device, system, equipment and medium | |
| TWI723410B (en) | Cloud resource management system, cloud resource management method, and non-transitory computer-readable storage medium | |
| CN110300134B (en) | Storage space adjusting method and device of cloud storage resource pool and cloud storage system | |
| CN108270842B (en) | Method, system and server for pushing rights and interests task | |
| CN110888714A (en) | Container scheduling method, device and computer-readable storage medium | |
| CN111770157B (en) | Business processing method and device, electronic equipment and storage medium | |
| CN110990159A (en) | Historical data analysis-based container cloud platform resource quota prediction method | |
| CN103986766B (en) | Adaptive load balancing job task dispatching method and device | |
| US20120221730A1 (en) | Resource control system and resource control method | |
| CN105607952B (en) | Method and device for scheduling virtualized resources | |
| CN113010260A (en) | Elastic expansion method and system for container quantity | |
| CN104243405A (en) | Request processing method, device and system | |
| CN111813523A (en) | Duration pre-estimation model generation method, system resource scheduling method, device, electronic equipment and storage medium | |
| KR101630125B1 (en) | Method for resource provisioning in cloud computing resource management system | |
| CN103248622B (en) | A kind of Online Video QoS guarantee method of automatic telescopic and system | |
| CN109428950B (en) | Automatic scheduling method and system for IP address pool | |
| CN116467082A (en) | Big data-based resource allocation method and system | |
| CN114866563A (en) | Capacity expansion method, device, system and storage medium | |
| CN107203256B (en) | Energy-saving distribution method and device under network function virtualization scene | |
| CN103488538A (en) | Application extension device and application extension method in cloud computing system | |
| CN109274991B (en) | Memory management method and system of smart television | |
| CN110399216B (en) | Method, system and device for distributing power consumption of whole machine box and readable storage medium | |
| CN114090256A (en) | Application delivery load management method and system based on cloud computing | |
| CN111932122B (en) | Distribution network disaster prevention material allocation method, device, terminal and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |