CN107132903B - Energy-saving management implementation method, device and network equipment - Google Patents

Abstract

The invention provides an energy-saving management implementation method, an energy-saving management implementation device and network equipment, wherein the energy-saving management implementation method comprises the following steps: periodically obtaining a current value of a variable parameter that identifies a processor's processing capability; determining the number of the current required running state processors and the number of the current required working state processors according to the single processor processing capacity value and the current value; obtaining adjustment information for adjusting the running state of the processors according to the number of running state processors required at the last moment, the number of working state processors required at the last moment, the number of current running state processors required at the current moment and the number of current working state processors required at the current moment; performing energy-saving management on the processor according to the adjustment information; wherein the running state processor comprises a working state processor and/or a ready state processor. The scheme provided by the invention can better realize energy-saving management and improve the resource utilization rate by adjusting the running state of the processor in real time according to the variable parameter for marking the processing capacity of the processor.

Description

A method, device and network device for realizing energy saving management

technical field

The present invention relates to the technical field of communications, and in particular, to a method, device and network device for realizing energy saving management.

Background technique

At present, the common sending and receiving modes of network card devices include interrupt mode and polling mode. No matter which mode is adopted, there are problems of uncontrollable CPU (Central Processing Unit, Central Processing Unit) utilization efficiency and general waste of system resources. For example, an existing data plane development kit named DPDK (Data Plane Development, data plane development), when its working mode is PMD (Pull Model Driver, polling interrupt mode driver) mode, has zero copy, no interruption and The advantages of batch processing, etc., are very suitable for network data packet analysis, processing and other operations, and have a certain performance improvement for data packet sending and receiving and multi-core operations. But PMD has a feature that the CPU that executes PMD is running at 100% all the time. Multiple CPUs running DPDK PMD on a network device idle, causing the temperature of the network device to increase passively. When an application program using this technology runs on a virtual machine, the CPU load of the virtual machine will also be artificially high, and the CPU resources of the system host will be wasted.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method, an apparatus and a network device for realizing energy saving management, so as to solve the problem of wasting CPU resources of the network device in the prior art.

In order to solve the above technical problems, an embodiment of the present invention provides a method for implementing energy conservation management, including:

Periodically obtain the current value of the variable parameter that identifies the processing capability of the processor;

According to the single processor processing capability value and the current value, determine the currently required number of processors in the running state and the number of processors in the currently required working state;

According to the number of processors in the running state required at the last moment, the number of processors in the working state required at the previous moment, the currently required number of processors in the running state, and the number of processors in the currently required working state, a method for adjusting the running state of the processor is obtained. Status adjustment information;

Perform energy-saving management on the processor according to the adjustment information;

Wherein, the running state processor includes a working state processor and/or a ready state processor.

Optionally, before the periodic acquisition of the current value of the variable parameter identifying the processing capability of the processor, the energy saving management implementation method further includes:

The variable parameter is determined, and a single processor processing capability value corresponding to the variable parameter is obtained.

Optionally, the step of determining the currently required number of processors in the running state and the number of currently required processors in the working state according to the single processor processing capability value and the current value includes:

According to the ratio of the current value to the processing capability value of the single processor and the preset threshold, and the sum of the value and the first fixed value, the currently required number of processors in the running state is obtained;

According to the ratio of the current value to the processing capability value of the single processor, and the sum of the second fixed value, the currently required number of processors in the working state is obtained.

Optionally, according to the number of processors in the running state required at the last moment, the number of processors in the working state required at the previous moment, the number of processors in the currently required running state and the number of processors in the currently required working state, obtain: The steps for adjusting the adjustment information for the operating state of the processor include:

If the currently required number of running state processors is less than the number of running state processors required at the previous moment, monitor whether the number of running state processors required at all moments within a preset time period after the current moment is less than the said number of running state processors the number of processors in the running state required at the last moment, if yes, obtain the adjustment information for converting the first preset number of processors in the ready state into the dormant state processors;

If the currently required number of processors in the running state is greater than the number of processors in the running state required at the previous moment, the adjustment information for waking up the second preset number of processors in the dormant state so as to be converted into the processors in the ready state is obtained.

Optionally, according to the number of processors in the running state required at the last moment, the number of processors in the working state required at the previous moment, the number of processors in the currently required running state and the number of processors in the currently required working state, obtain: The steps for adjusting the adjustment information for the operating state of the processor include:

If the currently required number of processors in the running state is less than the number of processors in the running state required at the previous moment, obtaining adjustment information for converting the third preset number of processors in the working state into processors in the ready state;

If the currently required number of processors in the running state is greater than the number of processors in the running state required at the previous moment, the adjustment information for converting the fourth preset number of ready state processors into working state processors is obtained.

Optionally, the step of performing energy-saving management on the processor according to the adjustment information includes:

The running state of the processor is adjusted in real time according to the adjustment information, and the load of each processor is adjusted accordingly.

The present invention also provides a device for realizing energy-saving management, including:

The acquisition module is used to periodically acquire the current value of the variable parameter that identifies the processing capability of the processor;

a determining module, configured to determine the currently required number of processors in the running state and the number of processors in the currently required working state according to the processing capability value of the single processor and the current value;

The first processing module is configured to, according to the number of processors in the running state required at the last moment, the number of processors in the working state required at the previous moment, the number of processors in the currently required running state and the number of processors in the currently required working state, Get the adjustment information used to adjust the running state of the processor;

a second processing module, configured to perform energy-saving management on the processor according to the adjustment information;

Wherein, the running state processor includes a working state processor and/or a ready state processor.

Optionally, the device for implementing energy saving management further includes:

A third processing module, configured to determine the variable parameter and obtain a single processor processing capability value corresponding to the variable parameter before the obtaining module performs the operation.

Optionally, the determining module includes:

a first processing sub-module, configured to obtain the currently required number of processors in the running state according to the ratio of the current value to the processing capability value of the single processor and a preset threshold, as well as the sum with the first fixed value;

The second processing sub-module is configured to obtain the currently required number of working processors according to the ratio of the current value to the processing capability value of the single processor, and the sum value of the current value and the second fixed value.

Optionally, the first processing module includes:

The third processing sub-module is configured to monitor whether the current required running state processors are less than the required running state processors at the previous moment, monitoring whether the running status is required at all moments within a preset time period after the current moment The number of state processors is less than the number of running state processors required at the previous moment, and if so, obtain adjustment information for converting the first preset number of ready state processors into dormant state processors;

The fourth processing sub-module is configured to wake up a second preset number of dormant state processors if the currently required number of running state processors is greater than the number of running state processors required at the previous moment, so as to transform them into Adjustment information for the ready state processor.

Optionally, the first processing module includes:

The fifth processing sub-module is used to obtain the processing of converting the third preset number of working state processors into ready state if the currently required number of running state processors is less than the number of running state processors required at the previous moment adjustment information of the device;

The sixth processing sub-module is used to obtain the processing of converting the fourth preset number of ready-state processors into working state if the currently required number of running state processors is greater than the number of running state processors required at the previous moment adjustment information for the device.

Optionally, the second processing module includes:

The seventh processing sub-module is configured to adjust the running state of the processors in real time according to the adjustment information, and adjust the load of each of the processors accordingly.

The present invention also provides a network device, comprising: the above-mentioned device for realizing energy saving management.

The beneficial effects of the above-mentioned technical solutions of the present invention are as follows:

In the above solution, the energy saving management implementation method can better realize energy saving management and improve resource utilization rate by adjusting the running state of the processor in real time according to the variable parameter that identifies the processing capability of the processor.

Description of drawings

FIG. 1 is a schematic flow chart 1 of a method for realizing energy saving management according to Embodiment 1 of the present invention;

2 is a second schematic flowchart of a method for implementing energy conservation management according to Embodiment 1 of the present invention;

3 is a schematic diagram of a software system of a session border controller SBC application example according to Embodiment 1 of the present invention;

FIG. 4 is a third schematic flowchart of a method for implementing energy-saving management according to Embodiment 1 of the present invention;

5 is a fourth schematic flowchart of a method for implementing energy-saving management according to Embodiment 1 of the present invention;

FIG. 6 is a schematic structural diagram of an apparatus for implementing energy-saving management according to Embodiment 2 of the present invention.

Detailed ways

In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

The present invention provides a variety of solutions for the problem of wasting CPU resources of network devices in the prior art, the details are as follows:

Example 1

As shown in FIG. 1 , the energy saving management implementation method provided by the first embodiment of the present invention includes:

Step 11: Periodically obtain the current value of the variable parameter that identifies the processing capability of the processor;

Step 12: According to the single processor processing capability value and the current value, determine the currently required number of processors in the running state and the number of processors in the currently required working state;

Step 13: According to the number of processors in the running state required at the last moment, the number of processors in the working state required at the previous moment, the number of processors in the currently required running state and the number of processors in the current Adjustment information of the operating state of the processor;

Step 14: Perform energy-saving management on the processor according to the adjustment information;

Wherein, the running state processor includes a working state processor and/or a ready state processor.

The regular time interval can be 1s, but not limited thereto; the variable parameter can be a single parameter or a combination of several related parameters.

The energy-saving management implementation method provided by the first embodiment of the present invention can better realize energy-saving management and improve resource utilization by adjusting the running state of the processor in real time according to the variable parameter that identifies the processing capability of the processor.

Further, before the periodic acquisition of the current value of the variable parameter identifying the processing capability of the processor, the energy saving management implementation method further includes: determining the variable parameter, and acquiring the processing capability of a single processor corresponding to the variable parameter. value.

Specifically, the step of determining the currently required number of processors in the running state and the currently required number of processors in the working state according to the processing capability value of a single processor and the current value includes: according to the current value and the single processor The ratio of the processor processing capacity value to the preset threshold value, and the sum of the first fixed value to obtain the currently required running state processors; according to the ratio of the current value to the single processor processing capacity value, and The sum with the second fixed value obtains the currently required number of processors in the working state.

The preset threshold may preferably be 95%. In this embodiment, the first fixed value and the second fixed value may preferably be 1.

For the case of mutual conversion between the ready state processor and the sleep state processor, the number of running state processors required at the previous moment, the number of working state processors required at the previous moment, and the currently required running state processors The steps of obtaining the adjustment information for adjusting the running state of the processor include:

If the currently required number of running state processors is less than the number of running state processors required at the previous moment, monitor whether the number of running state processors required at all moments within a preset time period after the current moment is less than the said number of running state processors the number of processors in the running state required at the last moment, if yes, obtain the adjustment information for converting the first preset number of processors in the ready state into the dormant state processors;

If the currently required number of processors in the running state is greater than the number of processors in the running state required at the previous moment, the adjustment information for waking up the second preset number of processors in the dormant state so as to be converted into the processors in the ready state is obtained.

It can be said that if the number of running state processors required at a later moment is continuously found to be less than the number of running state processors required at a previous moment, the ready state processor is converted into a dormant state processor, or it can be said that , the variable parameter value after a certain time is continuously smaller than the variable parameter value corresponding to the time, then the ready state processor is transformed into a dormant state processor.

For the case of mutual conversion between the ready state processor and the working state processor, the number of running state processors required at the previous moment, the number of working state processors required at the previous moment, and the currently required running state processors The steps of obtaining the adjustment information for adjusting the running state of the processor include:

If the currently required number of processors in the running state is less than the number of processors in the running state required at the previous moment, obtaining adjustment information for converting the third preset number of processors in the working state into processors in the ready state;

If the currently required number of processors in the running state is greater than the number of processors in the running state required at the previous moment, the adjustment information for converting the fourth preset number of ready state processors into working state processors is obtained.

Preferably, the step of performing energy saving management on the processor according to the adjustment information includes: adjusting the running state of the processor in real time according to the adjustment information, and correspondingly adjusting the load of each of the processors.

The processor may be a central processing unit (CPU) or other types of processors, which are not limited herein.

The following specifically describes the energy saving management implementation method provided in Embodiment 1 of the present invention.

Aiming at the problem of wasting CPU resources of network devices in the prior art, embodiments of the present invention provide a method for implementing energy-saving management by dynamically adjusting the number of CPUs executing a packet sending and receiving module according to a certain dynamic variable or a combination of multiple dynamic variables. In the case of a certain dynamic variable or a combination of a group of dynamic variables, the network device uses different numbers of CPUs to run the module for sending and receiving packets.

As shown in Figure 2, the method includes the following steps:

Step 21: The network device determines a single CPU processing capability value according to the actual situation and determines a dynamic variable or a combination of multiple dynamic variables used by the network device to adjust the number of CPUs and CPU utilization.

The network device first determines a numerical value or a combination of several numerical values to represent the processing capability value of a single CPU. The value of the processing capability of a single CPU can be the maximum packet throughput that a single CPU can process, the CPU occupancy rate when the packet transceiver module is running normally, or the temperature of the CPU, and so on.

Similar numerical values can also be a combination of several numerical values. The network device uses a single CPU processing capability value to calculate the number of CPUs currently required to execute the module for sending and receiving packets. At the same time, the network device determines a representation method for measuring the overall packet traffic load of the network device. It can be a dynamic variable or a combination of several dynamic variables, such as: the total packet throughput of the current network device, the total occupancy rate of each CPU executing the packet transceiver module, or the temperature of the CPU, and so on. Dynamic variables, or a combination of several dynamic variables.

Step 22: The network device periodically obtains the system dynamic variable value or the dynamic variable combination value, calculates the number of CPUs that need to execute the packet sending and receiving module, and determines whether to adjust the number of CPUs that execute the packet sending and receiving module.

The calculation formula that can be used in step 22 is:

The number of CPUs that execute the module for sending and receiving packets =

Network device dynamic variable value or dynamic variable combined value/single CPU processing capacity value +1

(The division calculation adopts the method of direct rounding)

A CPU that executes the module for sending and receiving packets has three operating states. One is the working state, in which the CPU executes the packet sending and receiving module, which is responsible for sending and receiving packets of the network port. One is the ready state. At this time, the CPU is in the running state, but it is not responsible for sending and receiving packets of the network port. It can start to be responsible for sending and receiving packets of the network port at any time. One is the sleep state, where the CPU sleeps to save energy. The ready state is a transition state between the working state and the dormant state, which can reduce the abnormal problem caused by the inability of the CPU to process packet traffic normally when the CPU wakes up from the dormant state.

Step 23: After determining to adjust the number of CPUs, increase or decrease the number of CPUs running the packet sending and receiving module according to specific conditions, and adjust the traffic load of the CPUs as required.

When adjusting the number of CPUs, you need to follow the "one fast and one slow" principle:

1) The application decides that it is faster to adjust the CPU from the dormant state to the working state. Taking into account no packet loss, traffic bursts, etc., as well as the decline of a single CPU capacity in a virtualized environment, when the system reaches the threshold of the processing capacity of a single CPU, such as 95%, the application will be in the The CPU in the sleep state wakes up, enters the ready state, and is ready to enter the working state at any time. The threshold value can be set according to the actual situation of the network equipment.

2) The application decides that it is slower to adjust the CPU from the working state to the sleep state. Only when it is found that the number of CPUs executing the packet sending and receiving module can be reduced for many times in a row, the application program will reduce the number of CPUs to ensure that the CPU executing the packet sending and receiving module can run smoothly and reduce the occurrence of abnormal situations such as packet loss.

If you need to wake up the CPU to enter the ready state, wake up the new CPU and prepare to take over the sending and receiving of some network ports.

If the CPU in the working state needs to be increased, the traffic load of the CPU in the working state is reduced, and part of the traffic load is allocated to the CPU in the ready state, and the CPU in the ready state begins to enter the working state.

If the CPU in the working state needs to be reduced, a CPU in the working state is selected to enter the ready state, and the traffic load originally undertaken by this CPU is allocated to other CPUs in the working state.

If the CPU in the ready state needs to enter the sleep state, the CPU in the ready state is migrated to the sleep state. Reduce energy consumption.

It can be seen from the above that the solution provided by the first embodiment of the present invention makes full use of the computing power of the CPU that executes the packet sending and receiving module. When the traffic load of the network device is low, the number of idle CPUs is reduced on the basis of ensuring the packet throughput performance. , to achieve the goal of energy-saving management. When the application program of the network device runs on the virtual machine, the CPU load of the virtual machine can be effectively reduced, the CPU resource utilization of the system host can be improved, and it is also very meaningful for the minimum configuration of the virtualization product and balancing the performance of each virtual machine.

Hereinafter, the energy saving management implementation method provided in Embodiment 1 of the present invention will be further described in detail by taking the SBC (Session Border Controller, Session Border Controller) representing the network device and the concurrent call volume representing the variable parameter value as an example.

As shown in FIG. 3 , it is a module related to network port packet processing in a session border controller SBC in an NGN (Next Generation Network)/IMS (IP Multimedia Subsystem, IP Multimedia Subsystem) network. Each network port of the SBC is guarded by a thread for sending and receiving packets. A packet sending and receiving thread can be responsible for sending and receiving packets of one network port, and can also be responsible for sending and receiving packets of other network ports. According to Figure 3, it is assumed that the SBC has 4 network ports and four threads for sending and receiving packets, and four CPUs are allocated to execute the module for sending and receiving packets. The main idea of the first embodiment of the present invention can be said to be: an adjustment module is added to the SBC to adjust the number of CPUs executing the packet sending and receiving module and the traffic load of each CPU, and the adjustment module periodically checks the current SBC executing the packet sending and receiving module. The total load of the CPU, increase or decrease the number of running CPUs as required, and adjust the packet traffic load of the CPU in the working state.

As shown in FIG. 4 , the solution provided in Embodiment 1 of the present invention specifically includes the following steps:

Step 41: Obtain the maximum processing capability value of a single CPU.

Since the main packet traffic of the SBC is the signaling media packet traffic generated by concurrent calls, an appropriate concurrent call volume value can be used as the value of the packet processing capability of a single CPU, assuming that the value is 1000 concurrent calls.

Step 42: Obtain the total packet processing load of the network device.

Since the main packet traffic of the SBC is the signaling media packet traffic generated by concurrent calls, the current total number of concurrent calls is used as the total load value of the packet traffic that needs to be processed by the CPU currently executing the packet sending and receiving module.

Step 43: Obtain the threshold for waking up the new CPU and preparing to enter the working state.

According to the SBC application scenario, 95% of the processing capacity of a single CPU can be used as the threshold for the new CPU to wake up from the sleep state and enter the state ready to receive packets (ie, the ready state). The number of CPUs to start can be calculated with the following formula:

Number of CPUs started =

The total load of the network port of the network device/(the value of the packet processing capacity of a single CPU × the threshold value)+1

(The division calculation adopts the method of direct rounding)

In the process of increasing the number of concurrent calls, the calculated number of CPUs, one of the CPUs is in the ready state, and the other CPUs are in the running state.

The thresholds required by other network devices can be determined according to their respective application scenarios.

Step 44: Calculate the number of CPUs executing the packet sending and receiving module according to the calculation formula, and adjust the number of CPUs and the traffic load of each CPU as required.

The adjustment module periodically obtains the current concurrent call volume of the SBC, and calculates the number of CPUs used to execute the module for sending and receiving packets. According to the SBC application scenario, this period can be set to 1 second.

The calculation formula that can be used in step 44 is:

The number of CPUs that execute the module for sending and receiving packets =

Network device dynamic variable value or dynamic variable combined value/single CPU processing capacity value +1

(The division calculation adopts the method of direct rounding)

When the concurrent call traffic is relatively low, for example: 700 concurrent calls, according to the calculation formula, it can be calculated that the number of CPUs currently in the working state is 1:

The number of CPUs that execute the module for sending and receiving packets = 700/1000+1

A CPU is used to execute the packet sending and receiving module thread, and this packet sending and receiving module thread is responsible for sending and receiving packets of all network ports. At this time, only one CPU is in the working state, and the other CPUs enter the sleep state.

When the concurrent calls start to increase and reach the threshold (1000×95%), according to the calculation formula of “number of CPUs activated”, it is necessary to wake up a dormant CPU and prepare to enter the working state. When the traffic increases to the point where two CPUs are required for processing, the CPU in the ready state can immediately start working, saving time for wake-up and reducing the occurrence of abnormal situations such as packet loss. At this time, one CPU is in a working state, one CPU is in a ready state, and the other CPUs are in a sleep state.

When the concurrent traffic fluctuates around 950, according to the principle of "one fast and one slow", the CPU in the ready state cannot be restored to the dormant state immediately. It is necessary to adjust the module to find that the concurrent traffic is lower than 950 for several consecutive times, indicating that the current concurrent calls are relatively stable. The CPU in the ready state can be returned to the sleep state.

When the concurrent call traffic increases to 1000, according to the calculation formula of "Number of CPUs that execute the packet sending and receiving module", it can be calculated that 2 CPUs need to be used to enter the working state. At this time, the CPU in the ready state starts to take over the sending and receiving work of some network ports. In order to balance the load of the two CPUs as much as possible, the new CPU in the working state can be responsible for sending and receiving packets for a certain number of network ports according to the actual situation. Responsible for handling the traffic of a small number of network ports.

When the concurrent traffic increases to 1900, you can refer to the processing flow when the concurrent traffic is 950 to wake up the new CPU and enter the ready state. At this time, two CPUs are in the working state, and one CPU is awakened and enters the ready state.

When the concurrent traffic increases to 2000, you can refer to the processing flow when the concurrent traffic is 1000 to let the CPU in the ready state enter the working state, and adjust the traffic load of the CPU in each working state, so that the load of each CPU can be balanced. At this time, the three CPUs are in a working state, and the other CPUs are in a dormant state.

For other CPU adjustment processes caused by traffic growth, refer to the above process.

When the concurrent traffic decreases, for example, from 2000 to 1900, according to the calculation formula, two CPUs are required to enter the working state, and one CPU is required to enter the ready state. The adjustment module selects a working state CPU to enter the ready state, and adjusts the traffic load of the remaining two working state CPUs, so that the two working state CPUs can normally process the sending and receiving work of all network ports of the network device.

When the concurrent traffic continues to decrease to 1800, according to the calculation formula, two CPUs are required to enter the working state, and the CPU that originally entered the ready state enters the sleep state.

For other CPU adjustment procedures caused by traffic reduction, refer to the above procedures.

To sum up, the solution provided by the first embodiment of the present invention can also be summarized as the process shown in FIG. 5 , or it can be said that the adjustment module adjusts the CPU that executes the packet sending and receiving module according to the calculation result according to the process shown in FIG. 5 . ,include:

Step 51: start;

Step 52: the software program obtains the overall traffic load of the current system;

Step 53: Calculate the number of CPUs according to the processing capacity of a single CPU;

Step 54: Determine whether the number of CPUs changes, if the number of CPUs does not change, go to Step 511, if the number of CPUs changes, go to Step 55;

Step 55: determine the content of the change, if it is necessary to increase the number of running CPUs, then enter step 56, if it is necessary to reduce the number of working state CPUs, then enter step 57, if it is necessary to increase the number of working state CPUs, then enter step 59;

Step 56: wake up the dormant CPU to enter the ready state, and then enter step 511;

Step 57: Select a working CPU to enter the ready state;

Step 58: Adjust the traffic load of the working CPU, and then go to Step 511;

Step 59: make the CPU in the ready state enter the working state;

Step 510: Adjust the traffic load of all working CPUs, and then go to Step 511;

Step 511: End.

Embodiment 2

As shown in FIG. 6 , the device for realizing energy saving management provided by Embodiment 2 of the present invention includes:

an acquisition module 61, configured to periodically acquire the current value of the variable parameter that identifies the processing capability of the processor;

A determination module 62, configured to determine the currently required number of processors in the running state and the number of processors in the currently required working state according to the processing capability value of the single processor and the current value;

The first processing module 63 is configured to calculate the number of processors in the running state required at the last moment, the number of processors in the working state required at the previous moment, the number of processors in the currently required running state, and the number of processors in the currently required working state. , get the adjustment information used to adjust the running state of the processor;

The second processing module 64 is configured to perform energy-saving management on the processor according to the adjustment information;

Wherein, the running state processor includes a working state processor and/or a ready state processor.

The regular time interval can be 1s, but not limited thereto; the variable parameter can be a single parameter or a combination of several related parameters.

The energy-saving management implementation device provided in the second embodiment of the present invention can better realize energy-saving management and improve resource utilization by adjusting the running state of the processor in real time according to the variable parameter that identifies the processing capability of the processor.

Further, the energy saving management implementation device further includes: a third processing module, configured to determine the variable parameter and obtain a single processor processing capability value corresponding to the variable parameter before the obtaining module performs the operation.

Specifically, the determining module includes: a first processing sub-module, configured to obtain the The currently required number of processors in the running state; a second processing submodule, configured to obtain the currently required work according to the ratio of the current value to the processing capability value of the single processor, and the sum of the current value and the second fixed value number of state processors.

The preset threshold may preferably be 95%. In this embodiment, the first fixed value and the second fixed value may preferably be 1.

In the case of the mutual conversion between the ready state processor and the sleep state processor, the first processing module includes: a third processing sub-module, which is used for if the number of the currently required running state processors is less than the previous time. The number of running state processors required, then monitor whether the number of running state processors required at all times is less than the number of running state processors required at the previous moment in the preset time period after the current moment, and if so, obtain the first Adjustment information of the preset number of ready state processors being transformed into sleep state processors;

The fourth processing sub-module is configured to wake up a second preset number of dormant state processors if the currently required number of running state processors is greater than the number of running state processors required at the previous moment, so as to transform them into Adjustment information for the ready state processor.

It can be said that if the number of running state processors required at a later moment is continuously found to be less than the number of running state processors required at a previous moment, the ready state processor is converted into a dormant state processor, or it can be said that , the variable parameter value after a certain time is continuously smaller than the variable parameter value corresponding to the time, then the ready state processor is transformed into a dormant state processor.

In the case of mutual conversion between the ready state processor and the working state processor, the first processing module includes: a fifth processing sub-module, used for if the currently required number of running state processors is less than the previous time If the number of processors in the running state is required, the adjustment information for converting the third preset number of processors in the working state into the processors in the ready state is obtained;

The sixth processing sub-module is used to obtain the processing of converting the fourth preset number of ready-state processors into working state if the currently required number of running state processors is greater than the number of running state processors required at the previous moment adjustment information for the device.

Preferably, the second processing module includes: a seventh processing sub-module, configured to adjust the running state of the processors in real time according to the adjustment information, and correspondingly adjust the load of each of the processors.

In this embodiment, the processor may preferably be a central processing unit (CPU), or may be other types of processors, which are not limited herein.

Wherein, the above-mentioned implementation embodiments of the energy-saving management implementation method are all applicable to the embodiments of the energy-saving management implementation device, and can also achieve the same technical effect.

In order to solve the above technical problem, the present invention also provides a network device, including: the above-mentioned device for realizing energy saving management.

Wherein, the implementation embodiments of the above-mentioned energy saving management implementation apparatus are all applicable to the embodiments of the network device, and the same technical effect can also be achieved.

It should be noted that many functional components described in this specification are referred to as modules/sub-modules in order to more particularly emphasize the independence of their implementations.

In this embodiment of the present invention, the modules/sub-modules may be implemented in software so as to be executed by various types of processors. For example, an identified executable code module may comprise one or more physical or logical blocks of computer instructions, which may be structured as objects, procedures, or functions, for example. Nonetheless, the executable code of the identified module need not be physically located together, but may include different instructions stored in different bits that, when logically combined, constitute the module and implement the specification of the module Purpose.

In practice, an executable code module may be a single instruction or many instructions, and may even be distributed over multiple different code segments, among different programs, and across multiple memory devices. Likewise, operational data may be identified within modules, and may be implemented in any suitable form and organized within any suitable type of data structure. The operational data may be collected as a single data set, or may be distributed over different locations (including over different storage devices), and may exist at least in part only as electronic signals on a system or network.

When a module can be implemented by software, considering the level of existing hardware technology, a module that can be implemented by software, regardless of cost, can build corresponding hardware circuits to implement corresponding functions. The hardware circuits include conventional very large scale integration (VLSI) circuits or gate arrays as well as off-the-shelf semiconductors such as logic chips, transistors, or other discrete components. A module may also be implemented in programmable hardware devices, such as field programmable gate arrays, programmable array logic, programmable logic devices, and the like.

The above are the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as It is the protection scope of the present invention.

Claims (13)

1. An energy-saving management implementation method is characterized by comprising the following steps:

periodically obtaining a current value of a variable parameter that identifies a processor's processing capability;

determining the number of the current required running state processors and the number of the current required working state processors according to the single processor processing capacity value and the current value;

obtaining adjustment information for adjusting the running state of the processors according to the number of running state processors required at the last moment, the number of working state processors required at the last moment, the number of current running state processors required at the current moment and the number of current working state processors required at the current moment, wherein the running state comprises a working state, a ready state and a dormant state;

performing energy-saving management on the processor according to the adjustment information;

wherein the number of the running state processors is equal to the sum of the number of the working state processors and the number of the ready state processors.

2. The energy conservation management implementation of claim 1, wherein prior to said periodically obtaining the current value of a variable parameter that identifies processor processing capacity, the energy conservation management implementation further comprises:

and determining the variable parameter, and acquiring a single processor processing capacity value corresponding to the variable parameter.

3. The energy conservation management implementation method of claim 1 wherein the step of determining a current number of required run state processors and a current number of required active state processors based on the single processor throughput capability value and the current value comprises:

obtaining the number of the processors in the current required running state according to the ratio of the current value to the single processor processing capacity value and a preset threshold value and the sum of the current value and a first fixed value;

and obtaining the number of the currently required working state processors according to the ratio of the current value to the single processor processing capacity value and the sum of the current value and a second fixed value.

4. The method according to claim 1, wherein the step of obtaining the adjustment information for adjusting the processor operating status according to the number of the operating status processors required at the previous time, the number of the working status processors required at the previous time, the number of the currently required operating status processors, and the number of the currently required working status processors comprises:

if the number of the processors in the current required running state is smaller than the number of the processors in the running state required at the last moment, monitoring whether the number of the processors in the running state required at all the moments is smaller than the number of the processors in the running state required at the last moment in a preset time period after the current moment, and if so, obtaining adjustment information for converting the processors in the first preset number of ready states into processors in a dormant state;

and if the number of the processors in the current required running state is greater than the number of the processors in the running state required at the last moment, obtaining adjustment information for awakening a second preset number of processors in the dormant state to convert the processors into processors in the ready state.

5. The method according to claim 1, wherein the step of obtaining the adjustment information for adjusting the processor operating status according to the number of the operating status processors required at the previous time, the number of the working status processors required at the previous time, the number of the currently required operating status processors, and the number of the currently required working status processors comprises:

if the number of the processors in the current required running state is smaller than the number of the processors in the running state required at the last moment, obtaining adjustment information for converting a third preset number of working state processors into ready state processors;

and if the number of the processors in the current required running state is greater than the number of the processors in the running state required at the last moment, obtaining adjustment information for converting the fourth preset number of ready-state processors into working-state processors.

6. The method of claim 1, wherein the step of managing the processor for power saving according to the adjustment information comprises:

and adjusting the running state of the processors in real time according to the adjustment information, and correspondingly adjusting the load of each processor.

7. An energy saving management implementation apparatus, comprising:

an acquisition module for periodically acquiring a current value of a variable parameter identifying a processing capability of a processor;

the determining module is used for determining the number of the currently required running state processors and the number of the currently required working state processors according to the single processor processing capacity value and the current value;

the first processing module is used for obtaining adjustment information for adjusting the running state of the processors according to the number of running state processors required at the last moment, the number of working state processors required at the last moment, the number of current running state processors required at the last moment and the number of current working state processors required at the current moment, wherein the running state comprises a working state, a ready state and a dormant state;

the second processing module is used for carrying out energy-saving management on the processor according to the adjustment information;

wherein the number of the running state processors is equal to the sum of the number of the working state processors and the number of the ready state processors.

8. The energy saving management implementing device of claim 7, wherein the energy saving management implementing device further comprises:

and the third processing module is used for determining the variable parameter and acquiring the processing capacity value of the single processor corresponding to the variable parameter before the acquisition module executes the operation.

9. The energy conservation management implementation apparatus of claim 7, wherein the determining module comprises:

the first processing submodule is used for obtaining the number of the processors in the current required running state according to the ratio of the current value to the single processor processing capacity value and a preset threshold value and the sum of the current value and a first fixed value;

and the second processing submodule is used for obtaining the number of the currently required working state processors according to the ratio of the current value to the single processor processing capacity value and the sum of the current value and a second fixed value.

10. The energy conservation management implementation apparatus of claim 7, wherein the first processing module comprises:

a third processing sub-module, configured to monitor whether the number of the operation state processors required at all times within a preset time period after the current time is less than the number of the operation state processors required at the previous time if the number of the operation state processors required at the current time is less than the number of the operation state processors required at the previous time, and if so, obtain adjustment information for converting the first preset number of ready state processors into dormant state processors;

and the fourth processing submodule is used for obtaining adjustment information for awakening a second preset number of processors in a dormant state to convert the processors into processors in a ready state if the number of the processors in the current required running state is greater than the number of the processors in the running state required at the last moment.

11. The energy conservation management implementation apparatus of claim 7, wherein the first processing module comprises:

a fifth processing submodule, configured to obtain adjustment information for converting a third preset number of working-state processors into ready-state processors if the number of currently required working-state processors is smaller than the number of working-state processors required at the previous time;

and the sixth processing submodule is used for obtaining adjustment information for converting a fourth preset number of ready-state processors into working-state processors if the number of the processors in the current required running state is greater than the number of the processors in the last running state.

12. The energy conservation management implementation apparatus of claim 7, wherein the second processing module comprises:

and the seventh processing submodule is used for adjusting the running state of the processors in real time according to the adjustment information and correspondingly adjusting the load of each processor.

13. A network device, comprising: the energy saving management implementing apparatus according to any one of claims 7 to 12.

Images