CN112433595A - Energy consumption supervision method of computer system - Google Patents
Energy consumption supervision method of computer system Download PDFInfo
- Publication number
- CN112433595A CN112433595A CN201910787213.6A CN201910787213A CN112433595A CN 112433595 A CN112433595 A CN 112433595A CN 201910787213 A CN201910787213 A CN 201910787213A CN 112433595 A CN112433595 A CN 112433595A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- computer system
- data
- processor
- rate
- 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.)
- Granted
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 20
- 230000000875 corresponding effect Effects 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000005070 sampling Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims abstract description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- 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
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/324—Power saving characterised by the action undertaken by lowering clock frequency
-
- 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
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/3243—Power saving in microcontroller unit
-
- 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
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/3296—Power saving characterised by the action undertaken by lowering the supply or operating voltage
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/10—Flow control between communication endpoints
- H04W28/14—Flow control between communication endpoints using intermediate storage
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
Abstract
The invention relates to an energy consumption supervision method of a computer system, which comprises the following steps: the system consists of a plurality of sensors, actuators and base stations which are in wireless connection; the sensor sampling period signal is sent to a base station, the base station can process corresponding data, and then an actuator is excited to make corresponding action; the base station is usually provided with a data buffer area for storing data which is acquired from the sensor but is not processed; and the processor sequentially fetches the cached data after finishing the previous round of data processing.
Description
Technical Field
The invention relates to an energy consumption supervision method of a computer system.
Background
In recent years, due to the rapid development of computer technology and communication technology, wireless embedded systems deployed in open and unknown environments have been widely used, such as traffic monitoring, environmental monitoring, home security systems, and the like. However, such systems typically have fairly stringent requirements for their quality of service (qos). For example, wireless distributed video surveillance systems require the identification of objects by analyzing successive video images in a very short time. However, the processing time of the image varies with the change of the content, background, size, etc. of the captured image, and becomes an uncertain value. In addition, as a wireless embedded system, the wireless link feature usually adopted requires a battery to supply power, and the system requires that the processor must reduce energy consumption and prolong the service life of the system. When the quality of service requirement of the system is high, it may cause a crash of the system or a transitional consumption of power supply resources. The classic real-time scheduling method can avoid the overload of tasks and delay the completion deadline of system tasks under the static assumed condition [4,5 ]. When the system is in a dynamically changing environment, the scheduling algorithm based on the control theory can also meet the real-time requirement of the system. In order to meet the requirement of real-time performance, the authors design a control mechanism to control the loading rate of tasks to achieve the utilization level of the multiprocessor, thereby ensuring the real-time performance of the system. However, existing work only guarantees real-time performance of the system, without considering the energy consumption of the system.
Disclosure of Invention
The invention designs an energy consumption monitoring method of a computer system, which solves the technical problems that the existing method only considers the energy consumption of the system and does not consider the real-time requirement of the system.
In order to solve the technical problems, the invention adopts the following scheme:
a method for energy consumption supervision of a computer system, comprising the steps of: the system consists of a plurality of sensors, actuators and base stations which are in wireless connection; the sensor sampling period signal is sent to a base station, the base station can process corresponding data, and then an actuator is excited to make corresponding action; the base station is usually provided with a data buffer area for storing data which is acquired from the sensor but is not processed; and the processor sequentially fetches the cached data after finishing the previous round of data processing.
Preferably, the voltage and operating frequency of the processor are reduced when there is no task or the load is light.
Preferably, each node is assumed to support m discrete operating modes, each operating mode having a different energy consumption and execution time for a standard task load, and a corresponding CPU operating frequency.
Preferably, in order to avoid the delay of the task, a model for representing the system workload needs to be found, the utilization rate of the CPU can directly reflect the condition of the system workload, and the utilization rate of the CPU is assumed to be calculated by the following formula
Wherein, nk is the number of processing tasks in the kth period, and es and ds are the execution time and the relative deadline of the tasks respectively; the utilization rate u (k) is expected not to exceed 100% and should be greater than a set value δ.
Preferably, given a CPU utilization setting δ (δ ≦ 100%), a buffer size γ, and a set of processor operating modes Mi＝{(ti,pi) 1.. m }, where m is the number of operating modes, ti is the execution time of mode i, and pi is the energy consumption of mode i; defining pki as the energy consumption of the mode i in the k period; the goal of the problem is then to assign the appropriate operating mode to the processor while ensuring that the total energy consumption of the system is minimized over n cycles, namely:
simultaneously, the constraint conditions are met:
(|Rate(k)-Rate|)·T≤γ,k＝1,...,n (2)
wherein T is the sampling period; in the inequality (2), the Rate is fixedTAnd the service rate (k) must not exceed the size of the buffer, gamma, in order to avoid overflow of the system buffer.
Preferably, when the utilization rate of the processor is lower than the schedulable upper limit δ, all the subtasks can be guaranteed to be completed smoothly, so that the real-time constraint of the system can be guaranteed when the condition (3) is met.
The energy consumption supervision method of the computer system has the following beneficial effects:
the invention provides a brand-new formalized problem description for the CPU operation mode distribution problem, establishes the CPU utilization rate and the energy consumption model thereof, and provides an operation mode distribution strategy (OMAS) based on a control theory, wherein the strategy adaptively adjusts the frequency of the CPU based on the CPU utilization rate and the user requirement. And under the condition of guaranteeing real-time scheduling, the energy consumption of the system is minimized.
Drawings
FIG. 1: the architecture of the present invention is schematically illustrated.
Detailed Description
The invention is further illustrated below with reference to fig. 1:
1.1 System model
A typical wireless embedded system consists of a plurality of wirelessly connected sensors, actuators and base stations. Assuming that the sensor sampling period signal is sent to the base station, the base station can process corresponding data and then activate the actuator to perform corresponding action. However, the workload of the base station varies with the amount of data received. In order to avoid data congestion, the base station is usually equipped with a data buffer for storing data that has not been processed since the data was obtained from the sensor. And the processor sequentially fetches the cached data after finishing the previous round of data processing.
FIG. 1 is an architectural depiction of a system, which has broader application. For example, in a surveillance sensing system consisting of a plurality of cameras, all the cameras periodically send pictures or videos to a processing center, and the processing center stores the pictures in a buffer and then sequentially identifies objects of interest and the like according to a pattern recognition algorithm such as image fusion and the like.
1.2CPU energy model
An effective way to save power consumption is to reduce the voltage and operating frequency of the processor when there are no tasks or the load is light. Assume that each node supports m discrete modes of operation. Each operating mode has a different energy consumption and execution time for the standard task load, and a corresponding CPU running frequency. For example, the Intel Xscale PXA270 processor can operate at 7 different frequencies, 13MHz, 104MHz, 208MHz, 312MHz, 416MHz, 520MHz, and 624 MHz. As shown in Table 1, when PAX270 is operated at different frequencies, it corresponds to different power consumption.
1.3 task model
Periodic tasks Ti are represented by triplets (r)i,ei,di) Where ri is the start time of the task Ti, ei is the execution time, and di is the completion time of the task. Each task hypothesis is composed of a series of subtasks Ti j|1≤j≤niThe subtasks are represented by a quadruplet
Wherein r isi jIs a subtask Ti jThe start time of (c) is,
is the default execution time of the execution,
is the relative deadline by which the subtask must be executed to complete,
is task Ti jThe workload (calculated amount). According to document [7 ]]The subtasks of a periodic task are also periodic tasks. Table 2 describes the meaning of all symbols in this chapter.
TABLE 2 symbol meaning table
The wireless embedded system works in an open environment, and the working mode of the CPU cannot be simply set to be a fixed mode. In order to avoid the delay of the task, a model for representing the system workload needs to be found, and the utilization rate of the CPU can directly reflect the condition of the system workload. Suppose that the CPU utilization can be calculated by the following equation [8 ]:
where nk is the number of processing tasks in the kth cycle, and es and ds are the execution time and relative deadline of the tasks, respectively. Generally, it is desirable that the utilization rate u (k) should not exceed 100% and should be greater than a set value δ (typically δ is set to 75%). The CPU operating mode allocation problem for power consumption optimization can be translated into a constrained optimization problem as follows:
given a CPU utilization setting δ (δ ≦ 100%), a buffer size γ, and a set of processor operating modes Mi＝{(ti,pi) I 1.. m }, where m is the number of operating modes, ti is the execution time of mode i, and pi is the energy consumption of mode i. And defining pki as the energy consumption of the mode i in the k-th period. The goal of the problem is then to assign the appropriate operating mode to the processor while ensuring that the total energy consumption of the system is minimized over n cycles, namely:
simultaneously, the constraint conditions are met:
(|Rate(k)-Rate|)·T≤γ,k＝1,...,n (2)
where T is the sampling period. In inequality (2), solidFixed Rate RateTAnd the service rate (k) must not exceed the size of the buffer, gamma, in order to avoid overflow of the system buffer. When the utilization rate of the processor is lower than the schedulable upper limit delta, all the subtasks can be guaranteed to be completed smoothly, so that the real-time constraint of the system can be guaranteed when the condition (3) is met.
The invention is described above with reference to the accompanying drawings, it is obvious that the implementation of the invention is not limited in the above manner, and it is within the scope of the invention to adopt various modifications of the inventive method concept and solution, or to apply the inventive concept and solution directly to other applications without modification.
Claims (6)
1. A method for energy consumption supervision of a computer system, comprising the steps of: the system consists of a plurality of sensors, actuators and base stations which are in wireless connection; the sensor sampling period signal is sent to a base station, the base station can process corresponding data, and then an actuator is excited to make corresponding action; the base station is usually provided with a data buffer area for storing data which is acquired from the sensor but is not processed; and the processor sequentially fetches the cached data after finishing the previous round of data processing.
2. The method of energy consumption supervision of a computer system according to claim 1, characterized by: the voltage and operating frequency of the processor are reduced when there is no task or the load is light.
3. The method of energy consumption supervision of a computer system according to claim 2, characterized by: assume that each node supports m discrete operating modes, each operating mode having a different energy consumption and execution time for a standard task load, and a corresponding CPU operating frequency.
4. The method of energy consumption supervision of a computer system according to claim 2, characterized by: in order to avoid the delay of the task, it is necessary to find a model characterizing the workload of the system,the utilization rate of the CPU can directly reflect the condition of the system workload, and it is assumed that the utilization rate of the CPU can be calculated by the following formula:
wherein, nk is the number of processing tasks in the kth period, and es and ds are the execution time and the relative deadline of the tasks respectively; the utilization rate u (k) is expected not to exceed 100% and should be greater than a set value δ.
5. The method of energy consumption supervision of a computer system according to claim 2, characterized by: given a CPU utilization setting δ (δ ≦ 100%), a buffer size γ, and a set of processor operating modes Mi＝{(ti,pi) 1.. m }, where m is the number of operating modes, ti is the execution time of mode i, and pi is the energy consumption of mode i; defining pki as the energy consumption of the mode i in the k period; the goal of the problem is then to assign the appropriate operating mode to the processor while ensuring that the total energy consumption of the system is minimized over n cycles, namely:
simultaneously, the constraint conditions are met:
(|Rate(k)-Rate|)·T≤γ,k＝1,...,n (2)
wherein T is the sampling period; in the inequality (2), the Rate is fixedTAnd the service rate (k) must not exceed the size of the buffer, gamma, in order to avoid overflow of the system buffer.
6. The method of energy consumption supervision of a computer system according to claim 5, characterized by: when the utilization rate of the processor is lower than the schedulable upper limit delta, all the subtasks can be guaranteed to be completed smoothly, so that the real-time constraint of the system can be guaranteed when the condition (3) is met.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910787213.6A CN112433595B (en) | 2019-08-25 | Energy consumption monitoring method for computer system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910787213.6A CN112433595B (en) | 2019-08-25 | Energy consumption monitoring method for computer system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112433595A true CN112433595A (en) | 2021-03-02 |
| CN112433595B CN112433595B (en) | 2024-04-30 |
Family
ID=
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101374140A (en) * | 2007-08-22 | 2009-02-25 | 湖南大学 | Node structure of wireless sensor network and MAC communication protocol thereof |
| CN103619056A (en) * | 2013-12-02 | 2014-03-05 | 华为终端有限公司 | Method and terminal for reporting sensor data |
| CN106900082A (en) * | 2017-02-04 | 2017-06-27 | 北京信息科技大学 | The data processing method and sensor network nodes of sensor network nodes |
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101374140A (en) * | 2007-08-22 | 2009-02-25 | 湖南大学 | Node structure of wireless sensor network and MAC communication protocol thereof |
| CN103619056A (en) * | 2013-12-02 | 2014-03-05 | 华为终端有限公司 | Method and terminal for reporting sensor data |
| CN106900082A (en) * | 2017-02-04 | 2017-06-27 | 北京信息科技大学 | The data processing method and sensor network nodes of sensor network nodes |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108958916B (en) | Workflow unloading optimization method under mobile edge environment | |
| CN106557369B (en) | Multithreading management method and system | |
| WO2021233261A1 (en) | Multi-task dynamic resource scheduling method | |
| Hu et al. | On exploring image resizing for optimizing criticality-based machine perception | |
| US10924995B2 (en) | Wake-up radio roaming | |
| CN104539972A (en) | Method and device for controlling video parallel decoding in multi-core processor | |
| JP2005182223A (en) | Information processor and operating system | |
| CN111580949B (en) | Automatic regulating method for network packet receiving mode | |
| CN112463293A (en) | Container-based expandable distributed double-queue dynamic allocation method in edge scene | |
| CN101561767A (en) | Method and device for executing tasks based on operating system | |
| CN112433595A (en) | Energy consumption supervision method of computer system | |
| JP2008146357A (en) | Programmable controller and support device thereof | |
| CN113434303A (en) | Batch-processed remote sensing image intelligent processing model prediction performance optimization system and method | |
| CN112433595B (en) | Energy consumption monitoring method for computer system | |
| US11360542B2 (en) | Method and apparatus for operating a processor in an electronic device | |
| CN109936471B (en) | Multi-cluster resource allocation method and device | |
| CN101349975B (en) | Method for implementing interrupt bottom semi-section mechanism in embedded operation system | |
| CN115686863A (en) | Hybrid polling method, device, equipment and readable storage medium | |
| Toma et al. | An energy-efficient middleware for computation offloading in real-time embedded systems | |
| WO2022253417A1 (en) | A computer software module arrangement, a circuitry arrangement, an arrangement and a method for improved autonomous adaptation of software monitoring of realtime systems | |
| CN113608845A (en) | Data processing method and device | |
| CN113037971A (en) | Method for multi-AI model parallel processing of multi-channel video | |
| CN110727513A (en) | Wireless embedded system energy consumption management method based on cybernetics | |
| CN113296941B (en) | Cache task scheduling method and device based on polygonal edge calculation | |
| CN104978006B (en) | A kind of low power idle under multithread mode waits method |
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 |