CN117251043A - OSI-based dynamic power management method and system - Google Patents
OSI-based dynamic power management method and system Download PDFInfo
- Publication number
- CN117251043A CN117251043A CN202311084456.6A CN202311084456A CN117251043A CN 117251043 A CN117251043 A CN 117251043A CN 202311084456 A CN202311084456 A CN 202311084456A CN 117251043 A CN117251043 A CN 117251043A
- Authority
- CN
- China
- Prior art keywords
- server
- state
- ipmi
- openstack
- ironic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 claims abstract description 48
- 230000001960 triggered effect Effects 0.000 claims abstract description 4
- 108010028984 3-isopropylmalate dehydratase Proteins 0.000 claims abstract 25
- 238000012544 monitoring process Methods 0.000 claims description 47
- 238000005265 energy consumption Methods 0.000 claims description 26
- 230000008569 process Effects 0.000 claims description 16
- 230000002618 waking effect Effects 0.000 claims description 13
- 230000005059 dormancy Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 230000005012 migration Effects 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- GFRROZIJVHUSKZ-FXGMSQOLSA-N OS I Natural products C[C@@H]1O[C@@H](O[C@H]2[C@@H](O)[C@@H](CO)O[C@@H](OC[C@@H](O)[C@@H](O)[C@@H](O)CO)[C@@H]2NC(=O)C)[C@H](O)[C@H](O)[C@H]1O GFRROZIJVHUSKZ-FXGMSQOLSA-N 0.000 claims 1
- 238000004134 energy conservation Methods 0.000 abstract description 11
- 238000011161 development Methods 0.000 abstract description 5
- 239000002184 metal Substances 0.000 description 14
- 230000006870 function Effects 0.000 description 13
- 239000002609 medium Substances 0.000 description 9
- 230000003068 static effect Effects 0.000 description 8
- 230000008901 benefit Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 230000007613 environmental effect Effects 0.000 description 6
- 238000012423 maintenance Methods 0.000 description 5
- 230000009467 reduction Effects 0.000 description 5
- 239000002699 waste material Substances 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000036541 health Effects 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 229910052799 carbon Inorganic materials 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 230000007958 sleep Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000000593 degrading effect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000003090 exacerbative effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000012120 mounting media Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
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
- G06F1/3215—Monitoring of peripheral devices
-
- 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
- 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
- G06F1/3287—Power saving characterised by the action undertaken by switching off individual functional units in the computer system
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Power Sources (AREA)
Abstract
The invention provides a dynamic power management method and a system based on OSI, wherein the method comprises the following steps: power consumption is reduced in a data center in two ways: when the workload is light, using the OpenStack Ironic to identify the physical server in the idle state, and sending a request for closing the power supply of the physical server in the idle state; when the workload increases, the increase of the demand is perceived through an OpenStack Ironic program, and a request for starting the power supply of the physical server is triggered to meet the service requirement; the state of the physical servers is monitored using the IPMI protocol. According to the invention, the power state of the host is dynamically adjusted according to the workload demand, automatic management is realized through OpenStack Ironic, optimal utilization of resources and maximum energy conservation are realized, and the method has important significance for reducing the running cost, improving the energy efficiency of a data center and sustainable development.
Description
Technical Field
The invention relates to the technical field of data center power management, in particular to a server and equipment dynamic power management for realizing a data center by using OpenStack Ironic; and more particularly to a dynamic OSI-based power management method and system.
Background
At present, the power management of the data center is manually managed, no automatic power management means exists yet, and the manual power management mode can cause the following problems:
1. energy waste: without DPM (dynamic power management), all servers and devices may always operate with the highest performance, even during low load periods or idle periods. This situation can result in a significant waste of energy, increasing the energy consumption and operating costs of the data center.
2. The energy cost is high: without dynamically adjusting the power usage, the data center will continue to operate with high power consumption, resulting in an increase in energy costs. Such conditions can adversely affect the finance and operation of the data center.
3. Environmental impact: high energy consumption will result in more carbon dioxide emissions and other environmental impact, exacerbating the carbon footprint of the data center. Data centers have attracted increasing attention as high energy-consuming facilities, their impact on the environment.
4. Premature equipment failure: long-term operation with high power consumption can lead to premature failure of servers and devices, shortening device life. This situation increases equipment maintenance and replacement costs for the data center.
5. Waste of resources: without DPM, server and device resources in the data center are also wasted. Resources are not fully utilized, so that the resources of the data center are wasted.
6. Service instability: in the high-load period, due to unreasonable resource allocation, the data center can have the problem of unstable service, so that the user experience is affected and the service quality requirement is met.
Dynamic power management (DynamicPowerManagement, DPM) is a technique by which the power usage of a device is dynamically adjusted according to the workload and requirements of the device. Dynamic power management DPM involves a number of key techniques and methods including:
1. and (3) energy consumption monitoring: DPM needs to monitor the energy consumption of the device in real time to understand the current power consumption state and energy usage. The energy consumption monitoring can be realized by sensors, power metering equipment, hardware performance counters and the like.
2. Modeling state space: in DPM, the status of the device is critical, including the operating status of the device, resource utilization, power status, etc. Building a state space model may abstract the state of a device into a state space to facilitate dynamic power management.
3. Prediction and decision algorithm: DPM needs to be able to predict future workload and energy demands in order to make power management decisions in time. The prediction and decision algorithm may be based on historical data and real-time data, employing statistical methods or machine learning algorithms to make predictions and decisions.
4. Frequency modulation and voltage regulation techniques: an important approach in DPM is to reduce power consumption by adjusting the operating frequency (frequency modulation) and operating voltage (voltage regulation) of the device. Frequency modulation and voltage regulation techniques may achieve reduced power consumption without degrading performance.
5. And (3) power management control: DPM requires adjustment of the power state by a power management control device. The power management control may be implemented through a hardware interface or software control.
6. And (3) self-adaptive optimization: some DPM systems use adaptive optimization techniques, such as reinforcement learning algorithms, to implement intelligent power management decisions. The algorithms can automatically learn and adjust strategies to optimize energy consumption and performance according to actual conditions.
OSI (OpenStackIronic) is a component in the OpenStack cloud computing platform, dedicated to the management and orchestration of bare metal servers (baremeans). The bare metal server is a physical server without a virtualization layer, directly runs an operating system and an application program, is closer to the traditional physical server than the traditional virtual machine, and is suitable for application scenes with higher requirements on performance and resources.
Technical implications of openstack infrared include:
IronicConductor: the ironicoconductor is a core component of the Ironic, is responsible for communication and interaction with a physical bare metal server, manages the state, resources and configuration of the bare metal server, and is responsible for performing various operations such as deployment, restart, startup and shutdown, etc.
2. Driver (Driver): the Ironic enables interaction with bare metal servers of different types and vendors through drivers. The Ironic supports a variety of drivers, including IPMI, iLO, iDRAC, etc., that allow the Ironic to communicate with bare metal servers of different brands and models.
3. Deployment and management: ironic allows users to deploy and manage bare metal servers through the OpenStackDashBoard or OpenStackAPI. The user can define the configuration and specification of the bare metal server on the openstack horizontal interface, and then the ironiconductor performs corresponding operations according to the user's needs and configuration.
4. Physical resource management: the Ironic may manage physical resources including CPU, memory, disk, network, etc. resources of the bare metal server. The Ironic can acquire the hardware information of the physical server through a driver, and allocate and configure the resources according to the requirements of users in the deployment process.
5. Automation and orchestration: ironic allows the user to automate and orchestrate through the OpenStackHeat. A user can define the configuration and deployment flow of the bare metal server by using the Heat template, so that the automatic management and arrangement of the bare metal server are realized.
6. Safety and stability: the Ironic is designed and realized by taking security and stability into consideration, including measures such as access control, identity authentication and data protection on the bare metal server, so as to ensure the security and stability of the bare metal server.
The existing DPM domain solution is static threshold management, which is a simple Dynamic Power Management (DPM) technique, in which an administrator sets a fixed power consumption threshold. When the power consumption of the server or the device exceeds a set threshold, the DPM system automatically places the server or the device in a low power consumption mode (such as a sleep state), thereby reducing the energy consumption. When the load increases, the system will automatically wake up the server.
The technical implications of existing static threshold management include:
1. threshold setting: in static threshold management, an administrator first sets a fixed power consumption threshold according to the actual situation and requirements of the data center. This threshold is typically determined based on the average load or power consumption level of the data center. The threshold value is set by considering the actual situation of the data center and the resource utilization requirement.
2. And (3) energy consumption monitoring: in order to implement static threshold management, the data center needs to monitor the energy consumption of servers and devices in real time. This may be achieved by installing energy monitoring sensors on the servers and devices or using electricity metering devices. The monitoring system will collect power consumption data in real time and compare with a preset threshold.
3. Low power consumption mode: when the power consumption of the server or device exceeds a preset threshold, the DPM system automatically places it in a low power mode, such as a sleep state or an off state. Thus, the equipment can be prevented from running with high power consumption for a long time, and the energy consumption and the resource waste are reduced.
4. Automatic awakening: once the load increases or devices are needed, the DPM system automatically wakes up the dormant or shut down server, causing it to resume normal operation. This ensures that the services of the data center can respond to the needs of the user in a timely manner.
5. Restriction: while static threshold management is a simple and easy to implement DPM technique, it suffers from a lack of flexibility. The static threshold is fixed and cannot automatically adapt to changes in load and changes in demand. Therefore, optimal energy consumption and resource utilization are not achieved in a highly dynamic data center environment.
In summary, static threshold management is a simple but limited DPM solution, suitable for a data center environment with relatively stable load, and can reduce energy consumption by setting a reasonable threshold. However, in highly dynamic and complex data center environments, static threshold management fails to meet real-time energy consumption optimization requirements.
Disclosure of Invention
In view of this, the present invention aims to provide a dynamic power management method based on OSI, which uses openstack infrared as a main technical means of dynamic power management (DynamicPowerManagement, DPM), provides remote management of power to a physical server, monitors node status, supports multiple hardware drivers and function plugins, can be integrated with a virtualization platform, and realizes more efficient resource management and energy utilization, and dynamically manages the power status of the physical server according to workload demands, thereby realizing optimal utilization of energy conservation and resources, solving the above problems existing in the current data center power management, realizing energy conservation and consumption reduction, improving energy efficiency, prolonging equipment life, improving resource utilization, and providing stable services.
The principle of the present invention to implement Dynamic Power Management (DPM) is that when there are a large number of virtual machines and physical servers in a data center, the total power consumption is typically quite large. However, not all hosts are always required to run at full load. Some physical servers may be in an idle state when the workload is light, at which time power consumption may be reduced by Dynamic Power Management (DPM) of the invention. The data center may dynamically adjust the power usage of the devices to optimize energy consumption and resource utilization according to actual workload and demand.
The present invention primarily utilizes OSI (OpenStackIronic) to implement Dynamic Power Management (DPM), an openStackIronic is a project for bare metal (BareMetal) deployment and management, which allows users to run virtual machines directly on physical servers without having to install operating systems or virtualization software thereon. Openstack infrared provides management functions for power to the physical server, and supports communication with the server through IPMI (IntelligentPlatformManagementInterface) and other protocols, so that the power state of the server can be remotely managed, including on, off, restarting and the like.
The invention uses OpenStackIronic to realize Dynamic Power Management (DPM), and the OpenStackIronic has the following technical characteristics and advantages:
1. highly flexible: the OpenStackIronic supports a plurality of hardware drivers and function plug-ins, and can adapt to physical servers of different manufacturers and models. This allows the openstack infrared to be compatible with a variety of hardware, enabling power management on different types of servers. Meanwhile, the required functional plug-ins can be flexibly selected according to actual requirements so as to meet the requirements of different scenes. The OpenStackIronic support function plug-in can add additional power management functions such as timing dormancy and automatic awakening by using the function plug-in, so that the requirements of dynamic power management are better met.
2. Remote power management: openStackIronic supports remote power management with a physical server via IPMI or other protocols. This allows an administrator to remotely control the power state of the server, including turning on, off, restarting, etc. The remote power management function facilitates maintenance and management of the server and reduces the need for physical operations.
3. High availability: because the OpenStackIronic can be integrated with a virtualization platform (such as OpenStack Nova), the OpenStackIronic directly operates a virtual machine on a physical server without a virtualization layer, the virtual machine operated on the physical server can obtain higher performance and stability, the availability and stability of the virtual machine are improved, and the OpenStackIronic is suitable for application scenes with higher requirements on high availability and performance. And by integrating the virtualization platform, the virtual machine can be dynamically created or closed on the idle node, so that more flexible resource management and energy utilization are realized.
4. Energy saving and resource utilization: by virtue of the openstack infrared-based dynamic power management, the power state of the physical server can be dynamically managed according to workload requirements, and the state of the physical server node can be monitored, including whether it is in an operational state, an idle state, or a dormant state. By monitoring the state of the nodes, whether idle nodes are available for shutdown or dormancy can be judged, and when the workload is light, part of idle hosts can be shutdown or dormancy so as to reduce energy consumption. When the workload is increased, the host computer in a dormant or shutdown state can be awakened and placed in a state of being capable of running the virtual machine, so that service requirements are met, energy conservation and optimal utilization of resources can be realized, and overall power consumption and running cost are reduced.
The invention provides a dynamic power management method based on OSI (OpenStackIronic), which comprises the following steps:
in a data center, the openstack infrared implementation Dynamic Power Management (DPM) procedure can reduce power consumption in two ways:
when the workload is light, using OpenStackIronic to identify a physical server in an idle state, and sending a request for closing the power supply of the physical server in the idle state;
the physical server in the idle state may be in a dormant state or a closed state, and by turning off the power supply of the physical server in the idle state, the power consumption of the physical server can be reduced and the energy can be saved.
When the workload increases, the increase of the demand is perceived through an OpenStackIronic program, and a request for starting the power supply of the physical server is triggered to meet the service requirement;
triggering a request for turning on the power supply of the physical server involves waking up the physical server in a dormant state or an off state, and putting the woken-up physical server in a state in which the virtual machine can be run;
monitoring the state of the physical server using the IPMI (IntelligentPlatformManagementInterface) protocol; the IPMI protocol is a hardware management interface standard, implemented through an independent management chip (BMC) integrated on a server, and allows a system administrator to remotely monitor, manage and control server hardware through a network in the case that an operating system is not available; the BMC periodically collects sensor data, such as temperature, voltage and the like, an administrator can be remotely connected to the BMC through an IPMI protocol to execute various management operations, including inquiring hardware states, setting alarm thresholds, restarting a server and the like, so that the operation and maintenance efficiency of the data center and the usability of the system are improved;
Further, the method for monitoring the state of the physical server by using the IPMI protocol comprises the following steps:
s1, configuring IPMI information: configuring IPMI information of each physical server before using the IPMI monitoring server, wherein the IPMI information comprises an IP address, a user name and a password; the IPMI information is used for establishing IPMI connection with the server so as to acquire hardware information and state of the server;
s2, establishing IPMI connection: establishing connection between the OpenStackIronic and a physical server through an IPMI tool so as to inquire the state and information of the server; IPMI tools (e.g., ipmitool or OpenIPMI) are typically used to execute IPMI commands. The connection may be through a Local Area Network (LAN) connection or through an Out-of-band management channel (Out-of-Bandwidth).
S3, inquiring the state and information of the server: after establishing IPMI connection between OpenStackIronic and a physical server, inquiring monitoring data of state and information of the server by using an IPMI command through the OpenStackIronic; the state and the information of the server can help to judge the health condition and the working state of the server;
s4, analyzing monitoring data: analyzing the state of the server according to the monitoring data of the server queried by the OpenStackIronic by using the IPMI command; if the state of the server is abnormal or reaches a set threshold value, triggering corresponding operation;
S5, triggering operation: obtaining the overall resource utilization rate of the server according to the state of the analysis server, and triggering corresponding operation for optimizing the overall resource utilization rate of the server through OpenStackIronic;
s6, recording and reporting: and recording the IPMI query result of each server through OpenStackIronic, and generating a report according to the requirement. The report may be used to monitor the status and health of the server, as well as the assessment of energy conservation effects.
Further, the method for triggering the corresponding operation of optimizing the overall resource utilization rate of the server through the openstack ironic in the step S5 includes:
if the overall resource utilization rate of the server is too high, triggering to start more physical servers;
if the overall resource utilization rate of the server is too low, temporarily closing the idle physical server to achieve the energy-saving effect;
if the overall resource utilization of the server is less than 30%, then triggering migration of the virtual machine to other machines and then hibernating the unloaded server. So as to achieve the effect of energy saving.
Further, the method for waking up the physical server in the dormant state or the closed state comprises the following steps:
configuring a target server, and enabling a Wake-on-LAN function in BIOS or UEFI setting of the target server before using a scheme of waking up the server; meanwhile, related settings are started in an operating system of the server so that the server can still receive a wake-up signal in a dormant or closed state;
Obtaining the MAC address of a target server; the MAC address is a unique identifier of a Network Interface Card (NIC) of the server. The MAC address of the NIC may be found on the server, or the MAC address corresponding to the allocated IP address may be found on the network device; transmitting a Wake-up signal using a Wake-on-LAN tool or command; the wake-up signal is a specific ethernet packet, which contains the MAC address of the target server, and is used to wake up the server, and the packet is sent to the network subnet where the server is located through a Local Area Network (LAN);
broadcasting a wake-up signal in the local area network, checking the MAC address of all servers receiving the signal, and if the server is matched with the MAC address of the target server, responding to the wake-up signal and starting a wake-up process by the server;
the method for responding to the wake-up signal and starting the wake-up process by the server comprises the following steps: resolving the signal and triggering a startup process of the server, which may involve starting a BIOS, operating system, etc., restores the server from a dormant or off state to an available state.
It should be noted that the process of waking up the server may be affected by network setup, hardware configuration, and operating system. Ensuring that the network connection of the server is normal and that the network device allows the broadcast signal to pass through is an important factor in successfully waking up the server.
In general, the process of waking up a server involves sending a specific packet to match the server's MAC address using the Wake-on-LAN protocol and triggering the server's startup process to restore the server from a dormant or off state to an available state.
Further, the judging basis of the workload comprises:
when the resource utilization rate of the server is lower than 30% -40%, the workload is determined to be lighter;
when the server resource utilization exceeds 70% or 80%, then the workload is deemed to be high.
Further, the monitoring data of the state and information of the query server in the step S3 includes:
the power state, temperature, fan speed of the server are queried.
Further, the method for analyzing the state of the server in the step S4 includes:
monitoring a state of a physical server node using openstack infrared, the state comprising: whether in an operational state, an idle state, or a dormant state;
by monitoring the state of the node, whether the physical server of the node has an idle node for closing or dormancy is judged, so that the energy consumption is reduced.
The present invention also provides a OSI (OpenStackIronic) -based dynamic power management system for performing an OSI-based dynamic power management method as described above, comprising:
Configuring an IPMI information module: IPMI information for configuring each physical server before using the IPMI monitoring server, wherein the IPMI information comprises an IP address, a user name and a password; the IPMI information is used for establishing IPMI connection with the server so as to acquire hardware information and state of the server;
establishing an IPMI connection module: the OpenStackIronic is used for establishing connection with a physical server through an IPMI tool so as to inquire the state and information of the server;
querying a server state and information module: after establishing IPMI connection between OpenStackIronic and a physical server, inquiring monitoring data of state and information of the server by using an IPMI command through the OpenStackIronic;
and (3) analyzing and monitoring a data module: the monitoring data are used for analyzing the state of the server according to the monitoring data of the server queried by the OpenStackIronic by using the IPMI command;
triggering an operation module: the method comprises the steps of obtaining the overall resource utilization rate of a server according to the state of an analysis server, and triggering corresponding operation for optimizing the overall resource utilization rate of the server through OpenStackIronic;
recording and reporting module: the IPMI query result of each server is recorded through OpenStackIronic, and a report is generated according to the need.
The dynamic power management method reduces the requirement of manual management by automatically adjusting the power state of the physical server. The power state of each host is not manually interfered by an administrator, and the automatic management is realized through OpenStackIronic. Thus saving time and effort of the administrator, improving work efficiency and enabling the administrator to concentrate more on other important tasks.
The energy consumption and operating costs of data centers are often substantial. The dynamic power management method dynamically adjusts the power state of the physical server according to the actual work load demand, can realize energy conservation and energy consumption reduction, and the saved energy cost is directly converted into the running cost saving of the data center.
Energy conservation and environmental protection are important issues facing today's society. The dynamic power management method can reduce carbon emission and energy consumption of the data center and reduce negative influence on the environment, thereby generating positive social benefit, being beneficial to promoting the development of the green data center and having great significance for realizing the sustainable development target.
The present invention also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor implements an OSI-based dynamic power management method as described above.
The present invention also provides a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed implements the OSI-based dynamic power management method as described above.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, based on the OpenStackIronic, dynamic Power Management (DPM) can dynamically adjust an energy-saving strategy and a virtual work load integration strategy by monitoring and analyzing the system state in real time, so that the optimal scheduling and dynamic allocation of resources are realized, the data center can more efficiently utilize hardware resources, and the stability and performance of the system are improved; by automatic energy-saving and load balancing operation, the manual intervention requirement is reduced, and the management efficiency and accuracy are improved, so that the overall working efficiency is improved;
the dynamic power management method reduces energy consumption and optimizes resource utilization, and brings remarkable cost saving to a data center; the energy-saving strategies such as power supply mode, processor frequency and voltage are dynamically adjusted, so that the system power consumption is minimized, and the expenditure of energy cost is reduced; meanwhile, through dynamic adjustment and load balancing of the virtual machine workload, reasonable allocation and utilization of resources are realized, idle waste of hardware equipment is reduced, and hardware investment and operation and maintenance cost are reduced.
The application of the dynamic power management method not only brings benefits to the data center, but also has positive influence on the whole society. Firstly, energy consumption and carbon emission are reduced through energy conservation and consumption reduction, the requirements of energy conservation and emission reduction and environmental protection are met, and the construction and sustainable development of a green data center are promoted. And secondly, through the optimal utilization of resources and the saving of cost, the consumption of natural resources is reduced, and the sustainable utilization of the resources and the environmental protection are facilitated. In addition, by improving the working efficiency and performance, high-availability and responsive services are provided for users, and user experience and satisfaction are improved.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention.
In the drawings:
FIG. 1 is a flowchart illustrating a method for monitoring the status of a physical server using the IPMI protocol according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a DPM system in accordance with an embodiment of the present invention;
FIG. 3 is a flowchart illustrating the operation of an OpenStackIronic implementation of Dynamic Power Management (DPM) in accordance with an embodiment of the present invention;
Fig. 4 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and products consistent with some aspects of the disclosure as detailed in the appended claims.
The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
Embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
The embodiment of the invention provides a dynamic power management method based on OSI, which comprises the following steps:
in a data center, the openstack infrared implementation Dynamic Power Management (DPM) procedure reduces power consumption in two ways:
when the workload is light, using OpenStackIronic to identify a physical server in an idle state, and sending a request for closing the power supply of the physical server in the idle state;
the physical server in the idle state may be in a dormant state or a closed state, and by turning off the power supply of the physical server in the idle state, the power consumption of the physical server can be reduced and the energy can be saved.
When the workload increases, the increase of the demand is perceived through an OpenStackIronic program, and a request for starting the power supply of the physical server is triggered to meet the service requirement;
the judging basis of the workload comprises the following steps:
when the resource utilization rate of the server is lower than 30% -40%, the workload is determined to be lighter;
when the server resource utilization exceeds 70% or 80%, then the workload is deemed to be high.
Triggering a request for turning on the power supply of the physical server involves waking up the physical server in a dormant state or an off state, and putting the woken-up physical server in a state in which the virtual machine can be run;
Monitoring the state of the physical server using the IPMI (IntelligentPlatformManagementInterface) protocol; the IPMI protocol is a hardware management interface standard, implemented through an independent management chip (BMC) integrated on a server, and allows a system administrator to remotely monitor, manage and control server hardware through a network in the case that an operating system is not available; the BMC periodically collects sensor data, such as temperature, voltage and the like, an administrator can be remotely connected to the BMC through an IPMI protocol to execute various management operations, including inquiring hardware states, setting alarm thresholds, restarting a server and the like, so that the operation and maintenance efficiency of the data center and the usability of the system are improved;
the method for monitoring the state of the physical server by using the IPMI protocol is shown in fig. 1, and comprises the following steps:
s1, configuring IPMI information of each physical server before using the IPMI monitoring server, wherein the IPMI information comprises an IP address, a user name and a password; the IPMI information is used for establishing IPMI connection with the server so as to acquire hardware information and state of the server;
s2, establishing connection between the OpenStackIronic and a physical server through an IPMI tool so as to inquire the state and information of the server;
IPMI tools (e.g., ipmitool or OpenIPMI) are typically used to execute IPMI commands. The connection may be through a Local Area Network (LAN) connection or through an Out-of-band management channel (Out-of-Bandwidth).
S3, after establishing IPMI connection between the OpenStackIronic and the physical server, inquiring monitoring data of the state and information of the server by using an IPMI command through the OpenStackIronic;
the monitoring data of the state and information of the query server in the step S3 comprises:
the power state, temperature, fan speed of the server are queried.
The state and information of the server may help determine the health and operational status of the server.
S4, analyzing the state of the server according to the monitoring data of the server queried by the OpenStackIronic by using the IPMI command;
the method for analyzing the state of the server comprises the following steps:
monitoring a state of a physical server node using openstack infrared, the state comprising: whether in an operational state, an idle state, or a dormant state;
by monitoring the state of the node, whether the physical server of the node has an idle node for closing or dormancy is judged, so that the energy consumption is reduced.
If the state of the server is found to be abnormal or the set threshold value is reached, triggering corresponding operation.
S5, obtaining the overall resource utilization rate of the server according to the state of the analysis server, and triggering corresponding operation for optimizing the overall resource utilization rate of the server through OpenStack Ironic;
the method for triggering the corresponding operation for optimizing the overall resource utilization rate of the server through OpenStackIronic comprises the following steps:
if the overall resource utilization rate of the server is too high, triggering to start more physical servers;
if the overall resource utilization rate of the server is too low, temporarily closing the idle physical server to achieve the energy-saving effect;
if the overall resource utilization of the server is less than 30%, then triggering migration of the virtual machine to other machines and then hibernating the unloaded server. So as to achieve the effect of energy saving.
S6, recording the IPMI query result of each server through the OpenStackIronic, and generating a report according to the need.
The report may be used to monitor the status and health of the server, as well as the assessment of energy conservation effects.
The method for waking up the physical server in the dormant state or the closed state comprises the following steps:
configuring a target server, and enabling a Wake-on-LAN function in BIOS or UEFI setting of the target server before using a scheme of waking up the server; meanwhile, related settings are started in an operating system of the server so that the server can still receive a wake-up signal in a dormant or closed state;
Obtaining the MAC address of a target server; the MAC address is a unique identifier of a Network Interface Card (NIC) of the server. The MAC address of the NIC may be found on the server, or the MAC address corresponding to the allocated IP address may be found on the network device; transmitting a Wake-up signal using a Wake-on-LAN tool or command; the wake-up signal is a specific ethernet packet, which contains the MAC address of the target server, and is used to wake up the server, and the packet is sent to the network subnet where the server is located through a Local Area Network (LAN);
broadcasting a wake-up signal in the local area network, checking the MAC address of all servers receiving the signal, and if the server is matched with the MAC address of the target server, responding to the wake-up signal and starting a wake-up process by the server;
the method for responding to the wake-up signal and starting the wake-up process by the server comprises the following steps: resolving the signal and triggering a startup process of the server, which may involve starting a BIOS, operating system, etc., restores the server from a dormant or off state to an available state.
It should be noted that the process of waking up the server may be affected by network setup, hardware configuration, and operating system. Ensuring that the network connection of the server is normal and that the network device allows the broadcast signal to pass through is an important factor in successfully waking up the server.
In general, the process of waking up a server involves sending a specific packet to match the server's MAC address using the Wake-on-LAN protocol and triggering the server's startup process to restore the server from a dormant or off state to an available state.
Referring to fig. 2, a DPM system technical architecture diagram of the present embodiment is shown;
referring to fig. 3, a flowchart of the operation of the openstack infrared implementation Dynamic Power Management (DPM) of the present embodiment is shown.
The embodiment of the invention also provides a dynamic power management system based on OSI (OpenStackIronic), which executes the OSI-based dynamic power management method, comprising the following steps:
configuring an IPMI information module: IPMI information for configuring each physical server before using the IPMI monitoring server, wherein the IPMI information comprises an IP address, a user name and a password; the IPMI information is used for establishing IPMI connection with the server so as to acquire hardware information and state of the server;
establishing an IPMI connection module: the OpenStackIronic is used for establishing connection with a physical server through an IPMI tool so as to inquire the state and information of the server;
querying a server state and information module: after establishing IPMI connection between OpenStackIronic and a physical server, inquiring monitoring data of state and information of the server by using an IPMI command through the OpenStackIronic;
And (3) analyzing and monitoring a data module: the monitoring data are used for analyzing the state of the server according to the monitoring data of the server queried by the OpenStackIronic by using the IPMI command;
triggering an operation module: the method comprises the steps of obtaining the overall resource utilization rate of a server according to the state of an analysis server, and triggering corresponding operation for optimizing the overall resource utilization rate of the server through OpenStackIronic;
recording and reporting module: the IPMI query result of each server is recorded through OpenStackIronic, and a report is generated according to the need.
The dynamic power management method of the embodiments of the present invention requires appropriate hardware and software support to implement the DPM program. The physical server needs to support power management functions such as support for remote Wake-on-LAN (Wake-on-LAN) or power control interfaces. In addition, the data center management system or virtualization management platform needs to be integrated with the DPM program to automate power management operations.
By using a DPM program, the data center can dynamically manage the power state of the physical servers according to actual workload requirements. This will bring about several advantages. First, overall power consumption is reduced, thereby reducing energy consumption and operating costs. And secondly, the service lives of the physical servers and the equipment are prolonged, and the risk of hardware faults is reduced. Thirdly, the energy efficiency and the environmental sustainability of the data center are improved, and the energy conservation and emission reduction are positively influenced.
Openstack infrared implementation Dynamic Power Management (DPM) is an efficient method of achieving power savings in a data center with a large number of virtual machines and physical servers. The power state of the host is dynamically adjusted according to the workload demand, so that the optimal utilization of resources and the maximum saving of energy are realized. This is of great importance for reducing operating costs, improving energy efficiency of data centers and sustainable development.
The embodiment of the invention also provides a computer device, and fig. 4 is a schematic structural diagram of the computer device provided by the embodiment of the invention; referring to fig. 4 of the drawings, the computer apparatus includes: input means 23, output means 24, memory 22 and processor 21; the memory 22 is configured to store one or more programs; when the one or more programs are executed by the one or more processors 21, the one or more processors 21 are caused to implement the OSI-based dynamic power management method as provided by the above-described embodiments; wherein the input device 23, the output device 24, the memory 22 and the processor 21 may be connected by a bus or otherwise, for example in fig. 4.
The memory 22 is used as a readable storage medium of a computing device and can be used for storing software programs and computer executable programs, and is used for storing program instructions corresponding to the dynamic power management method based on OSI according to the embodiment of the invention; the memory 22 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for functions; the storage data area may store data created according to the use of the device, etc.; in addition, memory 22 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device; in some examples, memory 22 may further comprise memory located remotely from processor 21, which may be connected to the device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input means 23 is operable to receive input numeric or character information and to generate key signal inputs relating to user settings and function control of the device; the output device 24 may include a display device such as a display screen.
The processor 21 executes various functional applications of the device and data processing by running software programs, instructions and modules stored in the memory 22, i.e., implements the OSI-based dynamic power management method described above.
The above-mentioned provided computer device can be used for executing the OSI-based dynamic power management method provided in the above-mentioned embodiment, and has corresponding functions and beneficial effects.
Embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are for performing the openstack infrared-based dynamic power management method as provided by the above embodiments, the storage medium being any of various types of memory devices or storage devices, the storage medium comprising: mounting media such as CD-ROM, floppy disk or tape devices; computer system memory or random access memory, such as DRAM, DDRRAM, SRAM, EDORAM, rambus (Rambus) RAM, etc.; nonvolatile memory such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc.; the storage medium may also include other types of memory or combinations thereof; in addition, the storage medium may be located in a first computer system in which the program is executed, or may be located in a second, different computer system, the second computer system being connected to the first computer system through a network (such as the internet); the second computer system may provide program instructions to the first computer for execution. Storage media includes two or more storage media that may reside in different locations (e.g., in different computer systems connected by a network). The storage medium may store program instructions (e.g., embodied as a computer program) executable by one or more processors.
Of course, the storage medium containing the computer executable instructions provided in the embodiments of the present invention is not limited to the OSI-based dynamic power management method described in the foregoing embodiments, but may also perform the relevant operations in the OSI-based dynamic power management method provided in any embodiment of the present invention.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
The foregoing description is only of the preferred embodiments of the invention and is not intended to limit the invention; various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for OSI-based dynamic power management comprising:
power consumption is reduced in a data center by two ways:
when the workload is light, using the OpenStack Ironic to identify the physical server in the idle state, and sending a request for closing the power supply of the physical server in the idle state;
when the workload increases, the increase of the demand is perceived through an OpenStack Ironic program, and a request for starting the power supply of the physical server is triggered to meet the service requirement;
the state of the physical servers is monitored using the IPMI protocol.
2. The OS I-based dynamic power management method of claim 1, wherein the method of monitoring the state of a physical server using an IPMI protocol comprises the steps of:
s1, configuring IPMI information of each physical server before using the IPMI monitoring server, wherein the IPMI information comprises an IP address, a user name and a password; the IPMI information is used for establishing IPMI connection with the server so as to acquire hardware information and state of the server;
s2, establishing connection between the OpenStack Ironic and a physical server through an IPMI tool so as to inquire the state and information of the server;
s3, after establishing IPMI connection between the OpenStack Ironic and the physical server, inquiring monitoring data of the state and information of the server by using an IPMI command through the OpenStack Ironic;
S4, analyzing the state of the server according to the monitoring data of the server queried by the OpenStack Ironic by using the IPMI command;
s5, obtaining the overall resource utilization rate of the server according to the state of the analysis server, and triggering corresponding operation for optimizing the overall resource utilization rate of the server through OpenStack Ironic;
s6, recording the IPMI query result of each server through the OpenStack Ironic, and generating a report according to the need.
3. The OSI-based dynamic power management method according to claim 2, wherein said method of step S5 of triggering a corresponding operation to optimize the overall resource usage of said server by OpenStack ionic comprises:
if the overall resource utilization rate of the server is too high, triggering to start more physical servers;
if the overall resource utilization rate of the server is too low, temporarily closing the idle physical server;
if the overall resource utilization of the server is less than 30%, then triggering migration of the virtual machine to other machines and then hibernating the unloaded server.
4. The OSI-based dynamic power management method of claim 1, wherein the method of waking up a physical server in a dormant state or an off state comprises:
Configuring a target server, and enabling a Wake-on-LAN function in BIOS or UEFI setting of the target server; meanwhile, related settings are started in an operating system of the server so that the server can still receive a wake-up signal in a dormant or closed state;
obtaining the MAC address of a target server; transmitting a Wake-up signal using a Wake-on-LAN tool or command;
broadcasting a wake-up signal in the local area network, checking the MAC address of all servers receiving the signal, and if the server is matched with the MAC address of the target server, responding to the wake-up signal and starting a wake-up process by the server;
the method for responding to the wake-up signal and starting the wake-up process by the server comprises the following steps: and analyzing the signal and triggering the starting process of the server to restore the server to the available state from the dormant or closed state.
5. The OSI-based dynamic power management method according to claim 1, wherein said workload determination basis comprises:
when the resource utilization rate of the server is lower than 30% -40%, the workload is determined to be lighter;
when the server resource utilization exceeds 70% or 80%, then the workload is deemed to be high.
6. The OSI-based dynamic power management method according to claim 2, wherein the monitoring data of the state and information of the query server of step S3 includes:
The power state, temperature, fan speed of the server are queried.
7. The OSI-based dynamic power management method according to claim 2, wherein the method of analyzing the state of the server of step S4 comprises:
monitoring a state of a physical server node using an OpenStack infrared, the state comprising: whether in an operational state, an idle state, or a dormant state;
by monitoring the state of the node, whether the physical server of the node has an idle node for closing or dormancy is judged, so that the energy consumption is reduced.
8. An OSI-based dynamic power management system, wherein performing the OSI-based dynamic power management method in accordance with claim 2 comprises:
configuring an IPMI information module: IPMI information for configuring each physical server before using the IPMI monitoring server, wherein the IPMI information comprises an IP address, a user name and a password; the IPMI information is used for establishing IPMI connection with the server so as to acquire hardware information and state of the server;
establishing an IPMI connection module: the OpenStack Ironic is used for establishing connection with a physical server through an IPMI tool so as to inquire the state and information of the server;
querying a server state and information module: after establishing IPMI connection between the OpenStack Ironic and the physical server, inquiring monitoring data of the state and information of the server by using an IPMI command through the OpenStack Ironic;
And (3) analyzing and monitoring a data module: the monitoring data are used for analyzing the state of the server according to the monitoring data of the server which are inquired by the OpenStack Ironic through the IPMI command;
triggering an operation module: the method comprises the steps of obtaining the overall resource utilization rate of a server according to the state of an analysis server, and triggering corresponding operation for optimizing the overall resource utilization rate of the server through OpenStack Ironic;
recording and reporting module: the IPMI query result of each server is recorded through the OpenStack Ironic, and a report is generated according to the need.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the OSI-based dynamic power management method of any one of claims 1-7.
10. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the OSI-based dynamic power management method of any one of claims 1-7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311084456.6A CN117251043A (en) | 2023-08-28 | 2023-08-28 | OSI-based dynamic power management method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311084456.6A CN117251043A (en) | 2023-08-28 | 2023-08-28 | OSI-based dynamic power management method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117251043A true CN117251043A (en) | 2023-12-19 |
Family
ID=89128497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311084456.6A Pending CN117251043A (en) | 2023-08-28 | 2023-08-28 | OSI-based dynamic power management method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117251043A (en) |
-
2023
- 2023-08-28 CN CN202311084456.6A patent/CN117251043A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10613900B2 (en) | Multi-tenant monitoring | |
| US8473768B2 (en) | Power control apparatus and method for cluster system | |
| US8095814B2 (en) | Computing device power management | |
| US8250382B2 (en) | Power control of servers using advanced configuration and power interface (ACPI) states | |
| US8271818B2 (en) | Managing under-utilized resources in a computer | |
| Femal et al. | Boosting data center performance through non-uniform power allocation | |
| US20080089338A1 (en) | Methods for remotely creating and managing virtual machines | |
| CN103188277A (en) | Load energy consumption management system, method and server | |
| US9411344B2 (en) | Energy saving monitoring method and device | |
| TWI551081B (en) | Remote server managing system and remote server managing method | |
| CN103701889A (en) | Data center energy saving method on basis of cloud computing | |
| CN101661327A (en) | Method for adjusting basic frequency of central processing unit (CPU) and device thereof | |
| WO2013127151A1 (en) | Power consumption capping control method, device and system | |
| CN104270430A (en) | Server remote dispatching method suitable for cloud computing | |
| WO2021078144A1 (en) | Power management method and device | |
| CN111147565A (en) | Cluster node control method, device and equipment and readable storage medium | |
| US9712647B2 (en) | Optimization of computing resources through monitoring and manipulating availabilty | |
| KR20100039140A (en) | Computer power saving method | |
| CN113986000A (en) | Server power-on method, system, device and medium | |
| CN114327023B (en) | Energy saving method, system, computer medium and electronic equipment of Kubernetes cluster | |
| CN114741180A (en) | Rack management system, method and controller | |
| CN117251043A (en) | OSI-based dynamic power management method and system | |
| CN215867831U (en) | Dormancy awakening device and NAS storage system | |
| CN111246549A (en) | Method and device for providing node dormancy and wakeup time | |
| Li et al. | An efficient approach for reducing power consumption in a production-run cluster |
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 |