CN106484499A - A kind of cloud management platform towards new forms of energy data center extending transversely - Google Patents
A kind of cloud management platform towards new forms of energy data center extending transversely Download PDFInfo
- Publication number
- CN106484499A CN106484499A CN201510556261.6A CN201510556261A CN106484499A CN 106484499 A CN106484499 A CN 106484499A CN 201510556261 A CN201510556261 A CN 201510556261A CN 106484499 A CN106484499 A CN 106484499A
- Authority
- CN
- China
- Prior art keywords
- energy
- module
- cloud
- strategy
- data center
- 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
- 238000012544 monitoring process Methods 0.000 claims abstract description 35
- 230000005611 electricity Effects 0.000 claims abstract description 12
- 238000004146 energy storage Methods 0.000 claims abstract description 12
- 238000004519 manufacturing process Methods 0.000 claims abstract description 8
- 230000007423 decrease Effects 0.000 claims abstract description 3
- 238000007726 management method Methods 0.000 claims description 32
- 238000000034 method Methods 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 10
- 230000004048 modification Effects 0.000 claims description 7
- 238000012986 modification Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 230000004087 circulation Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 5
- 230000005059 dormancy Effects 0.000 claims description 5
- 230000000977 initiatory effect Effects 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000009471 action Effects 0.000 claims description 2
- 230000004907 flux Effects 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 239000003607 modifier Substances 0.000 claims description 2
- 230000006855 networking Effects 0.000 claims description 2
- 230000000737 periodic effect Effects 0.000 claims description 2
- 230000004044 response Effects 0.000 claims description 2
- 238000012937 correction Methods 0.000 claims 1
- 238000009826 distribution Methods 0.000 abstract description 8
- 238000012423 maintenance Methods 0.000 abstract description 2
- 239000008186 active pharmaceutical agent Substances 0.000 description 13
- 230000008859 change Effects 0.000 description 6
- 238000010248 power generation Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 231100000749 chronicity Toxicity 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses one kind is towards having distributed new generator unit and energy-storage units, and generation of electricity by new energy and energy-storage units have the cloud management Platform Designing of the data center of transverse elasticity telescope features, its structure includes the functional unit of the conventional cloud management platform software stack of data center deployment；With the power pack being integrated in disposed cloud management platform software stack.The software and hardware deployable state information at the central controller shared data center of described power pack and cloud management platform；Coordinate the hardware level energy relations of distribution and electrical energy production module and Cloud Server that configuration mixes in new forms of energy data center between by energy monitoring/configuration module；By carrying out, with cloud platform central controller, the software levels energy relations of distribution that information exchange is coordinated and configured the live load mixing new forms of energy data center.The present invention improves the energy use efficiency of mixing new forms of energy data center, decreases management software deployment time and operation maintenance cost.
Description
Technical field
The present invention relates to new forms of energy data center operational management skill field.More particularly, to one kind, there is distributed new energy
Source generator unit and energy-storage units, and generation of electricity by new energy unit has the cloud management platform of the data center of transverse elasticity telescope features
Design.
Background technology
Green calculates will become one of important technical change of 21st century facing mankind.Currently, computer manufacturing and letter
Breath technical industry just manages to tackle two stern challenges the most：High energy consumption issues and high pollution predicament.By 2010, meter
Calculation machine system energy consumption has accounted for the 3% of the total power consumption in the whole world.But global server market just has just enter into comprehensive expansionary phase.Faced by
The situation that world energy sources crisis increasingly sharpens, depending on unduly the design of very big restriction computer system and operation to traditional electric energy.
Remove the economic expense that high energy consumption is brought, relevant environment problem also becomes the debatable focus of information industry circle in recent years.According to
The investigation of McKinsey ＆ Co., in following 10 years, the carbon dioxide annual emissions that global computer causes indirectly will reach 1.54 hundred million tons,
Become one of polluter the biggest in the world.Therefore, improve non-fossil energy proportion, development intelligent grid and distributed power generation, encourage
Green low-carbon technology and recycling economy, the ability improving reply climate change becomes the important issue that currently green calculates.Current
The energy supply of data center has begun to trial and uses new forms of energy, such as the Fructus Mali pumilae of the U.S. and Google using solar energy and
Wind energy is providing the partly even all of energy demand of its data center.
But main regenerative resource, such as solar energy and wind energy, there is intermittent and undulatory property, this can affect
Operational reliability and disposition flexibility to the computational load of data center.Continuous accumulation with information age data volume and growth,
In order to realize the Effec-tive Function of new forms of energy data center, the operation management system of data center must take effective live load to adjust
The collaborative energy production of degree combined innovation manages and to meet all kinds of big data newly emerged in large numbers applications to calculating speed, availability, peace
The different demands of full property.Currently for the existing correlational study of energy load coordinated management of new forms of energy data center, for example, apply for public
Cloth number is the Chinese patent literature of CN103297505A, and the data center's multiple-energy-source disclosing under a kind of dynamic cloud service request is online
Control system, drives the live load with different delayed time tolerance by dispatching the multi-source energy supplying system of data center, reaches
Optimize the purpose of data center's energy supplying system long term overhead in terms.
The construction of current new forms of energy data center, in addition to overall newly-built full green or enlargement green data center, due to
Existing environment and the restriction of economic budget, also assume a kind of interim and modularity expansion feature.That is, to existing traditional data
The power supply infrastructure at center carry out the transformation of batch terre verte, the green energy resource generating equipment of add module and energy storage device.
Because the differentiation of this construction for electricity is interim, chronicity, this will lead to a data center feed framework to develop becomes
One kind of multiple construction for electricities and deposit and interstitial content dynamic change isomery construction for electricity.Isomery power supply stand in this dynamic change
Under structure, the existing energy load collaborative management method for isomorphism new forms of energy data center design is difficult to provide efficient conformable tube
Reason.Further, since user for multiple use motivations and may have different optimization aim in the use of newly-increased green energy resource,
The supporting energy supply infrastructure of the new forms of energy data center of dynamic heterogeneous power supply and management software need to provide flexible green energy resource to adjust
Degree means are to meet the demand of user.And how to provide for the new forms of energy data center of green energy resource energy supply unit dynamic change
A kind of efficient energy management method, can in unified coordination and administration new forms of energy data center number dynamically changeable energy supply section
Point, rationally map out the work load, and reduces human cost and the time cost of management software deployment, and prior art still fails
Solution is provided.
Cloud computing management platform is a kind of calculating solution proposing for current Internet data center.Current interconnection
Network data center is made up of a large amount of par commercial servers, and these commercial servers have different hardware parameters and runnability.
Cloud computing platform has broken the border between software and hardware, makes these isomeries by providing a kind of service organization's framework of loose coupling
Commercial server can be managed concentratedly so that each Service Component supports that the resource of management member elastic telescopic on demand takes in platform
The operation mode of business, it is achieved thereby that the dynamically distributes of IT resource and elasticity adjustment.Inspired by cloud computing normal form, worked as data
In when in the heart there is isomery power supply unit, by disposing service-specific assembly in data center's cloud management software, you can it is right to realize
The centralized integration of dynamic heterogeneous confession electric resources, by using the Unified Communication framework of cloud management platform interior, realizes to dynamic heterogeneous
Cooperating type elastic management for electric resources and software layer live load.
Content of the invention
Defect for prior art and Improvement requirement, the present invention proposes a kind of cloud management platform, is deployed in and is currently integrated with newly
The novel hybrid energy data center of energy distributed generation unit and energy-storage units, its object is to for mixing new forms of energy data
In characteristic that in the heart generation of electricity by new energy unit dynamic elasticity stretches provide a kind of convenient, efficient, extendible coordination generator unit to produce
The solution with live load deployment, the energy use efficiency of lifting mixing new forms of energy data center can be configured, minimizing management is soft
Part deployment time and operation maintenance cost.
For this reason, the present invention adopts the following technical scheme that：
First, commonly use the functional unit of cloud management platform software stack in data center deployment, comprise computation module, storage assembly,
Networking component etc..
Secondly, a kind of power pack of the compatible current cloud platform specification being integrated in disposed cloud management platform, described energy are proposed
The software and hardware deployable state information at the central controller shared data center of amount assembly and cloud management platform；Described power pack is passed through
Hard between energy monitoring/configuration module is coordinated and configuration mixes in new forms of energy data center electrical energy production module and Cloud Server
The part level energy relations of distribution；Described power pack is coordinated and configures to mix by carrying out information exchange with cloud platform central controller
The software levels energy relations of distribution of the live load of new forms of energy data center.
Described power pack as shown in Figure 1 as an ingredient of cloud management platform software stack (middleware layer),
Including：Power pack central controller, message queue, outside API, internal database and can elastic telescopic energy monitoring
Module and energy reversal module.
The energy hardware production status that the energy monitoring module of described power pack is used for monitoring new forms of energy data center is set with IT
Standby energy (Cloud Server, switch etc.) consumption state, it includes what one group of software was realized, can carry out extending transversely in a large number
Plant capacity monitors driver element.Plant capacity monitors the high flux central authorities control master that driver element is deployed in data center's cloud system
On machine or slave, wherein each plant capacity monitors the hardware sensing that driver element passes through the energy device corresponding to it of enclosed inside
The API that device is communicated, each generation of electricity by new energy unit, intelligence from new forms of energy data center by the way of regular visit
Can energy allocation unit or Cloud Server acquisition energy information.
In described power pack, all of plant capacity monitoring driver element is all by the message queue in power pack and central authorities
Controller is communicated, and this achieves the property extending transversely that plant capacity in power pack monitors driver element.When mixing new forms of energy
When the hardware device configuration of data center occurs increase and decrease, only need to dispose built-in new on cloud system central control host or slave
Increase the relevant device energy monitoring driver element of the CommAPI of hardware device, you can realize the data acquisition monitoring to adding hardware.
The energy reversal module of described power pack be used for configuration modification new forms of energy data center energy production distribution state with
The processor power consumption modes of Cloud Server, it includes what one group of software was realized, plant capacity configuration driven unit that can be extending transversely.
Each plant capacity monitors the energy corresponding to it that dispensing unit passes through enclosed inside when receiving the configuring request of energy reversal module
The API that the hardware sensor of equipment is communicated to each generation of electricity by new energy unit in new forms of energy data center, energy-storage units,
Intelligent Energy allocation unit or Cloud Server carry out corresponding energy state configuration.
Described power pack central controller, as the main functional modules of power pack, is responsible for focusing on monitoring mould from energy
Block and the software and hardware status information of cloud central controller acquisition, make system mode adjustment configuration decisions according to outside input strategy,
And send to energy reversal module and the execution of cloud central controller.
Different from the interactive mode of the central controller of cloud platform according to power pack central controller, the control of power pack central authorities
The mode of operation of device processed is divided into aggressive mode and Passive Mode.In aggressive mode, power pack is in the friendship with cloud central controller
Mutually middle as controlling main body, actively monitor in the hardware level status information of energy device in new forms of energy data center and cloud platform and work
The software levels status information of load, and durations are carried out to software and hardware system running status according to the optimisation strategy of user preset
Ground optimizes to be revised, and in the most backward data center, each energy hardware and cloud central controller send control instruction.In Passive Mode,
Cloud platform central controller obtains to power pack when initiating system configuration operation (as virtual machine starts or migrates) and is based on energy
The reference decision-making of perception, power pack with the interacting of cloud central controller as passive response unit.Due to having multithreading
Support, two kinds of mode of operations are simultaneously present in power pack, that is, power pack may any in active state when
Carve and process the Passive Mode request that cloud central controller sends.
As shown in Figure 2, described power pack central controller includes information acquisition module, strategy selection module, and strategy is held
Row module and configuration module.
Described information acquisition module is responsible for collecting the system hardware operation information collecting from each plant capacity monitoring driver element
The system operating software operation information collecting with cloud central controller, and update in power pack internal database.Each equipment
The energy collecting hardware configuration information is pushed to message queue, information acquisition module by energy monitoring driver element information periodically
Then read, from message queue, the data that some plant capacities monitor driver element on demand.Information acquisition module passes through outside API (such as
REST API) communicating with cloud central controller obtains required cloud platform configuration information.
Described strategy selection module is responsible for selecting the implementation strategy of power pack central controller, and the implementation strategy of selection is probably
The user preset energy scheduling strategy of outside input or last time policy enforcement module are run and are finished the energy scheduling strategy selecting to redirect.
The energy Scheduling Mechanisms Analysis information acquisition module that described policy enforcement module is responsible for according to strategy selection module selects is read
The data taking, and make the configuration decisions of data center's energy hardware and cloud live load, it is the Core Feature of whole power pack
Unit.There are under aggressive mode and Passive Mode different working methods.
Regulative strategy selecting module first reads user profile, according to the tune of user preset to aggressive mode thread upon actuation
Degree strategy enters loop cycle and runs, and different scheduling strategies has different cycle of operations.Scheduling strategy can be at any time
Switch, be divided into synchronous and asynchronous two kinds of switching modes.Synchro cut is：When a scheduling strategy cycle period terminates
When, policy enforcement module jumps directly to specify strategy to enter the subsequent cycle cycle by condition judgement or returns strategy selection module
Enter different scheduling strategy circulations by reading exterior arrangement file；Asynchronised handover mode is：Policy enforcement module is from outside
Slave interrupt interface receives interrupting information and stops current strategies operation circulation, and Access strategy selecting module selects next periodic scheduling plan in advance
Slightly.
In active work mode, power pack central controller is generally circulated with the fixed cycle and executes its workflow, such as attached
Shown in Fig. 3, a typical circulation workflow comprises the following steps：
Step S1：In the initial time in each cycle, strategy selection module selects this cycle scheduling strategy to be executed, can
To be the mesh that when reading the user preset scheduling strategy in this cycle or a upper end cycle from configuration file, policy enforcement module is specified
Mark jump strategy.After selecting this cycle implementation strategy, strategy selection module regulative strategy performing module enters this loop cycle.
Step S2：Invoked policy enforcement module initiates request of data to information acquisition module, including from each plant capacity
Energy hardware configuration information and the cloud platform software configuration information from the collection of cloud central controller that monitoring driver element collects.
Step S3：The system status information obtaining is inputted corresponding decision unit and iterative optimization unit by policy enforcement module,
Obtain the system optimization configuration output result after decision-making.
Step S4：System configuration after optimizing and input information are contrasted by policy enforcement module, and configuration module to the back-end is sent out
Go out corresponding energy device configuration and live load schedule information, and specify next possible cycle jump strategy.
Step S5：Policy enforcement module enters the dormancy armed state of fixing duration or jumps out this loop cycle.
According to the difference of specific strategy, the cycle of operation of policy enforcement module has different durations.At policy enforcement module
When dormancy armed state, can be waken up by the system mode exceptional instructions of information acquisition module asynchronous push at any time, make exception
Reason decision-making, and send abnormality processing action command to configuration module.
Passive Mode thread is in backstage listening state upon actuation, initiates computation module or storage assembly when listening to cloud platform
During configuring request, activate a Passive Mode handling process.This result is computation module or the storage assembly of this cloud platform
Configuring request provides the decision-making of Energy-aware.After Passive Mode processes and is activated, Passive Mode thread regulative strategy first selects
Module selects corresponding energy scheduling strategy, and then regulative strategy performing module executes corresponding scheduling strategy, the most backward cloud central authorities
Controller returns the result of decision.Different from active work mode, there is not asynchronous mode, only in the strategy switching in Passive Mode
Energy presence is changed configuration file the strategy switching leading to and the strategy that policy enforcement module is specified by user and is redirected switching.
As shown in Figure 4, in passive work mode, power pack once specifically responds to the request of cloud central controller
Flow process is as follows：
Step S1：Passive Mode thread is in backstage listening state upon actuation, initiates computation module when listening to cloud platform
Or during storage assembly configuring request, recalls information acquisition module obtains data center and cloud platform running state information.
Step S2：Passive Mode thread dispatching strategy selection module reads user profile, according to the scheduling of user preset
The scheduling strategy of this execution of policy selection that tactful or last time policy enforcement module is specified.
Step S3：Policy enforcement module is according to the system status information obtaining in Scheduling Mechanisms Analysis step S1, and makes cloud
Dispatching platforms decision-making.
Step S4：Policy enforcement module sends scheduling decision by configuration module to cloud central controller, and is given possible
Jump strategy.
Described configuration module obtains specific equipment state modification information from policy enforcement module, is joined using corresponding plant capacity
The API that putting driver element provides sends status modifier instruction after being packaged；The live load schedule information obtaining is passed through outside
API (as REST API) sends the configuration modification carrying out high in the clouds live load to cloud central controller.
Power pack is integrated in data center systems middleware layer there is following advantage：(1) dispose conveniently, with application software
The power pack that form exists has platform-neutral, can be deployed in multiple host operating systems, and energy monitoring module is using envelope
The API installing is communicated with hardware detection sensor.(2) highly versatile, the power pack being integrated in middleware layer uses general leading to
Letter API (as REST API), can be with conventional cloud platform management software Seamless integration- with conventional cloud platform management software communication.
Brief description
Fig. 1 is power pack structural representation；
Fig. 2 is power pack central controller structural representation；
Fig. 3 is power pack active work mode flow chart；
Fig. 4 is power pack passive work mode flow chart；
Fig. 5 is that a kind of deployment in new forms of energy data center of green power generation unit transverse elastic telescopic for the cloud management platform is shown
It is intended to.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, below in conjunction with drawings and Examples, to this
Invention is further elaborated.It should be appreciated that specific embodiment described herein is only in order to explain the present invention, not
For limiting the present invention.
As shown in Figure 5, cloud management Platform deployment proposed by the invention is stretched in gradual elasticity in a kind of green power generation unit
The new forms of energy data center of contracting, it includes tradition power supply infrastructure framework 501, diesel generation equipment 502, modular green
Energy generator unit 503, automatic transfer switch 504, energy-storage units 505, two-way inversion system 506, Intelligent Energy distribution is single
First 507 data center information technoloy equipments.Modularity green energy resource generator unit selects solar photovoltaic equipment in the present embodiment, by too
Sun can photovoltaic panel, MPPT DC charger, energy-storage battery group and photovoltaic two-way inverter composition.Each modularity green energy resource is sent out
Electric unit drives one group of (scale in the present embodiment is a standard rack capacity, i.e. 42U) server and other IT facility such as
Switch etc..No matter all information technoloy equipments in data center, obtain energy from green power generation equipment or conventional electric power generation equipment, all logical
Cross Intelligent Energy allocation unit and drive and be beneficial to energy monitoring module acquisition information technoloy equipment operation power consumption information.
Described new forms of energy data center deployment has the cloud management system based on cloud management system Openstack realization of increasing income, number
According in the minds of physical node be divided into control node, network node, memory node and calculate node.Wherein said energy bins
What Openstack standard feature assembly was followed in the design of part writes specification, and with Openstack functional unit shared communication API.
The central controller of described power pack is deployed in the control node of Openstack cluster, by be deployed in identical or
Energy monitoring/configuration module in extension control node carries out information exchange to coordinate and to configure the electricity of mixing new forms of energy data center
The hardware level energy relations of distribution between module and Cloud Server can be produced；Additionally by the cloud central authorities with Openstack cluster
Controller carries out information exchange to coordinate and to configure the software levels energy relations of distribution of the live load mixing new forms of energy data center.
The message queue of described power pack adopts the default information queue AMQP of Openstack system, when in power pack
Plant capacity monitoring/configuration driven unit use multinode deployment scheme when, using long-range mistake between the message queue on each node
Journey call method (RPC) realizes information transmission.
The universal external api interface of described power pack, using the REST API meeting Openstack standard, can transmit JSON
Information with XML format.
The energy monitoring module of described power pack monitors driver element by a series of plant capacities being deployed in control node
Obtain the energy information of each hardware device in new forms of energy data center.To modular green energy resource generator unit, tradition power supply base
Infrastructure framework, energy-storage units, Intelligent Energy allocation unit data center information technoloy equipment all has corresponding plant capacity monitoring to drive
Unit corresponds to.Modularity green energy resource generator unit is led to corresponding plant capacity monitoring driver element using EPA
Letter；Intelligent Energy allocation unit and the network switch are led to corresponding plant capacity monitoring driver element using snmp protocol
Letter；Server is communicated with corresponding plant capacity monitoring driver element using Ethernet protocol.It is different according to system scale,
Plant capacity monitoring driver element process can be deployed in one or more extension control nodes.Control node server is each with described
Using ethernet communication between energy hardware device.Each plant capacity monitoring driver element is periodically read from corresponding energy hardware
Take status information, and be pushed in message queue.When energy hardware device, the number of particularly new energy equipment increases
When subtracting, only need to monitor driver element in control node server disposition or the corresponding plant capacity of disabling, you can realize to cloud platform
Low overhead Energy-aware elastic management.
Described power pack central controller is embodied as the multithreading finger daemon in cloud platform control node in the present embodiment,
Finger daemon includes aggressive mode thread and the Passive Mode thread of power pack central controller.
Regulative strategy selecting module first reads user profile, according to the tune of user preset to aggressive mode thread upon actuation
Degree strategy enters loop cycle and runs, and different scheduling strategies has different cycle of operations.Scheduling strategy can be at any time
Switch, be divided into synchronous and asynchronous two ways, synchro cut is：At the end of a scheduling strategy cycle period,
Power pack central controller is adjudicated by condition or is read exterior arrangement file and enters different scheduling strategy circulations；Asynchronised handover
Mode is：Current scheduling strategy program section from external interrupt interface interrupting information and stops current operation reserve, enters in advance
Next scheduling strategy selected element runs new scheduling strategy.
When power pack is operated in aggressive mode, the specific workflow of a cycle period of central controller is as follows：
Step S1：Initiate in cycle period, power pack obtains data center first and cloud platform running state information updates
Internal database, specifically includes：Each plant capacity monitoring driver element cycle is pushed to the energy hardware configuration letter in message queue
Breath；Power pack central controller is obtained from the background data base of Openstack cluster by its outside REST api interface
Cloud platform application state information；It is noted that the thing of each assembly of cloud platform being pushed in message queue by cloud central controller
Part information can be updated to the internal database of power pack at any time.
Step S2：Strategy program section will be excellent to the decision unit of the system status information the obtaining current optimisation strategy of input and iteration
Change unit, obtain the system optimization configuration output result after decision-making.
Step S3：System optimization configuration output result and input information are contrasted by strategy program section, configuration module to the back-end
Send corresponding energy device configuration and live load dispatch command.
Step S4：Policy enforcement module terminates current strategies cycle period after the dormancy armed state entering fixing duration.
Passive Mode thread is in backstage listening state upon actuation, when computation module initiates configuring request with storage assembly,
Passive Mode thread passes through energy monitoring module and cloud central controller obtains data center and cloud platform running state information, and ties
Share the default scheduling strategy in family and be directed to the scheduling decision that the configuring request that computation module or storage assembly initiate provides Energy-aware.
When power pack is operated in Passive Mode, the once concrete responding process of the request to cloud central controller for the power pack
As follows：
Step S1：Passive Mode thread is in backstage listening state upon actuation, when Passive Mode thread listens to
During the configuring request of the computation module on Openstack message queue RabbitMQ and storage assembly, it obtains data center first
And cloud platform running state information update the internal database of power pack, specifically include：Each plant capacity monitors driver element
Cycle is pushed to the energy hardware configuration information in message queue；Power pack central controller is by its outside REST api interface
The cloud platform application state information obtaining from the background data base of Openstack cluster；It should be noted that by the control of cloud central authorities
The event information of each assembly of cloud platform that device processed is pushed in message queue can be updated to the inside of power pack at any time
Data base.
Step S2：Passive Mode thread dispatching strategy selection module reads user profile, according to the scheduling of user preset
Policy selection strategy program section.
Step S3：The system status information obtaining in invoked strategy program piecewise analysis step S1, and make cloud platform tune
Degree decision-making.
Step S4：Strategy program section cloud central controller sends scheduling decision.
Claims (9)
1. a kind of cloud management platform being deployed in novel hybrid energy data center is it is characterised in that include：
The functional unit of the conventional cloud management platform software stack of data center deployment, it includes：Computation module, storage assembly, networking component；
A kind of power pack of the compatible current cloud platform specification being integrated in disposed cloud management platform, it includes：Can the energy monitoring module of elastic telescopic and energy reversal module, message queue, outside API, internal database and power pack central controller.
2. novel hybrid energy data center according to claim 1 is it is characterised in that novel hybrid energy data center is integrated with modular new forms of energy distributed generation unit and energy-storage units, thus constituting isomery construction for electricity；Wherein modular new forms of energy distributed generation unit and energy-storage units assume the feature of interim dynamic expanding.
3. cloud management platform according to claim 1 is it is characterised in that the energy monitoring module of described power pack is used for monitoring the energy hardware production status of new forms of energy data center and information technoloy equipment energy (Cloud Server, switch etc.) consumption state；It includes what one group of software was realized, can carry out the plant capacity monitoring driver element of a large amount of horizontal extension；Plant capacity monitoring driver element is deployed on high flux central control host or the slave of data center's cloud system, wherein each plant capacity monitors the API that driver element is communicated by the hardware sensor of enclosed inside energy device corresponding to it, and by the way of regular visit, each generation of electricity by new energy unit from new forms of energy data center, Intelligent Energy allocation unit or Cloud Server obtain energy information.
4. cloud management platform according to claim 1, it is characterized in that, the all of plant capacity of described power pack is monitored driver element and is all communicated with central controller by the message queue in power pack it is achieved that plant capacity monitors the horizontal extension of driver element in power pack；When there is increase and decrease in the hardware device configuration mixing new forms of energy data center, the relevant device energy monitoring driver element of the CommAPI of built-in adding hardware equipment only need to be disposed on cloud system central control host or slave, you can realize the data acquisition monitoring to adding hardware.
5. cloud management platform according to claim 1, it is characterized in that, the energy production that the energy reversal module of described power pack is used for configuration modification new forms of energy data center distributes state and the processor power consumption modes of Cloud Server, it includes what one group of software was realized, can horizontal extension plant capacity configuration driven unit；The API that each plant capacity monitoring dispensing unit is communicated by the hardware sensor of the corresponding energy device of enclosed inside when receiving the configuring request of energy reversal module carries out corresponding energy state configuration to the generation of electricity by new energy unit in new forms of energy data center, energy-storage units, Intelligent Energy allocation unit or Cloud Server.
6. cloud management platform according to claim 1, it is characterized in that, the central controller of described power pack is as the main functional modules of power pack, it is responsible for focusing on the software and hardware status information obtaining from energy monitoring module and cloud central controller, system mode adjustment configuration decisions are made according to outside input strategy, and sends to energy reversal module and the execution of cloud central controller.
7. cloud management platform according to claim 1 is it is characterised in that the mode of operation of described power pack central controller is divided into aggressive mode and Passive Mode；
In aggressive mode, power pack with the interacting of cloud central controller as controlling main body, actively monitor the hardware level status information of energy device in new forms of energy data center and the software levels status information of live load in cloud platform, and correction is optimized with carrying out durations to software and hardware system running status according to the optimisation strategy of user preset, in the most backward data center, each energy hardware and cloud central controller send control instruction；
In Passive Mode, cloud platform central controller obtains reference decision-making based on Energy-aware when initiating system configuration operation to power pack, power pack with the interacting of cloud central controller as passive response unit；
Support due to having multithreading, two kinds of mode of operations are simultaneously present in power pack, that is, power pack may process the Passive Mode request that cloud central controller sends any time in active state.
8. cloud management platform according to claim 1 is it is characterised in that described power pack central controller includes:Information acquisition module, strategy selection module, policy enforcement module and configuration module；
Described information acquisition module is responsible for collecting and is monitored, from each plant capacity, the system operating software operation information that the system hardware operation information that collects of driver element and cloud central controller collect, and updates in power pack internal database；The energy collecting hardware configuration information is pushed to message queue by each plant capacity monitoring driver element information periodically, and information acquisition module then reads, from message queue, the data that some plant capacities monitor driver element on demand；Information acquisition module passes through the cloud platform configuration information that outside API (as RESTAPI) is communicated needed for acquisition with cloud central controller；
Described policy enforcement module is responsible for the data reading according to the energy Scheduling Mechanisms Analysis information acquisition module that strategy selection module selects, and makes the configuration decisions of data center's energy hardware and cloud live load, is the core function unit of whole power pack.
9. cloud management platform according to claim 1 is it is characterised in that described policy enforcement module has different working methods under aggressive mode and Passive Mode；
Regulative strategy selecting module first reads user profile to aggressive mode thread upon actuation, and the scheduling strategy according to user preset enters loop cycle and runs, and different scheduling strategies has different cycle of operations；Scheduling strategy can switch at any time, is divided into synchronous and asynchronous two kinds of switching modes；Synchro cut is：At the end of a scheduling strategy cycle period, policy enforcement module jumps directly to the specified tactful entrance subsequent cycle cycle by condition judgement or return strategy selection module passes through the different scheduling strategies of reading exterior arrangement file entrance and circulates；Asynchronised handover mode is：Policy enforcement module from external interrupt interface interrupting information and stops current strategies operation circulation, and Access strategy selecting module selects next periodic scheduling strategy in advance；In active work mode, power pack central controller is generally circulated with the fixed cycle and executes its workflow, and a typical circulation workflow comprises the following steps：
Step S1：In the initial time in each cycle, strategy selection module selects this cycle scheduling strategy to be executed it may be possible to target jump strategy that when reading the user preset scheduling strategy in this cycle or a upper end cycle from configuration file, policy enforcement module is specified；After selecting this cycle implementation strategy, strategy selection module regulative strategy performing module enters this loop cycle；
Step S2：Invoked policy enforcement module initiates request of data to information acquisition module, monitors, including from each plant capacity, energy hardware configuration information and the cloud platform software configuration information from the collection of cloud central controller that driver element collects；
Step S3：The system status information obtaining is inputted corresponding decision unit and iterative optimization unit by policy enforcement module, obtains the system optimization configuration output result after decision-making；
Step S4：System configuration after optimizing and input information are contrasted by policy enforcement module, and configuration module to the back-end sends corresponding energy device configuration and live load schedule information, and specifies next possible cycle jump strategy；
Step S5：Policy enforcement module enters the dormancy armed state of fixing duration or jumps out this loop cycle；
According to the difference of specific strategy, the cycle of operation of policy enforcement module has different durations；When policy enforcement module is in dormancy armed state, can be waken up by the system mode exceptional instructions of information acquisition module asynchronous push at any time, make abnormality processing decision-making, and send abnormality processing action command to configuration module；
Passive Mode thread is in backstage listening state upon actuation, when listening to cloud platform initiation computation module or storage assembly configuring request, activates a Passive Mode handling process；This result provides the decision-making of Energy-aware for the computation module of this cloud platform or storage assembly configuring request；After Passive Mode processes and is activated, Passive Mode thread regulative strategy selecting module first selects corresponding energy scheduling strategy, and then regulative strategy performing module executes corresponding scheduling strategy, and the most backward cloud central controller returns the result of decision；Different from active work mode, there is not asynchronous mode in the tactful switching in Passive Mode, is only possible to the strategy that presence is switched by the strategy that user's modification configuration file leads to and policy enforcement module is specified and redirects switching；
In passive work mode, the once concrete responding process of the request to cloud central controller for the power pack is as follows：
Step S1：Passive Mode thread is in backstage listening state upon actuation, and when listening to cloud platform initiation computation module or storage assembly configuring request, recalls information acquisition module obtains data center and cloud platform running state information；
Step S2：The scheduling strategy of this execution of policy selection that Passive Mode thread dispatching strategy selection module reading user profile, the scheduling strategy according to user preset or last time policy enforcement module are specified；
Step S3：Policy enforcement module is according to the system status information obtaining in Scheduling Mechanisms Analysis step S1, and makes cloud platform scheduling decision；
Step S4：Policy enforcement module sends scheduling decision by configuration module to cloud central controller, and provides possible jump strategy；
Described configuration module obtains specific equipment state modification information from policy enforcement module, sends status modifier instruction using the API that corresponding plant capacity configuration driven unit provides after being packaged；The live load schedule information of acquisition is sent, by outside API, the configuration modification carrying out high in the clouds live load to cloud central controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510556261.6A CN106484499A (en) | 2015-09-02 | 2015-09-02 | A kind of cloud management platform towards new forms of energy data center extending transversely |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510556261.6A CN106484499A (en) | 2015-09-02 | 2015-09-02 | A kind of cloud management platform towards new forms of energy data center extending transversely |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106484499A true CN106484499A (en) | 2017-03-08 |
Family
ID=58237919
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510556261.6A Pending CN106484499A (en) | 2015-09-02 | 2015-09-02 | A kind of cloud management platform towards new forms of energy data center extending transversely |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106484499A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107220087A (en) * | 2017-06-02 | 2017-09-29 | 郑州云海信息技术有限公司 | A kind of MDC data center management systems based on operating service |
| CN108173692A (en) * | 2017-12-28 | 2018-06-15 | 山东华软金盾软件股份有限公司 | It is a kind of based on the whole network equipment sensory perceptual system being actively and passively combined and cognitive method |
| CN109150729A (en) * | 2017-06-28 | 2019-01-04 | 中国移动通信有限公司研究院 | A kind of data forwarding controlling method, device, system, medium and calculate equipment |
| CN110554756A (en) * | 2019-09-03 | 2019-12-10 | 西安交通大学 | green cloud service method of green data center |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515161A (en) * | 2008-02-18 | 2009-08-26 | 国际商业机器公司 | Centralized energy management method and system |
| CN102244677A (en) * | 2011-06-16 | 2011-11-16 | 中国科学院深圳先进技术研究院 | Green energy Cloud computing method and system |
| CN103377084A (en) * | 2012-04-11 | 2013-10-30 | 李涛 | Renewable energy based green data center load scheduling method and device |
| CN103384272A (en) * | 2013-07-05 | 2013-11-06 | 华中科技大学 | Cloud service distributed data center system and load dispatching method thereof |
| KR20130132067A (en) * | 2012-05-25 | 2013-12-04 | 우도영 | National power system using new renewable energy and information infrastructure |
| CN104092303A (en) * | 2014-07-03 | 2014-10-08 | 内蒙古大学 | Wind-solar distributed power generation independent micro-grid energy management coordination control system and method |
-
2015
- 2015-09-02 CN CN201510556261.6A patent/CN106484499A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515161A (en) * | 2008-02-18 | 2009-08-26 | 国际商业机器公司 | Centralized energy management method and system |
| CN102244677A (en) * | 2011-06-16 | 2011-11-16 | 中国科学院深圳先进技术研究院 | Green energy Cloud computing method and system |
| CN103377084A (en) * | 2012-04-11 | 2013-10-30 | 李涛 | Renewable energy based green data center load scheduling method and device |
| KR20130132067A (en) * | 2012-05-25 | 2013-12-04 | 우도영 | National power system using new renewable energy and information infrastructure |
| CN103384272A (en) * | 2013-07-05 | 2013-11-06 | 华中科技大学 | Cloud service distributed data center system and load dispatching method thereof |
| CN104092303A (en) * | 2014-07-03 | 2014-10-08 | 内蒙古大学 | Wind-solar distributed power generation independent micro-grid energy management coordination control system and method |
Non-Patent Citations (1)
| Title |
|---|
| 邓维 等: "云计算数据中心的新能源应用：研究现状与趋势", 《计算机学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107220087A (en) * | 2017-06-02 | 2017-09-29 | 郑州云海信息技术有限公司 | A kind of MDC data center management systems based on operating service |
| CN109150729A (en) * | 2017-06-28 | 2019-01-04 | 中国移动通信有限公司研究院 | A kind of data forwarding controlling method, device, system, medium and calculate equipment |
| CN109150729B (en) * | 2017-06-28 | 2021-11-19 | 中国移动通信有限公司研究院 | Data forwarding control method, device, system, medium and computing equipment |
| CN108173692A (en) * | 2017-12-28 | 2018-06-15 | 山东华软金盾软件股份有限公司 | It is a kind of based on the whole network equipment sensory perceptual system being actively and passively combined and cognitive method |
| CN110554756A (en) * | 2019-09-03 | 2019-12-10 | 西安交通大学 | green cloud service method of green data center |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103345298B (en) | Method of data center energy saving system based on virtual IT resource distribution technology | |
| CN101777769A (en) | Multi-agent optimized coordination control method of electric network | |
| Akella et al. | Distributed power balancing for the FREEDM system | |
| CN106484499A (en) | A kind of cloud management platform towards new forms of energy data center extending transversely | |
| CN101958808A (en) | A cluster task scheduling manager serving multi-grid access | |
| CN206272232U (en) | A multi-agent system for microgrid energy management | |
| CN111654067A (en) | Novel control method and device for dispatching power distribution system based on transmission and distribution coordination | |
| CN106385024A (en) | Modular electric energy router device and implementation method thereof | |
| Zheng et al. | Multi-agent based control system for multi-microgrids | |
| CN116050754A (en) | A method and system for collaborative management and control of virtual power plants based on digital twins | |
| CN115313393A (en) | Cloud group-edge power grid layered hierarchical group control and group dispatching system | |
| CN105634127A (en) | SDN controller based energy management method | |
| CN113357083A (en) | Intelligent control system and method for wind generating set | |
| CN106483876A (en) | A kind of energy scheduling architecture of new forms of energy data center | |
| CN102752374B (en) | System and method for storing and accessing power utilization efficacy data | |
| CN105119271B (en) | Energy management and control system and the method for tree are allocated based on two-dimentional bidirectional energy | |
| CN105576649A (en) | Power utilization scheduling method and system for microgrid | |
| CN104298536A (en) | Dynamic frequency modulation and pressure adjustment technology based data center energy-saving dispatching method | |
| CN117239758A (en) | Source network charge storage system based on scheduling control and scheduling control method | |
| CN202906977U (en) | A storage access system of electric energy using efficiency data | |
| Meng et al. | Distributed grid intelligence for future microgrid with renewable sources and storage | |
| CN117060573A (en) | Multi-source data driving regional source storage sensing and low-carbon operation intelligent control system | |
| Ku et al. | Energy information collecting agent for IoT big data system | |
| CN107706982A (en) | The charging pile control system of collective and distributive type control | |
| CN206162121U (en) | Little load of power grid controller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170308 |
|
| WD01 | Invention patent application deemed withdrawn after publication |