CN103107954A - Scheduling method based on green-energy-aware - Google Patents
Scheduling method based on green-energy-aware Download PDFInfo
- Publication number
- CN103107954A CN103107954A CN2013100459111A CN201310045911A CN103107954A CN 103107954 A CN103107954 A CN 103107954A CN 2013100459111 A CN2013100459111 A CN 2013100459111A CN 201310045911 A CN201310045911 A CN 201310045911A CN 103107954 A CN103107954 A CN 103107954A
- Authority
- CN
- China
- Prior art keywords
- task
- data center
- level
- solar energy
- green
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A scheduling method based on green-energy-aware uses solar energy as green energy and conducts optimizing and distributing on works reaching a data center. The scheduling method based on green-energy-aware comprises the following steps: predicting according to illuminated scope and weather conditions to acquire solar energy which can be used by the data center; preprocessing the works reaching to the data center, each work comprising a plurality of tasks, communication constraint relationships existing among all tasks, conducting grading on works needing to schedule according to the communication constraint relationships in works, and gradually distributing tasks without communication constraint relationships to every grade; under requirement of meeting work time limit, conducting scheduling on preprocessed work assembly according to grades, sequentially distributing task of every grade to every server, fully considering loss conditions of the servers and communication network in the distributing process, and aiming at achieving highest solar energy use ratio. The scheduling method based on green-energy-aware can be applied to scheduling of the data center of random dynamic, achieve the effect that the solar energy use ratio is highest and cost of the data center and carbonic emission are lowered.
Description
Technical field
The present invention relates to utilize in data center the dispatching method of solar energy, particularly a kind of dispatching method based on the green energy resource perception, belong to Internet technical field.
Background technology
Along with developing rapidly of Internet technology, for provide to the user fast, high-quality and safety service, data center becomes more and more important as service provider.At present, the data center that all has oneself of the enterprise of major company of many Internet-baseds (as Google, Amazon, Yahoo etc.) processes service request from all over the world in the world.The power consumption of data center is very large according to statistics, accounts for 1.3% of global power consumption total amount, and estimates will rise to 8% in this ratio of the year two thousand twenty.And wherein the electric power of most of data center all by the supply of coal electricity, causes serious pollution thereby the burning of coal meeting produces a large amount of carbon containing emissions to environment.Green energy resource is widely used as a kind of novel energy, and its advantage is cheaply, clean and be conducive to continuable long-range use.But compare with traditional colliery, oil equal energy source, the characteristics of green energy resource are: (1) is intermittent strong.The generation of green energy resource will follow the course of nature, so can not continue supply energy source, especially solar energy, wind energy are the most obvious.(2) predictability is poor.Although solar energy has certain rule every day, Accurate Prediction still very difficult, the factor of consideration comprises a lot: as observation place, light radiation, incident angle etc.(3) be difficult to storage.As wind energy instantaneity is strong, the time is short, each differs greatly constantly, almost can't store, even storage also will be stored by chemical cell, can aggravate like this pollution of environment.In addition, except calculation server can consume a large amount of energy consumptions, the communication network energy consumption of data center also accounted for very large ratio in data center.Data center network mainly realizes forwarding and the route of information between server by router and switch.In routing procedure, there is mulitpath to select from a station server to an other station server, because the quantity of the router of process and switch is different, the route cost of every paths is difference also.So when solving data center's energy consumption problem, the energy consumption of communication network also must be considered.
As known from the above, data center also will handle the problem of server and communication network high energy consumption well except taking full advantage of green energy resource, does so just and can reduce energy consumption, reduces the output of carbon containing emission, reaches the purpose of sustainable development.At present, in data center, the solution of energy consumption optimization problem mainly contains two kinds: the scheduling of (1) green energy resource and operation distribution method.For the historical data of solar energy by weather forecast and collection, prediction short-term (as 1 hour) solar energy generation; Then, the penalty mechanism function is set, when wherein system utilizes green energy resource, the penalty value is minimum, and job timeout's operation penalty value is the highest; Afterwards, according to the solar energy situation of prediction, under the condition that satisfies constraint, take penalty value minimum as target, the operation that arrives data center is distributed; Repeat above process, complete the distribution of All Jobs, make penalty minimum, namely reach the purpose of maximum using solar energy.(2) heat energy perception and job scheduling method.This method mainly is divided into following steps: at first utilize kalman filter method to estimate the arrival rate of following operation in a short time; Then, according to the operation arrival amount of estimating, macroscopical determining data center is opened the quantity of server; Afterwards, due to the isomerism of server, the heat cycle effect of every station server mechanism is different, the operation of workload maximum is dispensed to the server process of heat cycle effect mechanism the best; Repeat this step, until All Jobs assigns, reach the minimum purpose of system energy consumption.The advantage of above-mentioned first method is to have utilized green energy resource, and shortcoming is that model is too simple, does not consider data center's communication network; The advantage of second method is to reduce the energy consumption of system by rational dispatching method, but does not utilize green energy resource, can't reduce data center's cost and carbon containing emission.
Summary of the invention
The present invention is directed to the deficiency that there is high energy consumption in the available data center, a kind of dispatching method based on the green energy resource perception has been proposed, it obtains next solar energy that constantly can utilize of data center by prediction, under the condition of the energy consumption that takes into full account server and communication network, operation reasonably is dispensed to each server, to reach solar energy utilization ratio as supreme good.
The technical scheme that the present invention solves its technical problem employing is as follows:
A kind of dispatching method based on the green energy resource perception take solar energy as green energy resource, is optimized distribution with the operation that arrives data center, and its step is as follows:
Step 1 according to illumination amplitude and weather conditions, obtains by prediction the solar energy that data center can utilize future in a short time;
Step 2, preliminary treatment is carried out in the operation that arrives data center, and each operation is comprised of several tasks, has the communication constraint relation between each task, according to the communication constraint relation of doing in the industry, classification is carried out in the operation of needs scheduling, will successively be dispensed to every one-level without the task of restriction relation;
Step 3, satisfying under the demand of job time limit, dispatch according to rank for completing pretreated operation set, task in every one-level is dispensed to each server successively, take into full account the energy consumption of server and communication network in assigning process, and to reach the highest solar energy utilization ratio as purpose.
The concrete steps of described step 2 are as follows:
(1) after a group job arrived data center, first buffer memory was to the formation of data center;
(2) initialization i:=0 does not have the task of father's node to preserve as for the i level all;
(3) then i++ processes next stage, and find out all child nodes in the upper level task: if there is no child node, Output rusults directly quits a program;
(4) if any child node, check whether the father node of child node task has been kept in the level that traveled through before, if preserve, child node is saved to when the i level, otherwise does not preserve;
(5) repeating step (2), (3) and (4), until traveled through all tasks, Output rusults quits a program.
The concrete steps of described step 3 are as follows:
(1) initialization from the l level, begins to carry out task and distributes;
(2) suppose that Nl task arranged in the l level, task is dispensed to the M station server successively, and calculate each task after being dispensed to server, the calculating of generation and communication network energy consumption;
(3) calculate simultaneously the solar energy utilization ratio of task, select the highest task of utilance-server-assignment combination;
(4) if the identical distribution of a plurality of utilances combination is arranged simultaneously, select the shortest one group of time of implementation;
(5) repeating step (2), (3) and (4) are until all tasks of l level assign;
(6) complete all tasks of l level and distribute rear l++, jump to step (1);
(7) after having traveled through all ranks, EP (end of program), the highest task allocative decision of output solar energy utilization ratio.
Short-term described in step 1 is one hour.
The present invention is the dispatching method based on the green energy resource perception at a kind of data-oriented center, and what mainly solve is the problem that data center's Green energy takes full advantage of.Beneficial effect of the present invention is:
(1) the present invention is by rational Forecasting Methodology, estimate available solar energy in a short time in future, satisfy under the prerequisite of time limit demand guaranteeing operation, operation is carried out implementing reasonable distribution after preliminary treatment, obtain at last the highest effect of solar energy utilization ratio, reached the purpose that reduces data center's cost and carbon containing emission.
(2) the present invention is to improve solar energy utilization ratio as purpose, and complexity is low, and the speed of service is fast, so the method goes for the data center that scale is large, the task number is many.
(3) the present invention is from overall angle, take into full account the energy consumption of data center server and communication network, more complicated on model, operation is carried out the perception scheduling after dynamically arriving in real time, proper practical application scene, the results show by emulation experiment reliability of the present invention and superiority.
Description of drawings
Fig. 1 is the schematic diagram of data center.
Fig. 2 is method flow diagram of the present invention.
Fig. 3 is the pretreated flow chart of operation of the present invention.
Fig. 4 is the block diagram that task of the present invention is distributed.
Embodiment
The present invention is the dispatching method of the green energy resource perception at a kind of data-oriented center.Main stressing is a kind of dispatching method based on the solar energy perception in the present invention, and this method complexity is low, and the speed of service is fast, goes for large-scale data center.
See also Fig. 2, the dispatching method based on the green energy resource perception of the present invention is optimized distribution take solar energy as green energy resource with the operation that arrives data center, the structure of described data center as shown in Figure 1, the step of this dispatching method is as follows:
Step 1 according to illumination amplitude and weather conditions, obtains by prediction the solar energy that data center can utilize future in a short time.
The present invention passes through the method for prediction, estimates future in a short time, for example can be for the solar energy of data center in one hour; Two factors of prediction solar energy Main Basis: illumination amplitude and weather conditions, can calculate theoretically based on the illumination amplitude energy that next produces constantly, weather conditions have determined the loss of energy, and such as the fine day energy loss is minimum, severe snow sky loss is the highest.
Step 2, preliminary treatment is carried out in the operation that arrives data center, and each operation is comprised of several tasks, has the communication constraint relation between each task, according to the communication constraint relation of doing in the industry, classification is carried out in the operation of needs scheduling, will successively be dispensed to every one-level without the task of restriction relation.
After estimating the situation that following solar energy in a short time, what next step will be done is satisfying under the demand in time limit exactly, processes when job scheduling is abundant to solar energy.In the present invention, the main job model of considering is directed acyclic graph, each operation is comprised of several tasks, there is the communication constraint relation between task, so before schedule job in enormous quantities, must carry out preliminary treatment to operation, carry out classification according to the operation that the restriction relation of doing in the industry will be dispatched, to successively be dispensed to every one-level without the task of restriction relation, then dispatch successively according to rank.
The operation preprocess method of described step 2 sees also Fig. 3, and its concrete steps are as follows:
(1) after a group job arrived data center, first buffer memory was to the formation of data center;
(2) initialization i:=0 does not have the task of father's node to preserve as for the i level all;
(3) then i++ processes next stage, and find out all child nodes in the upper level task: if there is no child node, Output rusults directly quits a program;
(4) if any child node, check whether the father node of child node task has been kept in the level that traveled through before, if preserve, child node is saved to when the i level, otherwise does not preserve;
(5) repeating step (2), (3) and (4), until traveled through all tasks, Output rusults quits a program.
Step 3, satisfying under the demand of job time limit, dispatch according to rank for completing pretreated operation set, task in every one-level is dispensed to each server successively, take into full account the energy consumption of server and communication network in assigning process, and to reach the highest solar energy utilization ratio as purpose.
For completing pretreated operation set, what then will realize is with the task in every one-level, is up to purpose with solar energy utilization ratio, is dispensed to successively each server, and takes into full account the energy consumption cost of communication network in the process of distributing; The hypothesis of this method is time of implementation, power consumption and communication bandwidth and the communication power consumption of known task, according to following step, all tasks is dispensed to server.
The method for allocating tasks of described step 3 sees also Fig. 4, and its concrete steps are as follows:
(1) initialization from the l level, begins to carry out task and distributes;
(2) suppose that Nl task arranged in the l level, task is dispensed to the M station server successively, and calculate each task after being dispensed to server, the calculating of generation and communication network energy consumption;
(3) calculate simultaneously the solar energy utilization ratio of task, select the highest task of utilance-server-assignment combination;
(4) if the identical distribution of a plurality of utilances combination is arranged simultaneously, select the shortest one group of time of implementation;
(5) repeating step (2), (3) and (4) are until all tasks of l level assign;
(6) complete all tasks of l level and distribute rear l++, jump to step (1);
(7) after having traveled through all ranks, EP (end of program), the highest task allocative decision of output solar energy utilization ratio.
Claims (4)
1. the dispatching method based on the green energy resource perception, take solar energy as green energy resource, be optimized distribution with the operation that arrives data center, it is characterized in that, described dispatching method step is as follows:
Step 1 according to illumination amplitude and weather conditions, obtains by prediction the solar energy that data center can utilize future in a short time;
Step 2, preliminary treatment is carried out in the operation that arrives data center, and each operation is comprised of several tasks, has the communication constraint relation between each task, according to the communication constraint relation of doing in the industry, classification is carried out in the operation of needs scheduling, will successively be dispensed to every one-level without the task of restriction relation;
Step 3, satisfying under the demand of job time limit, dispatch according to rank for completing pretreated operation set, task in every one-level is dispensed to each server successively, take into full account the energy consumption of server and communication network in assigning process, and to reach the highest solar energy utilization ratio as purpose.
2. the dispatching method based on the green energy resource perception according to claim 1, is characterized in that, the concrete steps of described step 2 are as follows:
(1) after a group job arrived data center, first buffer memory was to the formation of data center;
(2) initialization i:=0 does not have the task of father's node to preserve as for the i level all;
(3) then i++ processes next stage, and find out all child nodes in the upper level task: if there is no child node, Output rusults directly quits a program;
(4) if any child node, check whether the father node of child node task has been kept in the level that traveled through before, if preserve, child node is saved to when the i level, otherwise does not preserve;
(5) repeating step (2), (3) and (4), until traveled through all tasks, Output rusults quits a program.
3. the dispatching method based on the green energy resource perception according to claim 1, is characterized in that, the concrete steps of described step 3 are as follows:
(1) initialization from the l level, begins to carry out task and distributes;
(2) suppose that Nl task arranged in the l level, task is dispensed to the M station server successively, and calculate each task after being dispensed to server, the calculating of generation and communication network energy consumption;
(3) calculate simultaneously the solar energy utilization ratio of task, select the highest task of utilance-server-assignment combination;
(4) if the identical distribution of a plurality of utilances combination is arranged simultaneously, select the shortest one group of time of implementation;
(5) repeating step (2), (3) and (4) are until all tasks of l level assign;
(6) complete all tasks of l level and distribute rear l++, jump to step (1);
(7) after having traveled through all ranks, EP (end of program), the highest task allocative decision of output solar energy utilization ratio.
4. the dispatching method based on the green energy resource perception according to claim 1, is characterized in that, the short-term described in step 1 is one hour.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310045911.1A CN103107954B (en) | 2013-02-05 | 2013-02-05 | A kind of dispatching method based on green energy resource perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310045911.1A CN103107954B (en) | 2013-02-05 | 2013-02-05 | A kind of dispatching method based on green energy resource perception |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103107954A true CN103107954A (en) | 2013-05-15 |
| CN103107954B CN103107954B (en) | 2015-08-26 |
Family
ID=48315531
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310045911.1A Expired - Fee Related CN103107954B (en) | 2013-02-05 | 2013-02-05 | A kind of dispatching method based on green energy resource perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103107954B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103530192A (en) * | 2013-10-25 | 2014-01-22 | 上海交通大学 | Low-energy-consumption reliability scheduling method based on solar energy sensing |
| CN104536826A (en) * | 2015-01-26 | 2015-04-22 | 中国人民解放军国防科学技术大学 | Wind and light multi-energy data center-oriented green scheduling method for real-time task |
| CN106886274A (en) * | 2017-01-22 | 2017-06-23 | 青海大学 | The management method and device of a kind of consumption of data center |
| CN107341043A (en) * | 2017-06-28 | 2017-11-10 | 东北大学 | A kind of emulation mode for the consumption of data center for assessing regenerative resource hybrid power supply |
| CN112235131A (en) * | 2020-09-25 | 2021-01-15 | 重庆邮电大学 | Data center network service configuration method based on clean energy time window |
| US20220410171A1 (en) * | 2009-07-21 | 2022-12-29 | The Research Foundation For The State University Of New York | Apparatus and method for efficient estimation of the energy dissipation of processor based systems |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271407A (en) * | 2008-05-13 | 2008-09-24 | 武汉理工大学 | Gridding scheduling method based on energy optimization |
| US20100228861A1 (en) * | 2009-03-04 | 2010-09-09 | International Business Machines Corporation | Environmental and computing cost reduction with improved reliability in workload assignment to distributed computing nodes |
| CN102063327A (en) * | 2010-12-15 | 2011-05-18 | 中国科学院深圳先进技术研究院 | Application service scheduling method with power consumption consciousness for data center |
| CN102902878A (en) * | 2012-08-17 | 2013-01-30 | 曙光信息产业(北京)有限公司 | Energy cost perception scheduling method |
-
2013
- 2013-02-05 CN CN201310045911.1A patent/CN103107954B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271407A (en) * | 2008-05-13 | 2008-09-24 | 武汉理工大学 | Gridding scheduling method based on energy optimization |
| US20100228861A1 (en) * | 2009-03-04 | 2010-09-09 | International Business Machines Corporation | Environmental and computing cost reduction with improved reliability in workload assignment to distributed computing nodes |
| CN102063327A (en) * | 2010-12-15 | 2011-05-18 | 中国科学院深圳先进技术研究院 | Application service scheduling method with power consumption consciousness for data center |
| CN102902878A (en) * | 2012-08-17 | 2013-01-30 | 曙光信息产业(北京)有限公司 | Energy cost perception scheduling method |
Non-Patent Citations (1)
| Title |
|---|
| 敬超 等: "面向动态可重构系统的低能耗调度算法", 《微电子学与计算机》, vol. 29, no. 9, 30 September 2012 (2012-09-30), pages 184 - 188 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220410171A1 (en) * | 2009-07-21 | 2022-12-29 | The Research Foundation For The State University Of New York | Apparatus and method for efficient estimation of the energy dissipation of processor based systems |
| CN103530192A (en) * | 2013-10-25 | 2014-01-22 | 上海交通大学 | Low-energy-consumption reliability scheduling method based on solar energy sensing |
| CN104536826A (en) * | 2015-01-26 | 2015-04-22 | 中国人民解放军国防科学技术大学 | Wind and light multi-energy data center-oriented green scheduling method for real-time task |
| CN104536826B (en) * | 2015-01-26 | 2015-10-21 | 中国人民解放军国防科学技术大学 | The green dispatching method of a kind of real-time task towards honourable multiple-energy-source data center |
| CN106886274A (en) * | 2017-01-22 | 2017-06-23 | 青海大学 | The management method and device of a kind of consumption of data center |
| CN106886274B (en) * | 2017-01-22 | 2019-10-29 | 青海大学 | A kind of management method and device of consumption of data center |
| CN107341043A (en) * | 2017-06-28 | 2017-11-10 | 东北大学 | A kind of emulation mode for the consumption of data center for assessing regenerative resource hybrid power supply |
| CN112235131A (en) * | 2020-09-25 | 2021-01-15 | 重庆邮电大学 | Data center network service configuration method based on clean energy time window |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103107954B (en) | 2015-08-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103107954B (en) | A kind of dispatching method based on green energy resource perception | |
| Kumar et al. | A multi-tenant cloud-based DC nano grid for self-sustained smart buildings in smart cities | |
| Kumar et al. | Vehicular delay-tolerant networks for smart grid data management using mobile edge computing | |
| Gelenbe | Energy packet networks: adaptive energy management for the cloud | |
| Gelenbe | Energy packet networks: smart electricity storage to meet surges in demand | |
| Gelenbe | Energy packet networks: ICT based energy allocation and storage | |
| CN106844051A (en) | The loading commissions migration algorithm of optimised power consumption in a kind of edge calculations environment | |
| James et al. | A Low Carbon Kubernetes Scheduler. | |
| Li et al. | Towards dynamic pricing-based collaborative optimizations for green data centers | |
| CN102157057A (en) | Wi-Fi (Wireless Fidelity)-based wireless meter reading device | |
| Neglia et al. | Geographical load balancing across green datacenters: A mean field analysis | |
| Askeland et al. | The role of 4th generation district heating (4GDH) in a highly electrified hydropower dominated energy system: The case of Norway | |
| Stratigakos et al. | A suitable flexibility assessment approach for the pre-screening phase of power system planning applied on the greek power system | |
| KR102513336B1 (en) | A control system for vpp platform and the method thereof | |
| Lin et al. | Adapting datacenter capacity for greener datacenters and grid | |
| Kenzhina et al. | Virtual power plant in industry 4.0: the strategic planning of emerging virtual power plant in Kazakhstan | |
| Huang et al. | Smart energy management system based on reconfigurable AI chip and electrical vehicles | |
| Bogdanov et al. | Energy transition for Japan: Pathways towards a 100% renewable energy system in 2050 | |
| Zamanidou et al. | Day‐ahead scheduling of a hybrid renewable energy system based on generation forecasting using a deep‐learning approach | |
| KR20230125561A (en) | Optimal energy operation method and system based on predictive optimization and load balancing in nanogrid | |
| Kiani et al. | On the fundamental energy trade-offs of geographical load balancing | |
| US20120166005A1 (en) | Shifting of computational load based on power criteria | |
| Gu et al. | Lowering down the cost for green cloud data centers by using ESDs and energy trading | |
| Zhang et al. | Minimizing electricity cost in geographical virtual network embedding | |
| Blanco et al. | Exploiting green energy to reduce the operational costs of multi-center web search engines |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150826 Termination date: 20160205 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |