CN106961351A - Intelligent elastic telescopic method based on Docker container clusters - Google Patents
Intelligent elastic telescopic method based on Docker container clusters Download PDFInfo
- Publication number
- CN106961351A CN106961351A CN201710123712.6A CN201710123712A CN106961351A CN 106961351 A CN106961351 A CN 106961351A CN 201710123712 A CN201710123712 A CN 201710123712A CN 106961351 A CN106961351 A CN 106961351A
- Authority
- CN
- China
- Prior art keywords
- load
- flexible
- container
- prediction
- elastic telescopic
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention is a kind of intelligent elastic telescopic method based on Docker container clusters, including simultaneously sampled data, data prediction are analyzed, perform flexible algorithm, perform the several steps of telescopic movable for monitoring.The present invention changes for different loads, the flexible error of prediction type stretches to make up adjustment by response type, flexible delayed stretched by prediction type of response type seamlessly transits, this method can preferably tackle periodic load and burst type load, rapid transverse direction Intelligent telescopic Docker containers, and pass through elastic telescopic tolerance and cool time, it is to avoid the generation of jitter conditions.
Description
Technical field
The present invention relates to field of cloud computer technology, more particularly to a kind of intelligent elastic telescopic based on Docker container clusters
Method.
Background technology
The development of internet is maked rapid progress, and is promoting the technologies such as Clustering, Distributed Calculation and cloud computing fast-developing
While, also deeply change daily life.Increasing application and service (for example mouth family website, electric business website,
Social interaction server etc.) all deploy around network.The life of people increasingly be unable to do without internet, so as to cause Internet Users
Flow explosion type expansion in amount surge and network.When cluster faces the excessive situation of workload amount, user terminal it is straight
See experience often to click certain website or ask certain to service and success can not be loaded always, caused result is exactly that request rings
It is elongated between seasonable, it is most likely that to cause the cluster for building service to can't bear the heavy load and paralyse.
Can be cluster and work by adjusting the performance of cluster at this stage for frequent, substantial amounts of load change
The dynamic relationship of load provides a kind of solution of universality, the i.e. telescopic method of cluster.For example strengthen cluster performance, a side
Face can be completed by lifting the performance of every server in cluster, so as to handle bigger workload.But this mode
Still there is limitation, because the performance of single server always has the upper limit, the overall performance of cluster also has the upper limit, works as work
When load continuous constantly becomes big, the performance issue of cluster is still present.On the other hand, can be by increasing an appropriate number of service
Device, the capacity for expanding cluster lift the overall performance of cluster.This mode is easy to operate, flexible, and cost is with respect to former side
Formula is less expensive.Its advantage is, in the case of conditions permit, and cluster scale in theory can be infinitely great.
When in face of dynamic load change, on the one hand, need to reduce the stand-by period of user's request as far as possible, to ensure clothes
Business quality.On the other hand, because present most cloud environments use virtual machine as base unit, virtual machine is given birth to from deployment is started to
Effect needs the time by minute level, in some scenes, when the cluster of dilatation just comes into force or come into force soon, and work is negative
Load has revert to normal level, and the virtual machine of dilatation is just in the state of relative free, thus expanding start-up virtual machine simultaneously
Do not maximize their value, which results in should not expense waste.Therefore, how to reduce the stand-by period of user's request and subtract
The time overhead of few flexible adjustment becomes very crucial.
The content of the invention
The present invention is by the improvement to existing cluster telescopic method, with reference to reason of problems in above-mentioned telescopic process
Analysis, by emerging Docker container techniques, it is proposed that a kind of Intelligent telescopic method based on Docker container clusters, in work
In the case of making load burst type growth, using the flexible dilatation of real-time response formula;In the case where periodic load increases, then adopt
With the flexible dilatation of prediction type.And in the relatively low situation of workload, cluster is shunk.Dependence work may finally be reached
Load situation of change and dynamic flexible adjustment is carried out to cluster, and be finally reached reduction user's request stand-by period, raising system
The target of performance.Carry out flexible adjustment in advance by Forecasting Methodology, and correct flexible according to the loading condition monitored in real time,
Allow stretch and become more intelligent, rationality.
Intelligent elastic telescopic method based on Docker container clusters, it is characterised in that comprise the following steps:
Monitor and sampled data,
Sub- monitoring module in 1-1, every main frame is monitored to the running status and parameter of the main frame, and timing is adopted
The data collected are sent to total monitoring module;
1-2, total monitoring module collect the data, are sent to historic load table, recent load table and the flexible alarm of rule
Device;
Data prediction is analyzed,
2-1. such as is sampled, weighted, being classified at the pretreatment to the load data in the historic load table;
Pretreated historic load is substituted into forecast model by 2-2., and following workload value is predicted;
Flexible algorithm is performed,
Results of the 3-1. based on step 1-2, obtains present load;If present load is more than the default load threshold upper limit
1.1 times or 0.9 times less than load threshold lower limit, then trigger flexible algorithm, perform step 3-2;Otherwise continue to obtain current
Load, and be compared with load threshold;
3-2. calculates present load rate of change according to recent load table, if present load rate of change is more than load and changed
Rate threshold value, explanation is that burst load is flexible using response type, calculates flexible using the real time load in step 1-2 current datas
The quantity of container;If present load rate of change is less than load change rate threshold, illustrates it is not burst load, stretched using prediction type
Contracting, the quantity of Expansion container is calculated using step 2-2 prediction load;
Perform telescopic movable,
4-1, total flexible actuator are sent according to the quantity of the Expansion container calculated in step 3-2 to the flexible actuator of son
Corresponding telescoping request, subsequently into the state of cooling, is denied to the telescopic movable reached, terminates until cool time during this period;
The flexible actuator of 4-2, son is received after telescoping request, is performed the establishment, operation, destruction task of container, is responsible for container
Whole life cycle, and carry out health examination, it is ensured that the Expansion container quantity of operation is consistent with step 3-2 result.
The method being predicted described in step 2-2 to following workload value is Time-Series analysis, machine learning, enhancing
Study, pattern match it is any.
The present invention has the following advantages:
1. reaction is quick
Docker containers are used to carry out intelligent elastic telescopic for base unit, second level start and stop that Docker containers have, money
The features such as source occupancy is few.And by real-time response formula elastic telescopic algorithm, system performance index is monitored in real time, answering
Under the scene loaded to burst type, real-time, fine granularity, the quantity of quick adjustment Docker containers can be reached.
2. it is more intelligent
Combine prediction type flexible flexible with response type, allowing stretch becomes more " intelligence ".It is independent flexible using both
Mode each has deficiency, and comprehensive use can allow their shortcomings complementary, carry out flexible adjustment in advance by Forecasting Methodology,
And correct flexible according to the loading condition monitored in real time, it is to avoid and many remaining sum expenses, while using elastic telescopic tolerance and cooling
Time, it is to avoid the generation of jitter phenomenon, make flexible more " rationality ".
3. simple and flexible
Docker containers are used completely, to solve asking for the flexible adjustment response time length of the existing cluster based on virtual machine
Topic, Docker container second levels, which start, to be stopped, and can be configured flexibly by simple Dockerfile, be created container, resource is accounted for
It is low with rate, make telescopic method more flexible and efficient.
Brief description of the drawings
Fig. 1 is the elastic telescopic control system topological diagram based on Docker container clusters;
Fig. 2 is the intelligent elastic telescopic method flow diagram based on Docker container clusters.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Elastic telescopic control system includes Docker containers cluster and elastic telescopic controller two parts, and wherein Docker holds
Device cluster is made up of multiple main frames, and each main frame includes several containers, a sub- monitor, the flexible actuator of a son;Bullet
Property expansion controller include total monitor, historic load table, one in the recent period load table, the actuator that always stretches, predict
Modeling device and flexible tactful siren, each equipment or component annexation are as shown in Figure 1.
The intelligent elastic telescopic method based on Docker container clusters, included in specific steps as shown in Figure 2:
Monitor sampled data:
Step 1) running status of every main frame is monitored by sub- monitoring module, and the data that timing acquiring is arrived are sent out
Deliver to total monitoring module:Sub- monitoring module is disposed on every main frame, be responsible for sampling, statistics and processing host and container it is real-time
With the CPU usage of history, memory usage, disk I/O, network throughput, request response time, request rate etc.;
Step 2) total monitoring module converges the data that sub- monitoring module is sent, and transmit data to historic load table, it is near
Phase loads table and the flexible alarm of rule;
Data prediction is analyzed:
Step 3) storage historic load, load data is sampled, weighted, is classified etc. pre-processes;
Step 4) by step 3) historic load substitute into forecast model, following workload value is predicted:
Using Forecasting Methodology such as Time-Series analysis, machine learning, enhancing study, pattern match etc., shift to an earlier date before periodic load arrival
Complete prediction task;
Perform flexible algorithm:
Step 5) compare present load and default load threshold, if present load is more than the 1.1 of the load threshold upper limit
Again or less than 0.9 times of load threshold lower limit, the flexible algorithm of triggering, performs step 6);Otherwise continue to obtain present load in real time,
And be compared with load threshold;
Step 6) according to recent load data table, present load rate of change is calculated, is born if present load rate of change is more than
Carry rate of change threshold value, then according to step 2) in present load calculate Expansion container quantity;If present load rate of change is less than
Load change rate threshold, then according to step 4) in prediction load value calculate Expansion container quantity;
Perform telescopic movable:
Step 7) total flexible actuator is based on step 6) flexible quantity, perform corresponding telescopic movable, and enter cooling
State, during this period, is denied to reach telescopic movable, terminates until cool time;
Step 8) the flexible actuator of son manages the whole life cycles such as establishment, operation, the destruction of container, and carry out healthy inspection
Look into, it is ensured that operation number of containers and step 6) result it is consistent.
So far, the intelligent elastic telescopic scheme based on Docker container clusters is realized.
Some involved key operations are defined as follows in above step:
Load estimation model:
Workload changes over time various in practice, and the present embodiment is regarded as a time series.Time series is past
Toward with certain regularity:For example for certain online shopping site, the trading volume in daily evening will be more than the transaction on daytime
Amount, the weekly trading volume at weekend are greater than trading volume in week etc..On long terms, trading volume change has rule skilful, from short
From the point of view of phase, trading volume change is again with randomness.The observation and statistics that the rule of change passes through a period of time are to be found
With it is captured.Its Hurst index is asked to a time series, if result interval is 0.5~1, represents that the sequence has length
Phase Memorability, also referred to as self-similarity.Result of calculation value is closer to 1, and self-similarity is stronger, and predictability is also stronger, can be by length
Phase load is expressed as periodic load.
Autoregression model (AR, Auto-regressive) is a kind of linear prediction algorithm, using target to be predicted before
Historical data value sequence in the same time, certain hiding rule is not found out therefrom, by one dependent variable ordered series of numbers of analysis and separately
Relation between one or more independent variable ordered series of numbers, sets up regression equation and is predicted.Assuming that in an autoregressive process,
Justice is the continuous sample in a random process, then it is linearly dependent on historical sample above, can with following formula come
Represent autoregression model:
yt+a1yt-1+a2yt-2+…+apyt-p=ηt
Wherein ai(i=1 ..., p) represents autoregressive coefficient, ηtWhite noise about during expression, its average represents the sequence for 0
Belong to stationary random process.
The present embodiment is modeled using the autoregression model in Time-Series analysis to history cycle load data, with complete
The prediction of paired future time instance workload.
Loading rate:
Judging the general features of burst type workload is:Within the extremely short time, it is original that workload amount, which increases,
Several times, show as tangent slope of the load curve at this and increase suddenly.It is possible thereby to which the change based on tangent slope is to workload
Whether it is that burst type is judged.When tangent slope change exceedes a certain upper threshold, judge workload situation for burst
Formula workload, then using response type elastic telescopic dilatation;If slope is expanded not less than threshold value using prediction type elastic telescopic
Hold..Load change rate threshold upper limit λuWith lower limit λlSet by keeper, it is assumed that the current request response time is γo, work negative
The upper limit threshold of load is γu/ lower threshold is γl, in the ideal case, load change rate threshold is met:
With
Define the request response time t of on-line monitoringo, the workload γ of t monitoringo(t), present load rate of change
λ, then computing formula be:
Elastic dilatation number of containers formula:
The single container workload γ (t) of t monitoring is defined, the upper limit threshold of workload is γu, target container
Quantity n computing formula is:
The tolerance of resilient expansion and cool time:
Noise may be introduced when starting or stop Docker containers to monitor control index, may be temporarily increased when such as starting negative
Carry, so, after each startup or stopping motion action, it should wait some time, referred to as cool time, such as exist
Dilatation terminates just to allow dilatation again after 3 minutes, just capacity reducing allows capacity reducing again after terminating 5 minutes, in order to shorten
Cool time, any expanding-contracting action must is fulfilled for 10% tolerance, shown in following dilatation formula:
γ (t) > γu*1.1
Wherein γ (t) is the container work load that t is monitored, γuThe upper limit threshold of workload, using tolerance and cold
But the method benefit of time has:
(1) stretched in the way of conservative.When the load increases, increase sharply container quantity to avoid the request of user not
It is very important that can be rejected, and the quantity for reducing container is less worried.
(2) shake is avoided.Prevent from, when loading also unstable, being carried out telescopic movable.
Elastic telescopic algorithm:
When cluster is started working, flexible algorithm also comes into force simultaneously.Timing uses cluster workload, and carries out corresponding
Record, detailed process is as follows:
Step 1) flexible algorithm is performed, often will be to current request response time and work by one section of Fixed Time Interval
Make load to be sampled, while the load factor that monitoring is obtained is separately input in historic load table H and recent load table P, with
Standby rear use.
Step 2) the loading rate λ at current time is calculated according to the data in load table P in the recent period, and utilize historic load
Data in table H, the workload γ to future time instance is calculated by autoregression modelp(t)。
Step 3) observation workload, reflection be current time cluster loading condition.When observation loads γo(t)
Or prediction load γp(t) upper limit threshold γ is exceededuWhen, dilatation algorithm will be triggered.When observation loads γo(t) or prediction load γp
(t) it is less than lower threshold γlWhen, capacity reducing algorithm will be triggered.
Dilatation algorithm:
Whether the current loading rate by calculating, be that burst type judges to workload, negative to work at present
Carry whether type is burst type, when tangent slope change exceedes a certain upper threshold, judge workload situation for burst type
Workload, then using response type elastic telescopic dilatation;If slope is not less than threshold value, using prediction type elastic telescopic dilatation.
Detailed process is as follows:
Step 1) when calling dilatation algorithm, present load rate of change λ is more than workload upper limit threshold γuWhen, container number
Measure n byCalculating is obtained;Present load rate of change λ is less than workload upper limit threshold γuWhen, number of containers n byCalculating is obtained.
Step 2) by step 1) result, in Docker container clusters start n container.
Step 3) after the completion of flexible extension is performed every time, it is required for, by the one-step cooling time, frequently stretching to prevent cluster
Contract " shake " phenomenon caused, in this cooling period, refuses telescopic movable.
The present embodiment is directed to sudden load and periodic load, being capable of intelligent elastic telescopic Docker container clusters.Profit
The features such as with the rapid starting/stopping of Docker containers, resources occupation rate less, realize reduction cluster workload, reduce user's request sound
Between seasonable, and the generation of cluster jitter conditions can be prevented effectively from.
Technological means disclosed in the present invention program is not limited only to the technological means disclosed in above-mentioned embodiment, also includes
Constituted technical scheme is combined by above technical characteristic.
Claims (2)
1. the intelligent elastic telescopic method based on Docker container clusters, it is characterised in that comprise the following steps:
Monitor and sampled data,
Sub- monitoring module in 1-1, every main frame is monitored to the running status and parameter of the main frame, and timing acquiring is arrived
Data send to total monitoring module;
1-2, total monitoring module collect the data, are sent to historic load table, recent load table and the flexible alarm of rule;
Data prediction is analyzed,
2-1. such as is sampled, weighted, being classified at the pretreatment to the load data in the historic load table;
Pretreated historic load is substituted into forecast model by 2-2., and following workload value is predicted;
Flexible algorithm is performed,
Results of the 3-1. based on step 1-2, obtains present load;If present load is more than the default load threshold upper limit
1.1 times or 0.9 times less than load threshold lower limit, then flexible algorithm is triggered, step 3-2 is performed;Otherwise continue to obtain current negative
Carry, and be compared with load threshold;
3-2. calculates present load rate of change according to recent load table, if present load rate of change is more than loading rate threshold
Value, explanation is that burst load is flexible using response type, and Expansion container is calculated using the real time load in step 1-2 current datas
Quantity;If present load rate of change is less than load change rate threshold, illustrate it is not burst load, it is flexible using prediction type,
The quantity of Expansion container is calculated using step 2-2 prediction load;
Perform telescopic movable,
4-1, total flexible actuator send corresponding according to the quantity of the Expansion container calculated in step 3-2 to the flexible actuator of son
Telescoping request, subsequently into the state of cooling, the telescopic movable reached is denied to during this period, is terminated until cool time;
The flexible actuator of 4-2, son is received after telescoping request, is performed the establishment, operation, destruction task of container, is responsible for the whole of container
Individual life cycle, and carry out health examination, it is ensured that Expansion container quantity and the step 3-2 of operation) result it is consistent.
2. the intelligent elastic telescopic method according to claim 1 based on Docker container clusters, it is characterised in that step
To the method that following workload value is predicted it is Time-Series analysis, machine learning, enhancing study, pattern described in 2-2
That matches somebody with somebody is any.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710123712.6A CN106961351A (en) | 2017-03-03 | 2017-03-03 | Intelligent elastic telescopic method based on Docker container clusters |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710123712.6A CN106961351A (en) | 2017-03-03 | 2017-03-03 | Intelligent elastic telescopic method based on Docker container clusters |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106961351A true CN106961351A (en) | 2017-07-18 |
Family
ID=59470812
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710123712.6A Pending CN106961351A (en) | 2017-03-03 | 2017-03-03 | Intelligent elastic telescopic method based on Docker container clusters |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106961351A (en) |
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107231264A (en) * | 2017-07-25 | 2017-10-03 | 北京百度网讯科技有限公司 | For the method and apparatus for the capacity for managing Cloud Server |
CN107370816A (en) * | 2017-07-26 | 2017-11-21 | 郑州云海信息技术有限公司 | A kind of dispositions method and device of Web applications |
CN107404409A (en) * | 2017-09-01 | 2017-11-28 | 广西大学 | Towards the container cloud elastic supply number of containers Forecasting Methodology and system of mutation load |
CN107689891A (en) * | 2017-09-01 | 2018-02-13 | 广西大学 | Container cloud elastic supply time point Forecasting Methodology and system under extensive prominent visit situation |
CN107908461A (en) * | 2017-12-01 | 2018-04-13 | 广东电网有限责任公司佛山供电局 | A kind of resource dynamic retractility device and implementation method based on Docker |
CN108092797A (en) * | 2017-11-21 | 2018-05-29 | 北京奇艺世纪科技有限公司 | A kind of Container Management method and device |
CN108616398A (en) * | 2018-05-08 | 2018-10-02 | 深圳大学 | A kind of container dynamic capacity reduction method based on DNS load-balancing techniques |
CN108762914A (en) * | 2018-04-17 | 2018-11-06 | 广东智媒云图科技股份有限公司 | A kind of Intelligent telescopic method, apparatus, electronic equipment and the storage medium of system architecture |
CN108762928A (en) * | 2018-05-29 | 2018-11-06 | 郑州云海信息技术有限公司 | A kind of processing method and system when elastic telescopic service execution operation |
CN108920153A (en) * | 2018-05-29 | 2018-11-30 | 华南理工大学 | A kind of Docker container dynamic dispatching method based on load estimation |
CN108965485A (en) * | 2018-09-30 | 2018-12-07 | 北京金山云网络技术有限公司 | Management method, device and the cloud platform of container resource |
CN109086120A (en) * | 2018-08-03 | 2018-12-25 | 北京奇虎科技有限公司 | Management method, device and the calculating equipment of resources of virtual machine |
CN109408242A (en) * | 2018-11-13 | 2019-03-01 | 郑州云海信息技术有限公司 | Inserting method and device on a kind of server resource |
CN109446032A (en) * | 2018-12-19 | 2019-03-08 | 福建新大陆软件工程有限公司 | The method and system of the scalable appearance of Kubernetes copy |
CN109710500A (en) * | 2018-11-29 | 2019-05-03 | 西安工程大学 | A kind of combination forecasting method based on automatic telescopic Docker cluster |
CN109800075A (en) * | 2017-11-16 | 2019-05-24 | 航天信息股份有限公司 | Cluster management method and device |
CN109814881A (en) * | 2017-11-21 | 2019-05-28 | 北京京东尚科信息技术有限公司 | Method and apparatus for disposing data-base cluster |
CN109933180A (en) * | 2019-03-22 | 2019-06-25 | 联想(北京)有限公司 | A kind of control method, device and electronic equipment |
CN109961151A (en) * | 2017-12-21 | 2019-07-02 | 同方威视科技江苏有限公司 | For the system for calculating service of machine learning and for the method for machine learning |
CN110175068A (en) * | 2019-04-16 | 2019-08-27 | 平安科技(深圳)有限公司 | Host number elastic telescopic method, apparatus and computer equipment in distributed system |
CN110198244A (en) * | 2019-06-19 | 2019-09-03 | 北京百度网讯科技有限公司 | Resource allocation method and device towards isomery cloud service |
CN110209467A (en) * | 2019-05-23 | 2019-09-06 | 华中科技大学 | A kind of flexible resource extended method and system based on machine learning |
CN110554904A (en) * | 2019-07-30 | 2019-12-10 | 无锡华云数据技术服务有限公司 | scaling strategy processing method, device, equipment and storage medium |
CN110647392A (en) * | 2018-06-26 | 2020-01-03 | 山东北邮信息技术产业研究院有限公司 | Intelligent elastic expansion method based on container cluster |
CN110990160A (en) * | 2019-12-27 | 2020-04-10 | 广西电网有限责任公司 | Static security analysis container cloud elastic expansion method based on load prediction |
WO2020135510A1 (en) * | 2018-12-29 | 2020-07-02 | 中兴通讯股份有限公司 | Burst load prediction method and device, storage medium and electronic device |
CN112532634A (en) * | 2020-11-30 | 2021-03-19 | 厦门亿联网络技术股份有限公司 | Capacity expansion and capacity reduction method and device for server in audio and video conference and storage medium |
CN113032157A (en) * | 2021-05-31 | 2021-06-25 | 睿至科技集团有限公司 | Automatic intelligent server capacity expansion and reduction method and system |
CN113553180A (en) * | 2021-07-20 | 2021-10-26 | 唯品会(广州)软件有限公司 | Container scheduling method and device and electronic equipment |
CN115202889A (en) * | 2022-09-13 | 2022-10-18 | 阿里巴巴(中国)有限公司 | Computing resource adjusting method and computing system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102176203A (en) * | 2011-02-25 | 2011-09-07 | 曙光信息产业(北京)有限公司 | Prediction method of load parameter |
CN105554102A (en) * | 2015-12-14 | 2016-05-04 | 中电科华云信息技术有限公司 | Elastic expansion method based on container cluster and application system thereof |
-
2017
- 2017-03-03 CN CN201710123712.6A patent/CN106961351A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102176203A (en) * | 2011-02-25 | 2011-09-07 | 曙光信息产业(北京)有限公司 | Prediction method of load parameter |
CN105554102A (en) * | 2015-12-14 | 2016-05-04 | 中电科华云信息技术有限公司 | Elastic expansion method based on container cluster and application system thereof |
Non-Patent Citations (1)
Title |
---|
苗立尧: ""一种基于Docker容器的集群分段伸缩方法"", 《计算机应用与软件》 * |
Cited By (43)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107231264A (en) * | 2017-07-25 | 2017-10-03 | 北京百度网讯科技有限公司 | For the method and apparatus for the capacity for managing Cloud Server |
CN107370816A (en) * | 2017-07-26 | 2017-11-21 | 郑州云海信息技术有限公司 | A kind of dispositions method and device of Web applications |
CN107404409A (en) * | 2017-09-01 | 2017-11-28 | 广西大学 | Towards the container cloud elastic supply number of containers Forecasting Methodology and system of mutation load |
CN107689891A (en) * | 2017-09-01 | 2018-02-13 | 广西大学 | Container cloud elastic supply time point Forecasting Methodology and system under extensive prominent visit situation |
CN107404409B (en) * | 2017-09-01 | 2020-01-31 | 广西大学 | Method and system for predicting quantity of container cloud flexible supply containers for sudden load |
CN109800075A (en) * | 2017-11-16 | 2019-05-24 | 航天信息股份有限公司 | Cluster management method and device |
CN108092797A (en) * | 2017-11-21 | 2018-05-29 | 北京奇艺世纪科技有限公司 | A kind of Container Management method and device |
CN109814881A (en) * | 2017-11-21 | 2019-05-28 | 北京京东尚科信息技术有限公司 | Method and apparatus for disposing data-base cluster |
CN107908461A (en) * | 2017-12-01 | 2018-04-13 | 广东电网有限责任公司佛山供电局 | A kind of resource dynamic retractility device and implementation method based on Docker |
CN109961151B (en) * | 2017-12-21 | 2021-05-14 | 同方威视科技江苏有限公司 | System of computing services for machine learning and method for machine learning |
CN109961151A (en) * | 2017-12-21 | 2019-07-02 | 同方威视科技江苏有限公司 | For the system for calculating service of machine learning and for the method for machine learning |
CN108762914A (en) * | 2018-04-17 | 2018-11-06 | 广东智媒云图科技股份有限公司 | A kind of Intelligent telescopic method, apparatus, electronic equipment and the storage medium of system architecture |
CN108616398B (en) * | 2018-05-08 | 2021-04-30 | 深圳大学 | DNS load balancing technology-based container dynamic capacity reduction method |
CN108616398A (en) * | 2018-05-08 | 2018-10-02 | 深圳大学 | A kind of container dynamic capacity reduction method based on DNS load-balancing techniques |
CN108920153B (en) * | 2018-05-29 | 2022-04-22 | 华南理工大学 | Docker container dynamic scheduling method based on load prediction |
CN108762928B (en) * | 2018-05-29 | 2022-03-22 | 郑州云海信息技术有限公司 | Processing method and system for elastic scaling service during execution operation |
CN108920153A (en) * | 2018-05-29 | 2018-11-30 | 华南理工大学 | A kind of Docker container dynamic dispatching method based on load estimation |
CN108762928A (en) * | 2018-05-29 | 2018-11-06 | 郑州云海信息技术有限公司 | A kind of processing method and system when elastic telescopic service execution operation |
CN110647392A (en) * | 2018-06-26 | 2020-01-03 | 山东北邮信息技术产业研究院有限公司 | Intelligent elastic expansion method based on container cluster |
CN109086120A (en) * | 2018-08-03 | 2018-12-25 | 北京奇虎科技有限公司 | Management method, device and the calculating equipment of resources of virtual machine |
CN108965485A (en) * | 2018-09-30 | 2018-12-07 | 北京金山云网络技术有限公司 | Management method, device and the cloud platform of container resource |
CN108965485B (en) * | 2018-09-30 | 2021-10-15 | 北京金山云网络技术有限公司 | Container resource management method and device and cloud platform |
CN109408242B (en) * | 2018-11-13 | 2020-08-04 | 郑州云海信息技术有限公司 | Server resource online and offline method and device |
CN109408242A (en) * | 2018-11-13 | 2019-03-01 | 郑州云海信息技术有限公司 | Inserting method and device on a kind of server resource |
CN109710500A (en) * | 2018-11-29 | 2019-05-03 | 西安工程大学 | A kind of combination forecasting method based on automatic telescopic Docker cluster |
CN109446032A (en) * | 2018-12-19 | 2019-03-08 | 福建新大陆软件工程有限公司 | The method and system of the scalable appearance of Kubernetes copy |
WO2020135510A1 (en) * | 2018-12-29 | 2020-07-02 | 中兴通讯股份有限公司 | Burst load prediction method and device, storage medium and electronic device |
CN109933180B (en) * | 2019-03-22 | 2021-06-15 | 联想(北京)有限公司 | Control method and device and electronic equipment |
CN109933180A (en) * | 2019-03-22 | 2019-06-25 | 联想(北京)有限公司 | A kind of control method, device and electronic equipment |
CN110175068A (en) * | 2019-04-16 | 2019-08-27 | 平安科技(深圳)有限公司 | Host number elastic telescopic method, apparatus and computer equipment in distributed system |
CN110209467A (en) * | 2019-05-23 | 2019-09-06 | 华中科技大学 | A kind of flexible resource extended method and system based on machine learning |
CN110198244B (en) * | 2019-06-19 | 2022-02-18 | 北京百度网讯科技有限公司 | Heterogeneous cloud service-oriented resource configuration method and device |
CN110198244A (en) * | 2019-06-19 | 2019-09-03 | 北京百度网讯科技有限公司 | Resource allocation method and device towards isomery cloud service |
US11449774B2 (en) | 2019-06-19 | 2022-09-20 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Resource configuration method and apparatus for heterogeneous cloud services |
CN110554904A (en) * | 2019-07-30 | 2019-12-10 | 无锡华云数据技术服务有限公司 | scaling strategy processing method, device, equipment and storage medium |
CN110990160A (en) * | 2019-12-27 | 2020-04-10 | 广西电网有限责任公司 | Static security analysis container cloud elastic expansion method based on load prediction |
CN112532634A (en) * | 2020-11-30 | 2021-03-19 | 厦门亿联网络技术股份有限公司 | Capacity expansion and capacity reduction method and device for server in audio and video conference and storage medium |
CN113032157A (en) * | 2021-05-31 | 2021-06-25 | 睿至科技集团有限公司 | Automatic intelligent server capacity expansion and reduction method and system |
CN113032157B (en) * | 2021-05-31 | 2021-08-24 | 睿至科技集团有限公司 | Automatic intelligent server capacity expansion and reduction method and system |
CN113553180A (en) * | 2021-07-20 | 2021-10-26 | 唯品会(广州)软件有限公司 | Container scheduling method and device and electronic equipment |
CN113553180B (en) * | 2021-07-20 | 2023-10-13 | 唯品会(广州)软件有限公司 | Container scheduling method and device and electronic equipment |
CN115202889A (en) * | 2022-09-13 | 2022-10-18 | 阿里巴巴(中国)有限公司 | Computing resource adjusting method and computing system |
CN115202889B (en) * | 2022-09-13 | 2022-12-09 | 阿里巴巴(中国)有限公司 | Computing resource adjusting method and computing system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106961351A (en) | Intelligent elastic telescopic method based on Docker container clusters | |
CN112000459B (en) | Method for expanding and shrinking capacity of service and related equipment | |
Jiang et al. | Cloud analytics for capacity planning and instant VM provisioning | |
CN107247651B (en) | Cloud computing platform monitoring and early warning method and system | |
CN104317658B (en) | A kind of loaded self-adaptive method for scheduling task based on MapReduce | |
WO2021159638A1 (en) | Method, apparatus and device for scheduling cluster queue resources, and storage medium | |
Tang et al. | Nanily: A qos-aware scheduling for dnn inference workload in clouds | |
CN110990138A (en) | Resource scheduling method, device, server and storage medium | |
CN107911399A (en) | A kind of elastic telescopic method and system based on load estimation | |
CN110796591B (en) | GPU card using method and related equipment | |
CN105022823B (en) | A kind of cloud service performance early warning event generation method based on data mining | |
CN116185584A (en) | Multi-tenant database resource planning and scheduling method based on deep reinforcement learning | |
CN105491079B (en) | The method and device of the required resource of adjustment application in cloud computing environment | |
CN109800075A (en) | Cluster management method and device | |
CN109992392A (en) | A kind of calculation resource disposition method, device and Resource Server | |
CN116643844A (en) | Intelligent management system and method for automatic expansion of power super-computing cloud resources | |
CN115774602A (en) | Container resource allocation method, device, equipment and storage medium | |
Ren et al. | Dynamic scheduling for energy minimization in delay-sensitive stream mining | |
CN114300082B (en) | Information processing method and device and computer readable storage medium | |
CN116149848A (en) | Load prediction method and device, electronic equipment and storage medium | |
CN113904940A (en) | Resource adjusting method and device, electronic equipment and computer readable storage medium | |
Qiu et al. | FLASH: Fast model adaptation in ML-centric cloud platforms | |
CN114988241A (en) | Elevator energy consumption on-line monitoring analysis system | |
KR20230089509A (en) | Bidirectional Long Short-Term Memory based web application workload prediction method and apparatus | |
Ullah et al. | Towards workload-aware cloud resource provisioning using a multi-controller fuzzy switching approach |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170718 |