CN108134842A - System, the method that a kind of cloud host is migrated according to load strategy - Google Patents
System, the method that a kind of cloud host is migrated according to load strategy Download PDFInfo
- Publication number
- CN108134842A CN108134842A CN201810075575.8A CN201810075575A CN108134842A CN 108134842 A CN108134842 A CN 108134842A CN 201810075575 A CN201810075575 A CN 201810075575A CN 108134842 A CN108134842 A CN 108134842A
- Authority
- CN
- China
- Prior art keywords
- host
- module
- load
- cloud
- strategy
- 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
-
- 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
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
Abstract
System, the method that a kind of cloud host is migrated according to load strategy, the system comprises information collection modules and monitoring module, the system carries out data collection and data analysis by described information collection module and the monitoring module, scheduling strategy is made according to the data of analysis, by cloud host from the host that the host of high load moves to low-load；The present invention proposes system, the method that a kind of cloud host is migrated according to load strategy, by cloud host according to load migration, realizes optimization computing resource, improves the effect of computing resource utilization rate.
Description
Technical field
The present invention relates to field of cloud computer technology more particularly to a kind of cloud host to be according to what load strategy was migrated
System, method.
Background technology
With the development of Internet technology, the cloud host of cloud computing is more and more universal, particularly external Amazon, the country
Ali's cloud, Baidu's cloud and Tencent's cloud etc., the server leasing trusteeship business that they make internet traditional in a disguised form makes the transition.
When cloud host generates, major provider can all consider, by currently newly-increased cloud host assignment to which place
It is proper in host, the foundation of distribution be which host load than relatively low, a cloud host assignment will be paid the utmost attention to
To that host；But in host cluster in operation, current loading condition did not represented after this hour, one day etc.
The load of periodic loading condition, i.e. host has periodic variation, it is therefore desirable to increase cyclic loading analysis,
By the way that either automatically or manually go will be on cloud host migration to the low host of load.
Invention content
A kind of system, the method migrated it is an object of the invention to propose cloud host according to load strategy, by cloud master
Machine realizes optimization computing resource, improves the effect of computing resource utilization rate according to load migration.
For this purpose, the present invention uses following technical scheme：
The system that a kind of cloud host is migrated according to load strategy, the system comprises information collection modules and monitoring mould
Block, the system carries out data collection and data analysis by described information collection module and the monitoring module, according to analysis
Data make scheduling strategy, by cloud host from the host that the host of high load moves to low-load；
Described information collection module is used to record the configuration of host cluster, while detects the performance change of host cluster
Situation；Described information collection module is additionally operable to record the quantity of the cloud host in the host cluster on every host and match
It puts；
The data that the monitoring module is used to collect described information collection module are analyzed, i.e., to the performance of host
State is periodically detected, and is then shown the data for analyzing detection.
Preferably, the system also includes scheduling strategy module is defined, the scheduling strategy module that defines is used to formulate cloud
Host migration strategy；
According to the data of monitoring module analysis detection, loading condition of the host within certain section of time cycle is obtained,
It formulates within the time cycle, by cloud host from the strategy on the host that the host of high load moves to low-load.
Preferably, the system also includes module of moving back, it is described move back module for when cloud host in certain section of time cycle
It is interior after high load host is moved on low-load host, after the time cycle is changed, may choose whether to move back
Into former host.
Preferably, the system also includes transferring module, the transferring module is used for cloud host from high load host
It moves on low-load host.
Preferably, the cloud host may be selected certainly during low-load host is moved to from high load host
Dynamic model block or manual mode.
A kind of method that cloud host is migrated according to load strategy is adjusted including information collection module, monitoring module, definition
Spend policy module, transferring module and module of moving back；
The process migrated including cloud host according to load strategy：
Step A：The configuration information of every host in described information collection module record host cluster, while detect place
The performance state of every host in host clusters collects the information of the configuration information of every host and performance state concurrent
It send to the monitoring module；
Step B：Described information collection module records the quantity and configuration information of the cloud host on every host, and sends
To the monitoring module；
Step C：The monitoring module receives the host of described information collection module transmission and the data information of cloud host
Afterwards, the performance state of host is periodically detected, and the data result of detection is periodically shown；
Step D:The host performance state for defining scheduling strategy module and being shown according to the monitoring module, obtains place
The host of high load is in the time cycle of high load in host clusters, formulates within the time cycle, by cloud host from height
The host of load moves to the strategy on the host of low-load；
Step E：The transferring module carries out cloud host according to the migration strategy for defining scheduling strategy module and formulating
Migration.
Preferably, the process of host's machine information is collected including described information collection module：
Step A1：By SNMP standard agreements, the data of the utilization rate of the cpu of host, memory, hard disk and bandwidth are collected
Information；
Step A2：The data information of collection is recorded, then analyzes the cyclically-varying of data.
Description of the drawings
Fig. 1 is general frame figure of the present invention；
The cloud host that Fig. 2 is the present invention migrates flow chart according to load strategy.
Wherein：Information collection module 1, defines scheduling strategy module 3, transferring module 4, module of moving back 5 at monitoring module 2.
Specific embodiment
Technical solution to further illustrate the present invention below with reference to the accompanying drawings and specific embodiments.
The system that a kind of cloud host of the present embodiment is migrated according to load strategy, as shown in Figure 1, the system comprises
Information collection module 1 and monitoring module 2, the system carry out data by described information collection module 1 and the monitoring module 2
Collection and data analysis, scheduling strategy is made according to the data of analysis, and cloud host is moved to low bear from the host of high load
On the host of load；
Described information collection module 1 is used to record the configuration of host cluster, while the performance for detecting host cluster becomes
Change situation；Described information collection module 1 is additionally operable to record the quantity of the cloud host in the host cluster on every host
And configuration；
The data that the monitoring module 2 is used to collect described information collection module 1 are analyzed, i.e., to the property of host
Energy state is periodically detected, and is then shown the data for analyzing detection.
After computing resource pond is formed, cloud host can be distributed in different physical hosts, i.e., on host, with the time and
The passage of business, limited fixation host resource, can carry different cloud host numbers, while according to different clouds respectively
The resource service condition of host, different hosts can be caused, which to put in different times, different load pressures, it is therefore desirable to
The host share loads of high load capacity are replaced on the host of underload by the way that cloud host is moved to from the host of high load capacity
Pressure.
The performance state of record host and cloud host within the different periods is collected by adding in information collection module 1
Variation, analyzed in the different periods, which host is in higher load condition, which host is in underload,
Then data are shown by monitoring module 2, cloud host is finally moved into underload host from high load capacity host
On.
Information collection module 1 can collect as in host cluster by how many hosts, by how many in each host
Cloud host, the numerical value typing of the initialization of host, cpu how many, the equipment such as the memory sum of separate unit host is how many
Information；The cpu utilization rates, memory service condition and network interface bandwidth that host can also be collected according to period unit simultaneously use
Situation.
And monitoring module 2 carries out the performance state of host periodic Data Detection, such as the period is 7 days, host
The load of machine A has 85% in the cpu utilization rates of 20. -23 points, and host B only has in the cpu utilization rates of 20. -23 points
20%, by data being shown after detection, can be apparent from the load condition of every host.
Preferably, the system also includes scheduling strategy module 3 is defined, the scheduling strategy module 3 that defines is used to formulate
Cloud host migration strategy；
According to the data of the monitoring module 2 analysis detection, loading condition of the host within certain section of time cycle is obtained,
It formulates within the time cycle, by cloud host from the strategy on the host that the host of high load moves to low-load.
Defining scheduling strategy module 3, i.e., the data that we are shown from monitoring module 2 see 20. -23 points of next Monday,
The load of the duty ratio host B of host A is higher by very much, we can be formulated the cloud master on host A in the moment point
Machine is moved on host B so that computing resource is optimized, will not waste of resource.
Preferably, the system also includes module 5 of moving back, it is described move back module 5 for when cloud host certain section of week time
In phase after high load host is moved on low-load host, after the time cycle is changed, it may choose whether back
It adjourns in former host.
Such as when the cloud host on host A in next week one 20. -23 points move on host B, if the period
After change, user can independently choose whether to move back in cloud host, i.e., the cloud host on host B is moved back onto host A.If
After the more period, load increases host B, and host A loads are lower, we may be selected by moving back.
Preferably, the system also includes transferring module 4, the transferring module 4 is used for cloud host from high load host
Machine is moved on low-load host.
Preferably, the cloud host may be selected certainly during low-load host is moved to from high load host
Dynamic model block or manual mode.
A kind of method that cloud host is migrated according to load strategy, including information collection module 1, monitoring module 2, definition
Scheduling strategy module 3, transferring module 4 and module 5 of moving back；
As shown in Fig. 2, the process migrated including cloud host according to load strategy：
Step A：Described information collection module 1 records the configuration information of every host in host cluster, detects simultaneously
The performance state of every host in host cluster collects the information of the configuration information of every host and performance state simultaneously
It is sent to the monitoring module 2；
Step B：Described information collection module 1 records the quantity and configuration information of the cloud host on every host, concurrently
It send to the monitoring module 2；
Step C：The monitoring module 2 receives the host of the transmission of described information collection module 1 and the data letter of cloud host
After breath, the performance state of host is periodically detected, and the data result of detection is periodically shown；
Step D:The host performance state for defining scheduling strategy module 3 and being shown according to the monitoring module 2, obtains
The host of high load is in the time cycle of high load in host cluster, formulates within the time cycle, by cloud host from
The host of high load moves to the strategy on the host of low-load；
Step E：The transferring module 4 according to the migration strategy for defining scheduling strategy module 3 and formulating, by cloud host into
Row migration.
Preferably, the process of host's machine information is collected including described information collection module 1：
Step A1：By SNMP standard agreements, the data of the utilization rate of the cpu of host, memory, hard disk and bandwidth are collected
Information；
Step A2：The data information of collection is recorded, then analyzes the cyclically-varying of data.
The technical principle of the present invention is described above in association with specific embodiment.These descriptions are intended merely to explain the present invention's
Principle, and it cannot be construed to limiting the scope of the invention in any way.Based on explanation herein, the technology of this field
Personnel would not require any inventive effort the other specific embodiments that can associate the present invention, these modes are fallen within
Within protection scope of the present invention.
Claims (7)
1. a kind of system that cloud host is migrated according to load strategy, it is characterised in that：The system comprises information to collect mould
Block and monitoring module, the system carry out data collection and data point by described information collection module and the monitoring module
Analysis, makes scheduling strategy, by cloud host from the host that the host of high load moves to low-load according to the data of analysis；
Described information collection module is used to record the configuration of host cluster, while detects the performance change feelings of host cluster
Condition；Described information collection module is additionally operable to record the quantity of the cloud host in the host cluster on every host and match
It puts；
The data that the monitoring module is used to collect described information collection module are analyzed, i.e., to the performance state of host
It is periodically detected, is then shown the data for analyzing detection.
2. a kind of system that cloud host is migrated according to load strategy according to claim 1, it is characterised in that：
The system also includes scheduling strategy module is defined, the scheduling strategy module that defines is used to formulate cloud host migration plan
Slightly；
According to the data of monitoring module analysis detection, loading condition of the host within certain section of time cycle is obtained, is formulated
Within the time cycle, by cloud host from the strategy on the host that the host of high load moves to low-load.
3. a kind of system that cloud host is migrated according to load strategy according to claim 2, it is characterised in that：
The system also includes module of moving back, it is described move back module for when cloud host within certain section of time cycle from high load place
After on host migration to low-load host, after the time cycle is changed, it may choose whether to move back into former host.
4. a kind of system that cloud host is migrated according to load strategy according to claim 1, it is characterised in that：
The system also includes transferring module, the transferring module is used to cloud host moving to low-load from high load host
On host.
5. a kind of system that cloud host is migrated according to load strategy according to claim 1, it is characterised in that：
Automatic module or hand may be selected during low-load host is moved to from high load host in the cloud host
Dynamic model formula.
6. a kind of method that cloud host is migrated according to load strategy, it is characterised in that：Including information collection module, monitoring mould
Block defines scheduling strategy module, transferring module and module of moving back；
The process migrated including cloud host according to load strategy：
Step A：The configuration information of every host in described information collection module record host cluster, while detect host
The information of the configuration information of every host and performance state is collected and is sent to by the performance state of every host in cluster
The monitoring module；
Step B：Described information collection module records the quantity and configuration information of the cloud host on every host, and is sent to institute
State monitoring module；
Step C：It is right after the monitoring module receives the host of described information collection module transmission and the data information of cloud host
The performance state of host is periodically detected, and the data result of detection is periodically shown；
Step D:The host performance state for defining scheduling strategy module and being shown according to the monitoring module, obtains host
The host of high load is in the time cycle of high load in cluster, formulates within the time cycle, by cloud host from high load
Host move to strategy on the host of low-load；
Step E：The transferring module moves cloud host according to the migration strategy for defining scheduling strategy module and formulating
It moves.
7. a kind of method that cloud host is migrated according to load strategy according to claim 6, it is characterised in that：
The process of host's machine information is collected including described information collection module：
Step A1：By SNMP standard agreements, the data letter of the utilization rate of the cpu of host, memory, hard disk and bandwidth is collected
Breath；
Step A2：The data information of collection is recorded, then analyzes the cyclically-varying of data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810075575.8A CN108134842A (en) | 2018-01-26 | 2018-01-26 | System, the method that a kind of cloud host is migrated according to load strategy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810075575.8A CN108134842A (en) | 2018-01-26 | 2018-01-26 | System, the method that a kind of cloud host is migrated according to load strategy |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108134842A true CN108134842A (en) | 2018-06-08 |
Family
ID=62400921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810075575.8A Pending CN108134842A (en) | 2018-01-26 | 2018-01-26 | System, the method that a kind of cloud host is migrated according to load strategy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108134842A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111090470A (en) * | 2019-10-15 | 2020-05-01 | 平安科技（深圳）有限公司 | Secure starting method and device of cloud host, computer equipment and storage medium |
| WO2021179487A1 (en) * | 2020-03-13 | 2021-09-16 | 平安科技（深圳）有限公司 | Method for dynamically allocating cloud host to host machine, electronic apparatus, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101719081A (en) * | 2009-12-01 | 2010-06-02 | 北京大学 | Method for scheduling virtual machines |
| US8327373B2 (en) * | 2010-08-24 | 2012-12-04 | Novell, Inc. | System and method for structuring self-provisioning workloads deployed in virtualized data centers |
| CN102981910A (en) * | 2012-11-02 | 2013-03-20 | 曙光云计算技术有限公司 | Realization method and realization device for virtual machine scheduling |
| CN103023802A (en) * | 2012-12-05 | 2013-04-03 | 暨南大学 | Web-cluster-oriented low energy consumption scheduling system and method |
| CN104010028A (en) * | 2014-05-04 | 2014-08-27 | 华南理工大学 | Dynamic virtual resource management strategy method for performance weighting under cloud platform |
-
2018
- 2018-01-26 CN CN201810075575.8A patent/CN108134842A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101719081A (en) * | 2009-12-01 | 2010-06-02 | 北京大学 | Method for scheduling virtual machines |
| US8327373B2 (en) * | 2010-08-24 | 2012-12-04 | Novell, Inc. | System and method for structuring self-provisioning workloads deployed in virtualized data centers |
| CN102981910A (en) * | 2012-11-02 | 2013-03-20 | 曙光云计算技术有限公司 | Realization method and realization device for virtual machine scheduling |
| CN103023802A (en) * | 2012-12-05 | 2013-04-03 | 暨南大学 | Web-cluster-oriented low energy consumption scheduling system and method |
| CN104010028A (en) * | 2014-05-04 | 2014-08-27 | 华南理工大学 | Dynamic virtual resource management strategy method for performance weighting under cloud platform |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111090470A (en) * | 2019-10-15 | 2020-05-01 | 平安科技（深圳）有限公司 | Secure starting method and device of cloud host, computer equipment and storage medium |
| WO2021179487A1 (en) * | 2020-03-13 | 2021-09-16 | 平安科技（深圳）有限公司 | Method for dynamically allocating cloud host to host machine, electronic apparatus, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9432462B2 (en) | Distributed metering and monitoring system | |
| US20180307734A1 (en) | Correlating performance data and log data using diverse data stores | |
| Logothetis et al. | In-situ {MapReduce} for Log Processing | |
| US20190179815A1 (en) | Obtaining performance data via an application programming interface (api) for correlation with log data | |
| CN101667034B (en) | Scalable monitoring system supporting hybrid clusters | |
| CN103152393B (en) | A kind of charging method of cloud computing and charge system | |
| CN102404126B (en) | Charging method of cloud computing during application process | |
| CN108845878A (en) | The big data processing method and processing device calculated based on serverless backup | |
| CN103955509A (en) | Quick search method for massive electric power metering data | |
| CN111355606B (en) | Web application-oriented container cluster self-adaptive expansion and contraction system and method | |
| US8375228B2 (en) | Multiple-node system power utilization management | |
| CN102662750A (en) | Virtual machine resource optimal control method and control system based on elastic virtual machine pool | |
| CN104133727A (en) | Load distribution method based on real-time resources | |
| CN110888714A (en) | Container scheduling method, device and computer-readable storage medium | |
| CN108920153A (en) | A kind of Docker container dynamic dispatching method based on load estimation | |
| CN102045396A (en) | Load balancing method of server document | |
| WO2021093365A1 (en) | Gpu video memory management control method and related device | |
| CN104462432A (en) | Self-adaptive distributed computing method | |
| CN103841129A (en) | Cloud computing resource information acquisition server, cloud computing resource information acquisition client and information processing method | |
| CN102339233A (en) | Cloud computing centralized management platform | |
| CN112835698A (en) | Heterogeneous cluster-based dynamic load balancing method for request classification processing | |
| da Silva et al. | A science-gateway workload archive to study pilot jobs, user activity, bag of tasks, task sub-steps, and workflow executions | |
| CN107291539A (en) | Cluster program scheduler method based on resource significance level | |
| CN108134842A (en) | System, the method that a kind of cloud host is migrated according to load strategy | |
| US10691653B1 (en) | Intelligent data backfill and migration operations utilizing event processing architecture |
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 |
Application publication date: 20180608 |
|
| RJ01 | Rejection of invention patent application after publication |