CN103916438B - Cloud testing environment scheduling method and system based on load forecast - Google Patents
Cloud testing environment scheduling method and system based on load forecast Download PDFInfo
- Publication number
- CN103916438B CN103916438B CN201310003538.3A CN201310003538A CN103916438B CN 103916438 B CN103916438 B CN 103916438B CN 201310003538 A CN201310003538 A CN 201310003538A CN 103916438 B CN103916438 B CN 103916438B
- Authority
- CN
- China
- Prior art keywords
- main frame
- load
- test main
- test
- module
- 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.)
- Active
Links
Landscapes
- Debugging And Monitoring (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention discloses a cloud testing environment scheduling method based on load forecast. The cloud testing environment scheduling method based on the load forecast comprises the steps that a load changing database of testing host machines is established, and selection strategies of physical nodes are configured; the average load of all the physical nodes is obtained; according to the average load and a low threshold value, the testing host machines needing to be migrated on the physical nodes are judged; according to the load changing database, loads of the physical nodes are forecasted, and forecasting results are fed back; according to the forecasting results and the node selection strategies, the testing host machines needing to be scheduled are migrated to the selected physical nodes. According to the cloud testing environment scheduling method based on the load forecast, through the forecasting of the loads of all the physical nodes in the cloud testing environment, the most suitable target node of the cloud testing host machines needing to be scheduled is found, the dual targets of load balance and energy conservation of the cloud testing environment are achieved, the performance bottleneck of the cloud testing environment is eliminated, in this way, the stability of the whole cloud testing environment is improved, and the energy consumption of the system is lowered. The invention further discloses a cloud testing environment scheduling system based on the load forecast.
Description
Technical field
The invention belongs to test environment dispatching method and system, and in particular to a kind of cloud test environment based on load estimation
Dispatching method and its system.
Background technology
Recent years, cloud computing have become the hot technology of current IT circles, and by cloud computing, user can be elastic on demand
Using the resource of magnanimity, so as to optimized integration facility is serviced.Virtualization is the core technology of cloud computing, and it is by bottom physics
Equipment is included unified resource pool and is managed, and the resource that bottom is provided is provided using virtual machine upwards as least unit, so as to
Improve the service efficiency and the efficiency of management of computing resource.Under the ordering about of demand, the virtual machine quantity of data center is increasingly
It is many, new challenge is proposed to the scheduling of resource of cluster virtual machine.In large-scale cluster virtual machine, the load meeting of virtual machine
With user application and demand and Jing often changes, when load dynamic change cause the resource of physical node cannot meet it is all should
During seeervice level target, one or several virtual machine (vm) migration on the node can be saved to other by dynamic migration technology
Put to solve.In the process, the loading condition for how accurately rapidly obtaining each physical node again is a key issue.
If the load variation tendency of each physical host can be understood before scheduling in advance, it is to need the virtual machine of migration to find to be best suitable for
Destination node, it would be beneficial in the load balancing of whole cloud platform.
Due to the multiformity applied in cloud platform, the resources of virtual machine scheduling strategy for finding a suitable all situations is ratio
More difficult, therefore the present invention focuses on the test main frame scheduling of resource in cloud test environment.Cloud test is based on cloud computing
A kind of novel test scheme, has the advantages that low cost, computing capability be strong, test process is simple.Survey in cloud test environment
The load of examination main frame often presents certain regularity, and this provides possibility for the prediction of load.At present, in multinode resource
Scheduling aspect, existing method often only consider impacts of the CPU to loading, and do not consider the other factors such as internal memory, disk, and
The load change of the physical node in transition process is not accounted for, it is impossible to the load shape of physical node after a priori precognition is migrated
Condition.
Applicant has applied for entitled " the STE dynamic generation based on cloud computing on May 23rd, 2012
The patent of invention of system and its implementation ", Application No. 201210162008.9." the test wrapper proposed in the patent of invention
The algorithm adopted by border dynamic dispatching module " is simple, and energy consumption is big, and load balancing effect is not good, therefore the present invention is to aforementioned invention
" the test environment dynamic dispatching module " proposed in patent is improved and has been supplemented.
Load balance and energy-conservation double goal of the present invention according to cloud test platform, while considering to survey under cloud test environment
The load rule of examination main frame, proposes a kind of cloud test environment dispatching method and its system based on load estimation, and it can be surveyed for cloud
Test ring border provides transparent load balance, does not carry out any change to the application in test main frame.
The content of the invention
The present invention proposes a kind of cloud test environment dispatching method based on load estimation and its system, by predicting that cloud is surveyed
The load of each physical node in test ring border, is that the cloud test main frame for needing scheduling finds most suitable destination node, makes whole
Cloud test environment reaches the double goal of load balancing and energy-conservation, eliminates the performance bottleneck of cloud test environment, whole so as to improve
The stability of cloud test environment, reduces the energy consumption of system.
The present invention proposes a kind of cloud test environment dispatching method based on load estimation, comprises the following steps:Step one:
Set up test main frame load delta data storehouse, and configure the selection strategy of the physical node in the cloud test environment module, choosing
Select strategy specifically to preserve in the form of configuration file;Step 2:It is each in test main frame scheduler module acquisition cloud test environment module
Average load in the current slot of individual physical node, the length of the time period is setting in advance;Step 3:The test
Host schedules module judges physical node in the cloud test environment module according to the average load and the threshold ones of setting
The upper test main frame for needing to be migrated;Step 4:According to the load delta data storehouse, test main frame load prediction module pair
The load of the physical node in the cloud test environment module is predicted, and will predict the outcome and feed back to the test main frame and adjust
Degree module;Step 5:According to predicting the outcome, the test main frame scheduler module is according to node selection strategy, it would be desirable to scheduling
Test main frame moves to the physical node of selection.
In a kind of cloud test environment dispatching method based on load estimation of the present invention, step 6 is further included:If step
The average load of physical node in rapid two then disables the node less than the threshold ones of default.
It is in a kind of cloud test environment dispatching method based on load estimation of the present invention, in the step 2, described average
Load includes that CPU usage, memory usage, disk utilization rate and the network bandwidth take.
In a kind of cloud test environment dispatching method based on load estimation of the present invention, in the step 3, if the thing
The average load of reason node is less than the threshold ones of default, then all test main frames on the physical node need migration,
The physical node is disabled after migration;If the average load of the physical node is constantly selected more than the high threshold of default
Select the maximum test main frame of occupancy resource to be migrated, till the load of the physical node is less than high threshold.
In a kind of cloud test environment dispatching method based on load estimation of the present invention, in the step 4, the load
Predicted operation is comprised the following steps:Step A1:The test main frame load prediction module obtains each survey on the physical node
The historic load of examination main frame;Step A2:The test main frame load prediction module is by the historic load of each test main frame and institute
The load stated in test main frame load delta data storehouse is compared, and is the load change song that each test main frame finds matching
Line;Step A3:The test main frame load prediction module obtains every according to the load change curves of the matching of each test main frame
The prediction load of individual test main frame, the prediction load of each test main frame is added the prediction load for obtaining physical node.
In a kind of cloud test environment dispatching method based on load estimation of the present invention, in the step 5, the test
Host migration operation includes:Step B1:For the test main frame of each needs migration, the test main frame scheduler module is described
Physical node is selected in cloud test environment module so that the load estimation value of the physical node adds test main frame to be migrated
Load estimation value scope between the threshold ones and high threshold of default, if without suitable physical node, no
Generation is migrated, if there is multiple physical nodes to meet, selects physical node according to the selection strategy of physical node;Step B2:Institute
State test main frame scheduler module test main frame to be migrated to be migrated to selected physical node.
The invention allows for a kind of cloud test environment scheduling system based on load estimation, including:Node selection strategy
Module, for configuring the selection strategy of physical node;Test main frame scheduler module, for dispatching physical node in cloud test environment
On test main frame;Test main frame load prediction module, for predicting the load of the test main frame in cloud test environment;Survey
Examination load on host computers delta data storehouse, is connected with the test main frame load prediction module, for storing the negative of the test main frame
Carry delta data;Cloud test environment module, for running the test main frame;Wherein, the node selection strategy module and institute
State the connection of test main frame scheduler module;The test main frame scheduler module respectively with the test main frame load prediction module and institute
State the connection of cloud test environment module;The test main frame load prediction module loads delta data storehouse with the test main frame respectively
Connect with the cloud test environment module.
The technical characteristic and its advantage that the present invention is adopted is following aspect:
The present invention realizes a kind of cloud test environment dispatching method and its system by using load estimation, design, and this is
The advantage of system is:By the load for predicting each physical node in cloud test environment, it is that the cloud test main frame for needing scheduling finds
Most suitable destination node, makes whole cloud test environment reach the double goal of load balancing and energy-conservation, eliminates cloud test environment
Performance bottleneck, so as to improve the stability of whole cloud test environment, reduce the energy consumption of system.
Description of the drawings
Fig. 1 is flow chart of the present invention based on the cloud test environment dispatching method of load estimation;
Workflow diagrams of the Fig. 2 for load estimation;
Fig. 3 is the workflow diagram of test main frame migration;
Fig. 4 is the structural representation that cloud test environment of the present invention based on load estimation dispatches system.
Specific embodiment
Specific embodiments of the present invention are further described in detail with reference to the accompanying drawings and examples, but should not be with
This limits the scope of the invention.
In as shown in Figure 1 to Figure 4,1- test main frame scheduler modules, 2- node selection strategy modules, 3- test main frame load estimations
Module, 4- test main frames load delta data storehouse, 5- cloud test environment modules.
As shown in figure 1, the cloud test environment dispatching method based on load estimation of the present invention, including:
Step one:Test main frame load delta data storehouse 4 is set up, and configures the physical node in cloud test environment module 5
Selection strategy, selection strategy specifically in the form of configuration file preserve;
Step 2:Test main frame scheduler module 1 obtains the current slot of each physical node in cloud test environment module 5
Interior average load, the length of time period is setting in advance;
Step 3:Test main frame scheduler module 1 judges cloud test environment mould according to average load and the threshold ones of setting
The test main frame for being migrated is needed in block 5 on physical node;
Step 4:According to load delta data storehouse 4, test main frame load prediction module 3 is in cloud test environment module 5
The load of physical node is predicted, and will predict the outcome and feed back to test main frame scheduler module 1;
Step 5:According to predicting the outcome, test main frame scheduler module 1 is according to node selection strategy, it would be desirable to the survey of scheduling
Physical node of the examination host migration to selection.
Further include step 6:If low door of the average load of certain physical node less than default in previous steps two
Limit value, then disable the node with energy-conservation.
In step 2, the average load of physical node includes but is not limited to CPU usage, memory usage, disk and uses
Rate and network bandwidth occupancy etc., can increase the metric of average load according to practical situation.
In step 3, judge that the criterion migrated by the needs of the test main frame on which physical node is:If something
The average load of reason node is less than the threshold ones of default, then all test main frames on the physical node need migration,
The physical node is disabled after migration with energy-conservation;If the average load of certain physical node is more than the high threshold of default, should
Partial test main frame on physical node needs migration, selects the test main frame for taking resource maximum to be migrated, until the thing
The load of reason node is less than till high threshold.
In step 4, load estimation is comprised the following steps:
Step A1:Test main frame load prediction module 3 obtains the historic load of each test main frame on the physical node;
Step A2:The historic load of each test main frame is become by test main frame load prediction module 3 with test main frame load
The load change curves changed in data base 4 are compared, and are the load change most matched during each test main frame finds data base
Curve, the standard for most matching for the load change curves in historic load curve and the data base of test main frame Euclidean distance most
It is little;
Step A3:Test main frame load prediction module 3 is pre- according to the load change curves for most matching of each test main frame
The load of each test main frame one time point of future is surveyed, the prediction load of each test main frame is added and is obtained physical node
Prediction load.
In step 5, test main frame migration operation includes:
Step R1:For the test main frame of each needs migration, test main frame scheduler module 1 is in cloud test environment module 5
It is middle to select suitable physical node so that load estimation of the load estimation value of the physical node plus test main frame to be migrated
After value, its scope is not moved between the threshold ones and high threshold of default if it can not find suitable physical node
Move, if there are multiple physical nodes to meet the requirements, a physical node is selected according to the selection strategy of physical node;
Step B2:Test main frame scheduler module 1 migrates test main frame to be migrated to selected physical node.
As shown in figure 4, the cloud test environment scheduling system based on load estimation of the present invention, including:Test main frame is dispatched
Module 1, storehouse 4, the cloud test of node selection strategy module 2, test main frame load prediction module 3, test main frame load delta data
Environment module 5.Node selection strategy module 2 is connected with test main frame scheduler module 1;Test main frame scheduler module 1 respectively with survey
Examination host load prediction module 3 and cloud test environment module 5 connect;Test main frame load prediction module 3 respectively with test main frame
Load delta data storehouse 4 and cloud test environment module 5 connect.
Test main frame scheduler module 1 is needed for judging the test main frame in cloud test environment module 5 in which physical node
Migrate, and according to load estimation situation and node selection strategy, the test main frame of each needs migration is moved to most suitable
Physical node on.One threshold ones of default and a high threshold, when the load of certain physical node is less than low door
During limit value, all test main frames on the physical node are all migrated out, then shut down the physical node.When certain physics
When the load of node is more than high threshold, the occupancy resource on the physical node maximum test main frame is migrated out, if
The load of the physical node after migration then continues the test main frame that migration resource occupation takes second place, until this still greater than high threshold
The load of physical node is less than till high threshold.According to load estimation situation, test main frame scheduler module 1 is in cloud test environment
Suitable physical node is selected in module 5 so that the load estimation value of the physical node is plus the negative of test main frame to be migrated
Carry predictive value after its scope between the threshold ones and high threshold of default, if it can not find suitable physical node not
Generation is migrated, if there is multiple physical nodes to meet the requirements, selects a physical node according to the selection strategy of physical node.
The effect of node selection strategy module 2 is configuration node selection strategy, when test main frame to be migrated has multiple mesh
Physical node a physical node is selected according to selection strategy when meeting the requirements.Node selection strategy specifically can be configured to selection
The most abundant node of cpu resource, the node for selecting memory source most abundant, the node for selecting disk resource most abundant, selection net
The most abundant node of network bandwidth resources, the node for selecting prediction load minimum etc..Node selection strategy can be according to the concrete of user
Depending on requirement.
The effect of test main frame load prediction module 3 is the load for predicting each physical node in cloud test environment module 5,
Load includes the CPU usage of the physical node, memory usage, disk utilization rate, network bandwidth occupancy.Test main frame is loaded
Prediction module 3 is by the historic load of certain test main frame and test main frame to be loaded the load change curves in delta data storehouse 4
Compare, find out the load change curves for most matching the test main frame, so as to predict the load of the test main frame, and then to something
On reason node, the prediction load summation of all of test main frame, obtains the load estimation value of the physical node.
Test main frame load delta data storehouse 4 is used to store the typical load change curves of different type test main frame.
Load change curves are stored in the way of two-dimentional point set, i.e., one curve is made up of several points, and each point has a two dimension to sit
Mark, wherein abscissa represents the moment, and vertical coordinate represents load value.The data base only need to set up one before scheduling System Operation
It is secondary, need not resettle when repeatedly dispatching later.
Cloud test environment module 5 is to build the virtual machine set on cloud computing environment, and each test main frame is one
Virtual machine.The LAN that cloud test environment module 5 is made up of multiple stage computers, and every computer is respectively mounted Linux behaviour
Make system.Whole cloud test environment module 5 uses CloudStack as cloud platform software.Multiple test main frames constitute a test
Domain, for performing a specific test event.The visual specific test event of test main frame number included in test domain
Depending on requirement.Test main frame in each test domain is divided into main test main frame and auxiliary test main frame, and main test main frame is that test is soft
The main test main frame of most of indexs such as part function, security reliability, performance, ease for use;Auxiliary test main frame is usually used to full
The different testing requirement of foot or special test event, such as compatibility test, simulation real world testing, across comparison test etc..
Embodiment 1:Load estimation is operated
As shown in Fig. 2 test main frame load prediction module 3 is by the historic load of certain test main frame and test main frame are born
The load change curves carried in delta data storehouse 4 compare, and find out the load change curves for most matching the test main frame, so as to pre-
The load of the test main frame, and then the prediction load summation to all of test main frame on certain physical node are surveyed, the physics is obtained
The load estimation value of node.Test main frame load prediction module 3 obtains going through for each test main frame on the physical node first
History is loaded, then the load change curves in the historic load of each test main frame and test main frame load delta data storehouse 4 are entered
Row is compared, and is the load change curves most matched during each test main frame finds data base, and the standard for most matching is test main frame
Historic load curve and data base in load change curves Euclidean distance it is minimum, finally according to each test main frame most
The load change curves of matching predict the load of each test main frame one time point of future, and the prediction of each test main frame is born
Carry and be added the prediction load for obtaining physical node.
Embodiment 2:Test main frame migration operation
When in cloud test environment module 5, certain physics node load is less than the threshold ones of default or higher than default
High threshold when, it will occur test main frame migration.Migration operation includes:Test main frame scheduler module 1 is in cloud test wrapper
Suitable physical node is selected in border module 5 so that the load estimation value of the physical node is plus test main frame to be migrated
After load estimation value, its scope is between the threshold ones and high threshold of default, if it can not find suitable physical node
There is no migration, if there are multiple physical nodes to meet the requirements, a physical node is selected according to the selection strategy of physical node;
Test main frame scheduler module 1 migrates test main frame to be migrated to selected physical node.In the present embodiment, one is migrated
The concrete steps of test main frame are as shown in Figure 3:1) original test main frame is preserved and is worked.2) the original survey of loop iteration ground transmission
The page of examination main frame is to target physical node, until iterationses reach certain numerical value.3) original test main frame is hung up, transmission
Last ripple page and CPU state.4) original test main frame is deleted, starts new test main frame on target physical node.Jing
Test, migrates the test main frame that a virtual memory is respectively 128MB, 256MB, 512MB, 1024MB, and required time is respectively about
For 3 minutes, 4 minutes, 13 minutes, 16 minutes.
The foregoing is only presently preferred embodiments of the present invention, not for limit the present invention practical range.Belonging to any
Has usually intellectual in technical field, without departing from the spirit and scope of the present invention, when various variations and retouching can be made, originally
The protection domain that invention protection domain should be defined by claims is defined.
Claims (6)
1. a kind of cloud test environment dispatching method based on load estimation, it is characterised in that
Including:
Node selection strategy module (2), for configuring the selection strategy of physical node;
Test main frame scheduler module (1), for dispatching the test main frame in cloud test environment on physical node;
Test main frame load prediction module (3), for predicting the load of the test main frame in cloud test environment;
Test main frame load delta data storehouse (4), is connected with the test main frame load prediction module (3), described for storing
The load delta data of test main frame;
Cloud test environment module (5), for running the test main frame;
Comprise the following steps:
Step one:Test main frame load delta data storehouse (4) is set up, and cloud test is configured by node selection strategy module (2)
The selection strategy of the physical node in environment module (5), selection strategy are specifically preserved in the form of configuration file;
Step 2:Test main frame scheduler module (1) obtain each physical node in the cloud test environment module (5) it is current when
Between average load in section, the length of the time period is setting in advance;
Step 3:The test main frame scheduler module (1) judges the cloud according to the average load and the threshold ones of setting
The test main frame for being migrated is needed in test environment module (5) on physical node;
Step 4:According to load delta data storehouse (4), test main frame load prediction module (3) is to the cloud test environment
The load of the physical node in module (5) is predicted, and will predict the outcome and feed back to the test main frame scheduler module (1),
The load estimation operation is comprised the following steps:Step A1:The test main frame load prediction module (3) obtains the physical node
On each test main frame historic load;Step A2:The test main frame load prediction module (3) is by each test main frame
The load that historic load and the test main frame are loaded in delta data storehouse (4) is compared, and is that each test main frame finds
The load change curves matched somebody with somebody;Step A3:The test main frame load prediction module (3) is according to the negative of the matching of each test main frame
The prediction load that change curve obtains each test main frame is carried, the prediction load addition of each test main frame is obtained into physical node
Prediction load;
Step 5:According to predicting the outcome, the test main frame scheduler module (1) is according to node selection strategy, it would be desirable to scheduling
Test main frame moves to the physical node of selection.
2. the cloud test environment dispatching method based on load estimation as claimed in claim 1, it is characterised in that further include
Step 6:If the average load of physical node disables the node less than the threshold ones of default in step 2.
3. the cloud test environment dispatching method based on load estimation as claimed in claim 1, it is characterised in that the step 2
In, the average load includes that CPU usage, memory usage, disk utilization rate and the network bandwidth take.
4. the cloud test environment dispatching method based on load estimation as claimed in claim 1, it is characterised in that the step 3
In, if the average load of the physical node is less than the threshold ones of default, all test masters on the physical node
Machine needs migration, and the physical node is disabled after migration;If the average load of the physical node is more than the high threshold of default
Value, then constantly select the test main frame for taking resource maximum to be migrated, until the load of the physical node is less than high threshold
Till.
5. the cloud test environment dispatching method based on load estimation as claimed in claim 1, it is characterised in that the step 5
In, the test main frame migration operation includes:
Step B1:For the test main frame of each needs migration, the test main frame scheduler module (1) is in the cloud test environment
Physical node is selected in module (5) so that load of the load estimation value of the physical node plus test main frame to be migrated
The scope of predictive value is between the threshold ones and high threshold of default, if without suitable physical node, not moving
Move, if there are multiple physical nodes to meet, physical node is selected according to the selection strategy of physical node;
Step B2:The test main frame scheduler module (1) migrates test main frame to be migrated to selected physical node.
6. a kind of cloud test environment based on load estimation dispatches system, it is characterised in that include:
Node selection strategy module (2), for configuring the selection strategy of physical node;
Test main frame scheduler module (1), for dispatching the test main frame in cloud test environment on physical node;
Test main frame load prediction module (3), for predicting the load of the test main frame in cloud test environment;
Test main frame load delta data storehouse (4), is connected with the test main frame load prediction module (3), described for storing
The load delta data of test main frame;
Cloud test environment module (5), for running the test main frame;
Wherein, the node selection strategy module (2) is connected with the test main frame scheduler module (1);The test main frame is adjusted
Degree module (1) is connected with the test main frame load prediction module (3) and the cloud test environment module (5) respectively;The survey
Examination host load prediction module (3) loads delta data storehouse (4) and the cloud test environment module with the test main frame respectively
(5) connect.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310003538.3A CN103916438B (en) | 2013-01-06 | 2013-01-06 | Cloud testing environment scheduling method and system based on load forecast |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310003538.3A CN103916438B (en) | 2013-01-06 | 2013-01-06 | Cloud testing environment scheduling method and system based on load forecast |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103916438A CN103916438A (en) | 2014-07-09 |
CN103916438B true CN103916438B (en) | 2017-04-12 |
Family
ID=51041841
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310003538.3A Active CN103916438B (en) | 2013-01-06 | 2013-01-06 | Cloud testing environment scheduling method and system based on load forecast |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103916438B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104333569A (en) * | 2014-09-23 | 2015-02-04 | 同济大学 | Cloud task scheduling algorithm based on user satisfaction |
CN106547607B (en) * | 2015-09-17 | 2020-02-11 | 中国移动通信集团公司 | Dynamic migration method and device for virtual machine |
CN107562512B (en) * | 2016-07-01 | 2020-06-26 | 华为技术有限公司 | Method, device and system for migrating virtual machine |
CN106201718A (en) * | 2016-07-05 | 2016-12-07 | 北京邮电大学 | A kind of cloud computing resources dynamic retractility method based on load estimation |
CN106293944B (en) * | 2016-08-11 | 2019-12-10 | 上海交通大学 | non-consistency-based I/O access system and optimization method under virtualized multi-core environment |
CN106899660B (en) * | 2017-01-26 | 2021-05-14 | 华南理工大学 | Cloud data center energy-saving scheduling implementation method based on rolling grey prediction model |
CN107220108B (en) * | 2017-07-06 | 2020-06-30 | 山东超越数控电子股份有限公司 | Method and system for realizing load balance of cloud data center |
CN109194703B (en) * | 2018-06-29 | 2021-06-18 | 平安科技(深圳)有限公司 | Processing method of communication load between cloud platform hosts, electronic device and medium |
CN110196807A (en) * | 2019-05-21 | 2019-09-03 | 苏州浪潮智能科技有限公司 | A kind of implementation method and device of wisdom load |
CN117272329A (en) * | 2023-11-21 | 2023-12-22 | 明阳时创(北京)科技有限公司 | Distributed virtualization vulnerability-based scanning method and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102236582A (en) * | 2011-07-15 | 2011-11-09 | 浙江大学 | Method for balanced distribution of virtualization cluster load in a plurality of physical machines |
CN102790793A (en) * | 2012-05-08 | 2012-11-21 | 北京邮电大学 | Decision method and control module facing to cloud computing virtual machine migration |
CN103685542A (en) * | 2013-12-23 | 2014-03-26 | 重庆广播电视大学 | Method, device and system for migrating cloud virtual machine |
-
2013
- 2013-01-06 CN CN201310003538.3A patent/CN103916438B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102236582A (en) * | 2011-07-15 | 2011-11-09 | 浙江大学 | Method for balanced distribution of virtualization cluster load in a plurality of physical machines |
CN102790793A (en) * | 2012-05-08 | 2012-11-21 | 北京邮电大学 | Decision method and control module facing to cloud computing virtual machine migration |
CN103685542A (en) * | 2013-12-23 | 2014-03-26 | 重庆广播电视大学 | Method, device and system for migrating cloud virtual machine |
Also Published As
Publication number | Publication date |
---|---|
CN103916438A (en) | 2014-07-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103916438B (en) | Cloud testing environment scheduling method and system based on load forecast | |
CN108829494B (en) | Container cloud platform intelligent resource optimization method based on load prediction | |
Kumar et al. | Renewable energy-based multi-indexed job classification and container management scheme for sustainability of cloud data centers | |
CN108182105B (en) | Local dynamic migration method and control system based on Docker container technology | |
CN102546700B (en) | Resource scheduling and resource migration methods and equipment | |
CN102932279B (en) | A kind of cloud environment data center multi dimensional resource dispatching method | |
CN110231976B (en) | Load prediction-based edge computing platform container deployment method and system | |
CN110221915B (en) | Node scheduling method and device | |
CN102904955B (en) | The self-adapting stretching control system of Web application in cloud computing platform and method thereof | |
CN111966453B (en) | Load balancing method, system, equipment and storage medium | |
CN103473139A (en) | Virtual machine cluster resource allocation and scheduling method | |
CN106020934A (en) | Optimized deploying method based on virtual cluster online migration | |
CN104035542B (en) | A kind of virtual machine migration method and system calculated with energy consumption for cooling balance | |
CN104572307A (en) | Method for flexibly scheduling virtual resources | |
CN106201700A (en) | The dispatching method that a kind of virtual machine migrates online | |
CN101715003A (en) | Load control optimization method of interoperable platform | |
CN104375897A (en) | Cloud computing resource scheduling method based on minimum relative load imbalance degree | |
CN103957261A (en) | Cloud computing resource distributing method based on energy consumption optimization | |
WO2015051685A1 (en) | Task scheduling method, device and system | |
CN105893113A (en) | Management system and management method of virtual machine | |
CN107861796A (en) | A kind of dispatching method of virtual machine for supporting cloud data center energy optimization | |
CN104572279B (en) | A kind of virtual machine dynamic dispatching method of supporting node binding | |
CN104199739A (en) | Speculation type Hadoop scheduling method based on load balancing | |
CN103677960A (en) | Game resetting method for virtual machines capable of controlling energy consumption | |
CN106201693A (en) | Dispatching method in a kind of virtualized environment and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |