CN108900583A - A kind of SiteServer LBS and equalization methods of intelligent automation - Google Patents
A kind of SiteServer LBS and equalization methods of intelligent automation Download PDFInfo
- Publication number
- CN108900583A CN108900583A CN201810622285.0A CN201810622285A CN108900583A CN 108900583 A CN108900583 A CN 108900583A CN 201810622285 A CN201810622285 A CN 201810622285A CN 108900583 A CN108900583 A CN 108900583A
- Authority
- CN
- China
- Prior art keywords
- control module
- sub
- task
- son
- submodule
- 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
-
- 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
- H04L67/63—Routing a service request depending on the request content or context
Abstract
The invention discloses a kind of SiteServer LBS of intelligent automation and its equalization methods, load balance process for multitask high concurrent, system carries out task schedule according to each task execution node current state and realizes load balancing, and combine the realization of cloud computing virtualization technology that part idle node is automatically closed when the system is idle, it automatically turns on new node when the system is overloaded to be added in load balancing network, to realize the purpose of dynamic resource scheduling.By the dynamic management to system resource, resource is saved, has reduced costs.
Description
Technical field
The invention belongs to field of computer technology, are related to load-balancing technique and cloud computing virtualization technology, can be used for more
The load balance process of task high concurrent, and in particular to a kind of SiteServer LBS and its equalization methods of intelligent automation.
Background technique
Load (task, access request) is exactly balanced, shares the (service of multiple operating units by load balancing
Device, component) on executed.It is the ultimate solution party for solving high-performance, Single Point of Faliure (High Availabitity), scalability (horizontal extension)
Case.
Many products with load-balancing function have been had already appeared currently on the market, and Nginx, LVS, HAProxy are mesh
Preceding most popular three kinds of load balancing software.Nginx is a high performance Web and Reverse Proxy, as negative
Carry equalization server Nginx can according to scheduling rule realize dynamic, static page separation, can according to poll, ip Hash,
The various ways such as URL Hash, weight do load balancing to back-end server, while also supporting the health examination of back-end server.
LVS is the abbreviation of Linux Virtual Server, that is, Linux virtual server.It has good reliability, can expand
Malleability and operability, to realize optimal performance with cheap cost.LVS is the open source of a realization load balancing cluster
Software project, LVS framework can logically be divided into dispatch layer, Server cluster layer and shared storage.HAProxy is a use
The freedom and open-source software that C language is write provide high availability, load balancing, and answering based on TCP and HTTP
Use programmatic agent.
But the prior art and product are only able to achieve and load to node that is existing, having disposed corresponding service application
Weighing apparatus processing, can only manually be established if adding new node new node, disposed corresponding service application modify load balancing group again
It configures, new node could be added into load balancing network accordingly in part, and can not achieve automatic addition.Also, works as and be
System pressure becomes smaller, and when node state becomes the free time, can not cut off idle node, cause the wasting of resources.
Summary of the invention
In view of the deficienciess of the prior art, the purpose of the present invention is to propose to a kind of SiteServer LBSs of intelligent automation
And its equalization methods, task schedule is carried out according to each task execution node current state and realizes load balancing, and combines cloud computing
Idle node is automatically closed in virtualization technology realization when the system is idle, automatically turns on new node when the system is overloaded and load is added
In equalising network, realizes dynamic resource scheduling, economize on resources, reduce cost.
To achieve the goals above, the technical scheme adopted by the invention is as follows:A kind of load balancing system of intelligent automation
System, including main control module, sub- control module and cloud platform management module, wherein:
Main control module is used to receive the task execution quantity that each height control module is sent, and appointing according to all sub- control modules
Business number of executions judges current system operating status:When the average task execution quantity of every height control module is greater than the overload of setting
When threshold value, system running state is overload, and cloud platform management module is given in the instruction for sending the new son control module of unlatching;When every height control
When the average task execution quantity of module is less than the Tidle threshold of setting, timing is carried out, system begins in defined timing time
When the case where meeting Tidle threshold eventually, system running state is the free time, and cloud platform pipe is given in the instruction for sending the new son control module of closing
Manage module;
The service operation state sent simultaneously for receiving each height control module, the service operation state of group control module are
When delay machine, the son control module of delay machine is rejected;
It is also used in the task requests for receiving upper layer application, different son control moulds is dispatched to according to load balancing
Block processing.
Further, the main control module includes sub- control degree submodule, dispatching platforms submodule and condition monitoring submodule
Block, wherein:
Sub- control degree submodule is used for the task requests to each height control module schedules upper layer application;Dispatching platforms submodule
Block, for sending the instruction for opening or closing sub- control module to cloud platform management module;Condition monitoring submodule, for receiving son
Module operating status is controlled, and judges the operating status of system according to the task execution quantity of each height control module.
Further, the sub- control module include state report submodule and task processing submodule, wherein:
State report submodule is used to report self-operating state to main control module, including:Service operation state and task
Number of executions, when group control module normal operation, otherwise it is delay machine that service operation state, which is OK,;Task handles submodule and uses
In the task requests of processing main control module scheduling.
Further, the cloud platform management module, including Virtual Machine Manager submodule and mirror image manage submodule,
In:
Virtual Machine Manager submodule is used to receive the control instruction of main control module, and antithetical phrase control module is opened and closed;
Mirror image management submodule is for Mirror Info needed for providing the new son control module of unlatching.
It is dispatched to that different son control resume modules are specific as follows according to load balancing, current according to each sub- control module
Execution status of task therefrom chooses the smallest sub- control module of task amount and goes to handle new request, and there are the identical sub- control moulds of task amount
It selects newly to request in the forward son control resume module of sub- control queue sequence when block.
A kind of load-balancing method of intelligent automation, includes the following steps:
Step 1. activation system;
Step 2. main control module opens the monitoring requested to upper layer application and the task requests for receiving upper layer application;
Step 3. main control module analyzes the operating status of each sub- control module operating status and system, when all sub- control modules
When performed task average is greater than the overloading threshold of setting, system running state is overload, and enters step 5 processing;
When the task average performed by all sub- control modules is less than the Tidle threshold of setting, timing is carried out, defined
When system meets the case where Tidle threshold always in timing time, then defines system and be in Idle state, enter step 4 processing;
The processing of step 4. Idle state
When the system is idle, the label least N number of sub- control module of task amount is that idle son controls module, and N >=1 is not reallocated
New task is to idle son control module, until main control module sends the finger for closing idle son control module after its task is disposed
Cloud platform management module is enabled to complete the closing to idle son control module;
Step 5. overloads state processing
When the system is overloaded, cloud platform management module, cloud platform are given in the instruction that main control module sends the new son control module of unlatching
Management module opens new son control module.
Further, the step 1 specifically includes:
Main control module is disposed and is fabricated to master control mirror image by the first step, opens one using master control mirror image in cloud platform
A virtual machine is as main controlled node, and after main controlled node starting, main control module monitors son control module status information;
Sub- control module and corresponding service application are disposed good and are fabricated to sub- control mirror image by second step, are utilized in cloud platform
Son control mirror image opens several virtual machines as son control node in advance, and after sub- control node starting, sub- control module continues to main control module
Report itself service operation state and execution status of task;
Third step, main control module receive the status information of son control module, are then added into son if it is new son control module
Control module queues.
Further, in the step 5, cloud platform management module realizes antithetical phrase control by the opening and closing to virtual machine
The opening and closing of module;And son control module queues are added in the son control module of unlatching and execute task.
Compared with prior art, the present invention at least has following beneficial effect:
The present invention realizes the dynamic management of resource, and it is flat that cloud can be dynamically managed according to current system execution status of task
The virtual computing resource of task is handled in platform.Part idle node is automatically closed when the system is idle, when the system is overloaded automatically
It opens new node to be added in load balancing network, economizes on resources, reduces cost;Equalization methods of the invention can be suitable for a variety of
Cloud computing platform system, such as Eucalyptus, OpenStack, CloudStack, OpenNebula.The present invention passes through mirror image
Rapid deployment node realizes the fast construction of task processing node, improves the efficiency of the dynamic equalization of system.
Detailed description of the invention
Fig. 1 is architecture diagram of the invention;
Fig. 2 is sub- control degree flow chart of the invention;
Fig. 3 is dispatching platforms sub-process figure of the invention.
In figure:1, main control module;11, sub- control degree submodule;12, dispatching platforms submodule;13, condition monitoring submodule
Block;2, sub- control module;21, state report submodule;22, task handles submodule;3, cloud platform management module;31, virtual machine
Manage submodule;32, mirror image manages submodule.
Specific embodiment
Below in conjunction with attached drawing, the present invention will be described in detail:
Referring to Fig.1, the SiteServer LBS of intelligent automation of the present invention, including main control module 1, sub- control module 2 and Yun Ping
Platform management module 3.Wherein:
Main control module 1, including sub- control degree submodule 11, dispatching platforms submodule 12 and condition monitoring submodule 13 are used
In the request for receiving upper layer application, and different son control modules 2 is dispatched to according to load balancing and is handled.Monitoring son control module
2 state, discovery, which has, then rejects queue for the sub- control module 2 when sub- 2 delay machine of control module.According to performed by current sub- control module 2
Task scale, dynamic dispatching cloud computing virtual resource.Wherein, the concrete function of each module is as follows:
Sub- control degree submodule 11 goes the request of processing upper layer application for dynamic dispatching sub- control module 2, is supervised according to state
The state for controlling each sub- control module 2 that submodule 13 obtains carries out load balance process, specially:The service of group control module 2 is transported
The sub- control module 2 is eliminated into sub- control queue when row state is delay machine, otherwise the task execution shape current according to each sub- control module 2
State therefrom chooses the smallest sub- control module 2 of task amount and goes to handle new request, selection when sub- control module 2 identical there are task amount
New request is handled in the forward son control module 2 of sub- control queue sequence.It, will new son control when listening to the information of new son control module 2
Module 2 is added in sub- control queue.
Dispatching platforms submodule 12 is used for dynamic dispatching cloud platform virtual computing resource, and transfers to cloud platform management module 3
Realize the opening and closing of virtual machine in cloud platform.The current operating status of system is judged according to condition monitoring submodule 13, when
When system is in Idle state, scheduling cloud platform management module 3 closes partial virtual machine, when system is in overload state, then dispatches
Cloud platform management module 3 opens new virtual machine.
Condition monitoring submodule 13, the son control module 2 reported for receiving state report submodule 21 in sub- control module 2
State, including service operation state and execution status of task.When the report letter for not receiving sub- control module 2 that exceeds schedule time
When breath, then this sub- 2 service operation state of control module is set as delay machine.Setting system Tidle threshold and system overload threshold value, system
When Tidle threshold indicates that the task average performed by the sub- control module 2 each in system is less than the threshold value, show that system pressure is smaller;
When system overload threshold value indicates that the task average performed by the sub- control module 2 each in system is greater than the threshold value, show system pressure
It is larger.Monitoring system operating status carries out timing, in defined timing when there is the case where meeting system Tidle threshold
It when interior system meets the case where Tidle threshold always, then defines system and is in Idle state, once occurring not in timing course
The case where meeting system Tidle threshold, then stops timing and timing of being zeroed;It is fixed if there is the case where meeting system overload threshold value
Adopted system is in overload state.
Son control module 2, including state report submodule 21 and task handle submodule 22, adjust for handling main control module 1
Corresponding service application execution task is called in the request that degree comes, and by state report submodule 21, the shape into main control module 1
State monitoring submodule 13 reports the state of sub- control module 2, including service operation state and execution status of task.Group control module 2
When normal operation, otherwise it is delay machine that service operation state, which is OK, and execution status of task is that the sub- control module 2 is currently performed
Task quantity.
Task handles submodule 22, for handling the request come scheduled in 1 neutron control degree submodule 11 of main control module, opens
Opening new thread calls corresponding service application to complete desired executing for the task of request.
Cloud platform management module 3, including Virtual Machine Manager submodule 31 and mirror image manage submodule 32, for according to master control
The management to cloud platform virtual computing resource is realized in the scheduling of module 1, including when the system is idle, closes part free virtual
Machine opens new virtual machine, and manage mirror image when the system is overloaded, i.e., the official API of calling Openstack is mirrored storage
Get off, and record the relevant information of mirror image, such as mirror image name, when opening son control node, system provide son control Mirror Info to
Openstack determines to be opened virtual machine by which mirror image, realizes rapid deployment.
Wherein, Virtual Machine Manager submodule 31, for the scheduling result according to condition monitoring submodule 13 in main control module 1
Openstack cloud platform virtual computing resource is managed, the scheduling result that wherein condition monitoring submodule 13 is sent is URL form.
It receives URL request and obtains specific scheduling of resource content, call the official Nova API of Openstack platform to open and close empty
Quasi- calculate node, wherein mirror image needed for opening new node is provided by mirror image management submodule 32, the ginseng of resource needed for new node
Number is parsed from URL request and is obtained.
Mirror image manages submodule 32, for managing cloud platform mirror image, to pass through mirror image rapid deployment system.By main control module
1 deployment is good and is fabricated to the mirror image that can open virtual machine as master control mirror image, by sub- control module 2 and corresponding service application deployment
Sub- control mirror image is got well and be fabricated to, calls official's Glance API memory image of Openstack platform, and when opening new node
Corresponding mirror image is provided.
The detailed process of the load-balancing mechanism of intelligent automation of the present invention is described as follows:
It is divided into two sub-processes:Sub- control degree sub-process and platform scheduling sublayer process.
Process 1:Sub- control degree sub-process, as shown in Fig. 2,
Step 1, activation system,
The first step:Main control module 1 is disposed good and is fabricated to master control mirror image, master control mirror is utilized in Openstack platform
As opening a virtual machine as main controlled node.After main control module 1 starts, the status information of son control module 2 is monitored;
Second step:Sub- control module 2 and corresponding service application are disposed good and are fabricated to sub- control mirror image, it is flat in Openstack
Partial virtual machine is opened in advance as son depending on demand using son control mirror image in platform controls node.After son control node starting, it is deployed in son
The son control module 2 controlled on node continues to report oneself state information, including service operation state and task execution to main control module 1
State, otherwise it is delay machine that service operation state, which is OK, when sub- 2 normal operation of control module, and execution status of task is the sub- control module 2
Currently performed task amount;
Third step:Main control module 1 receives the status information of sub- control module 2, if it is new node then by corresponding sub- control mould
Sub- control queue is added in block 2;
4th step:Main control module 1 opens the monitoring requested upper layer application.
Step 2, processing request,
The first step:The request of the reception upper layer application of main control module 1;
Second step:Main control module 1 successively analyzes the state of each sub- control module 2, the clothes of group control module 2 from sub- control queue
When operating status of being engaged in is delay machine, which is rejected into queue.Otherwise the task execution shape current according to each sub- control module 2
State therefrom chooses the smallest sub- control module 2 of task amount and goes to handle new request, selection when sub- control module 2 identical there are task amount
New request is handled in the son control module 2 of node queue's front;
Third step:Selected son control module 2 opens a new thread and corresponding service application is called to complete desired by request
The task of execution.
Process 2:Dispatching platforms sub-process, as shown in figure 3,
Step 1, activation system,
With the step 1 of process 1;
Step 2, analysis system operating status,
The first step:Judge system running state, when there is the case where meeting system Tidle threshold, carries out timing, work as meter
When reaching threshold value, then decision-making system is currently at Idle state, once there is discontented pedal system Tidle threshold in timing course
The case where then stop timing and timing of being zeroed;Decision-making system is in overload if there is the case where meeting system overload threshold value
State;
Second step:If system is in Idle state, Idle state processing is carried out by step 3, if system is in overload state
Then overload state processing is carried out by step 4;
Step 3, Idle state is handled,
The first step:Main control module 1 chooses the smallest N number of sub- control module 2 of task amount as idle from sub- control module queues
Son control module, corresponding virtual machine are idle node, and reallocation new task does not control module to idle son, until idle son controls mould
The current task execution of block is completed;
Second step:Main control module 1 initiates the URL request including idle node IP information to cloud platform management module 3;
Third step:Cloud platform management module 3 parses URL request, and the ID of virtual machine is obtained according to the IP of idle node, calls
The official API that computer is managed in cloud platform closes the virtual machine of corresponding ID.
Step 4, overload state processing,
The first step:Master control selects the mirror image of new node to be started, i.e., the son control mirror image to complete in step 1, and sets
Determine resource distribution, including memory size, virtual nucleus number, disk space, these information are sent to Virtual Machine Manager submodule by URL
Block 31;
Second step:Virtual Machine Manager submodule 31 parses URL request, and relevant parameter is passed to Openstack platform
Official Nova API opens new virtual machine using sub- control mirror image;
Third step:After virtual machine starting, the son control module 2 newly opened reports the status information of itself to main control module 1;
4th step:After main control module 1 obtains the status information for the son control module 2 newly opened, sub- control module is added it to
Queue.
Above description is only example of the present invention, does not constitute any limitation of the invention.Obviously for this
It, all may be in the premise without departing substantially from the principle of the invention, structure after understanding summary of the invention and principle for the professional in field
Under, the amendment and improvement of algorithm are carried out, but these amendments and improvement based on inventive algorithm are in claim of the invention
Within protection scope.
Claims (8)
1. a kind of SiteServer LBS of intelligent automation, which is characterized in that including main control module (1), sub- control module (2) and cloud
Platform management module (3), wherein:
Main control module (1) is used to receive the task execution quantity that each height control module (2) sends, and according to all sub- control modules
Average task execution quantity judges current system operating status:When the average task execution quantity of every height control module is greater than setting
Overloading threshold when, system running state is overload, sends and opens the instruction of new son control module and give cloud platform management module (3);
When the average task execution quantity of every height control module is less than the Tidle threshold of setting, timing is carried out, in defined timing
When interior average task execution quantity is consistently less than Tidle threshold, system running state is the free time, sends and closes idle son control mould
Cloud platform management module (3) are given in the instruction of block;
The service operation state sent simultaneously for receiving each height control module (2), the service operation state of group control module are
When delay machine, son control module (2) of delay machine is eliminated into sub- control module queues;
It is also used in the task requests for receiving upper layer application, different son control modules is dispatched to according to load balancing
(2) it handles.
2. a kind of SiteServer LBS of intelligent automation as described in claim 1, which is characterized in that the main control module
It (1) include sub- control degree submodule (11), dispatching platforms submodule (12) and condition monitoring submodule (13), wherein:
Sub- control degree submodule (11) is used for the task requests to each height control module (2) scheduling upper layer application;Dispatching platforms
Module (12), for sending the instruction for opening or closing sub- control module to cloud platform management module (3);Condition monitoring submodule
(13), for receiving sub- control module operating status, and judge according to the task execution quantity of each height control module (2) fortune of system
Row state.
3. a kind of SiteServer LBS of intelligent automation as described in claim 1, which is characterized in that the sub- control module
(2) include state report submodule (21) and task processing submodule (22), wherein:
State report submodule (21) is used to report self-operating state to main control module (1), including:Service operation state and appoint
It is engaged in number of executions, when group control module (2) normal operation, otherwise it is delay machine that service operation state, which is OK,;Task handles submodule
Block (22) is used to handle the task requests of main control module (1) scheduling.
4. a kind of SiteServer LBS of intelligent automation as described in claim 1, which is characterized in that the cloud platform management
Module (3), including Virtual Machine Manager submodule (31) and mirror image management submodule (32), wherein:
Virtual Machine Manager submodule (31) is used to receive the control instruction of main control module (1), antithetical phrase control module (2) open and
It closes;Mirror image management submodule (32) is for Mirror Info needed for providing the new son control module of unlatching.
5. a kind of SiteServer LBS of intelligent automation according to claim 1, which is characterized in that according to load balancing
It is specific as follows that strategy is dispatched to different son control module (2) processing, according to the current execution status of task of each sub- control module (2),
It therefrom chooses the smallest sub- control module (2) of task amount to go to handle new request, whens sub- control module (2) identical there are task amount selects
In the forward new request of son control module (2) processing of sub- control queue sequence.
6. a kind of load-balancing method of intelligent automation, which is characterized in that include the following steps:
Step 1. activation system;
Step 2. main control module (1) opens the monitoring requested to upper layer application and the task requests for receiving upper layer application;
Step 3. main control module (1) analyzes the operating status of each sub- control module (2) operating status and system, when all sub- control moulds
When task average performed by block is greater than the overloading threshold of setting, system running state is overload, and enters step 5 processing;
When the task average performed by all sub- control modules is less than the Tidle threshold of setting, timing is carried out, in defined timing
When system meets the case where Tidle threshold always in time, then defines system and be in Idle state, enter step 4 processing;
The processing of step 4. Idle state
When the system is idle, the label least N number of sub- control module (2) of task amount is that idle son controls module, and N >=1 is not reallocated new
Task to idle son control module, until main control module (1) sends the finger for closing idle son control module after its task is disposed
Cloud platform management module (3) are enabled to complete the closing to idle son control module;
Step 5. overloads state processing
When the system is overloaded, cloud platform management module (3), Yun Ping are given in the instruction that main control module (1) sends the new son control module of unlatching
Platform management module (3) opens new son control module.
7. a kind of equalization methods of the SiteServer LBS of intelligent automation as claimed in claim 6, which is characterized in that the step
Rapid 1 specifically includes:
Main control module (1) is disposed good and is fabricated to master control mirror image by the first step, opens one using master control mirror image in cloud platform
Virtual machine is as main controlled node, and after main controlled node starting, main control module (1) monitors son control module (2) status information;
Sub- control module (2) and corresponding service application are disposed good and are fabricated to sub- control mirror image by second step, and son is utilized in cloud platform
Control mirror image opens several virtual machines as son control node in advance, and after sub- control node starting, sub- control module (2) continues to main control module
(1) itself service operation state and execution status of task are reported;
Third step, main control module (1) receive the status information of son control module (2), are then added if it is new son control module (2)
Enter sub- control module queues.
8. a kind of equalization methods of the SiteServer LBS of intelligent automation as claimed in claim 6, which is characterized in that the step
In rapid 5, factor control module subordinate is on virtual machine, therefore to pass through the opening and closing to virtual machine real for cloud platform management module (3)
The opening and closing of existing antithetical phrase control module (2);And son control module queues are added in the son control module of unlatching and execute task.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810622285.0A CN108900583A (en) | 2018-06-15 | 2018-06-15 | A kind of SiteServer LBS and equalization methods of intelligent automation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810622285.0A CN108900583A (en) | 2018-06-15 | 2018-06-15 | A kind of SiteServer LBS and equalization methods of intelligent automation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108900583A true CN108900583A (en) | 2018-11-27 |
Family
ID=64344944
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810622285.0A Pending CN108900583A (en) | 2018-06-15 | 2018-06-15 | A kind of SiteServer LBS and equalization methods of intelligent automation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108900583A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110716800A (en) * | 2019-10-09 | 2020-01-21 | 广州华多网络科技有限公司 | Task scheduling method and device, storage medium and electronic equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102427475A (en) * | 2011-12-08 | 2012-04-25 | 曙光信息产业(北京)有限公司 | Load balance scheduling system in cloud computing environment |
CN104202254A (en) * | 2014-08-14 | 2014-12-10 | 江苏省邮电规划设计院有限责任公司 | An intelligent load balancing method based on a cloud calculation platform server |
US20160100330A1 (en) * | 2014-10-03 | 2016-04-07 | At&T Intellectual Property I, L.P. | Scalable network function virtualization |
CN107341045A (en) * | 2017-07-13 | 2017-11-10 | 郑州云海信息技术有限公司 | A kind of scheduling virtual machine management method and scheduler |
-
2018
- 2018-06-15 CN CN201810622285.0A patent/CN108900583A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102427475A (en) * | 2011-12-08 | 2012-04-25 | 曙光信息产业(北京)有限公司 | Load balance scheduling system in cloud computing environment |
CN104202254A (en) * | 2014-08-14 | 2014-12-10 | 江苏省邮电规划设计院有限责任公司 | An intelligent load balancing method based on a cloud calculation platform server |
US20160100330A1 (en) * | 2014-10-03 | 2016-04-07 | At&T Intellectual Property I, L.P. | Scalable network function virtualization |
CN107341045A (en) * | 2017-07-13 | 2017-11-10 | 郑州云海信息技术有限公司 | A kind of scheduling virtual machine management method and scheduler |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110716800A (en) * | 2019-10-09 | 2020-01-21 | 广州华多网络科技有限公司 | Task scheduling method and device, storage medium and electronic equipment |
CN110716800B (en) * | 2019-10-09 | 2021-07-09 | 广州华多网络科技有限公司 | Task scheduling method and device, storage medium and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11934864B1 (en) | System and method for controlled sharing of consumable resources in a computer cluster | |
CN104657220B (en) | Scheduling model and method based on deadline and expense restriction in mixed cloud | |
CN107291547B (en) | Task scheduling processing method, device and system | |
CN110297699B (en) | Scheduling method, scheduler, storage medium and system | |
CN106445675B (en) | B2B platform distributed application scheduling and resource allocation method | |
CN108762896A (en) | One kind being based on Hadoop cluster tasks dispatching method and computer equipment | |
CN104123182B (en) | Based on the MapReduce task of client/server across data center scheduling system and method | |
CN109495398A (en) | A kind of resource regulating method and equipment of container cloud | |
US20190230044A1 (en) | System and Method for Optimizing Resource Utilization in a Clustered or Cloud Environment | |
CN114138486A (en) | Containerized micro-service arranging method, system and medium for cloud edge heterogeneous environment | |
CN107122233A (en) | A kind of adaptive real-time scheduling methods of many VCPU towards TSN business | |
US11838384B2 (en) | Intelligent scheduling apparatus and method | |
CN109905329A (en) | The flow queue adaptive management method that task type perceives under a kind of virtualized environment | |
CN104052677B (en) | The soft load-balancing method and device of data mapping | |
CN112162835A (en) | Scheduling optimization method for real-time tasks in heterogeneous cloud environment | |
CN108563495A (en) | The cloud resource queue graded dispatching system and method for data center's total management system | |
Sharma et al. | A Dynamic optimization algorithm for task scheduling in cloud computing with resource utilization | |
CN100477630C (en) | Queue dispatching method and apparatus in data network | |
CN108900583A (en) | A kind of SiteServer LBS and equalization methods of intelligent automation | |
Khabbaz et al. | Modelling and analysis of a novel deadline-aware scheduling scheme for cloud computing data centers | |
CN107766154A (en) | The conversion method and device of server | |
Hamzeh et al. | A new approach to calculate resource limits with fairness in kubernetes | |
CN107426109A (en) | A kind of traffic scheduling method, VNF modules and flow scheduling server | |
WO2014160954A1 (en) | System and method for network provisioning | |
CN109446641B (en) | Multi-stage reliability modeling analysis method of cloud computing service system |
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: 20181127 |
|
RJ01 | Rejection of invention patent application after publication |