CN104572304A - Method for intelligently maintaining and managing cluster rendering tasks in balance manner - Google Patents
Method for intelligently maintaining and managing cluster rendering tasks in balance manner Download PDFInfo
- Publication number
- CN104572304A CN104572304A CN201510039489.8A CN201510039489A CN104572304A CN 104572304 A CN104572304 A CN 104572304A CN 201510039489 A CN201510039489 A CN 201510039489A CN 104572304 A CN104572304 A CN 104572304A
- Authority
- CN
- China
- Prior art keywords
- task
- rendering
- dimension pipe
- supervision
- rendering task
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Image Generation (AREA)
Abstract
The invention belongs to the field of cluster rendering cloud computing, and provides a method for intelligently maintaining and managing cluster rendering tasks in a balance manner. The method has the advantages that an intelligent rendering task maintaining and managing coordinating module is set in a cluster rendering system and is responsible for sequential connection between maintaining and managing staffs and running rendering nodes; rendering supervision tasks are dynamically and sequentially distributed to the maintaining and managing staffs by the aid of adaptive selection strategies on the basis of task balance of the maintaining and managing staffs, accordingly, the work efficiency and quality of the maintaining and managing staffs can be optimally improved, and the high running efficiency of the rendering working nodes can be maintained; the method can be applied to constructing, optimizing and upgrading the Saas (software as a service) cluster rendering system and improving the internal management efficiency and the integral running efficiency of the system.
Description
Technical field
The present invention relates to Cluster Rendering field of cloud calculation field, particularly relate to a kind of Cluster Rendering task dimension pipe method of intelligent equalization.
Background technology
Adopt figure to surpass calculation cluster and the important support that output has become cartoon making and film later stage is played up to animation, film scene.It not only can realize the significantly lifting of rendering efficiency, also effectively can improve the level of resources utilization of expensive calculation server.Although Cluster Rendering system can automatically be played up and Output rusults according to the tasks carrying that enters, find in actual motion, animation or film special efficacy render process can not no supervisions completely, otherwise may not obtain the Output rusults of entirely accurate.Such as due to some scene and script formulation, the rendering result exported does not conform to expectation, or some script formulation can cause general renderer operational failure.These all need supervisor's Timeliness coverage, and redress rendering task, reschedule execution, otherwise completely etc. to be rendered complete after give user and could find.So, externally service quality performance is bad on the one hand, also can reduce the real work efficiency of system on the other hand to a great extent.In this context, how to improve the supervisory efficiency of dimension pipe personnel, the important technological problems building Cluster Rendering system is optimized in the efficient work that can maintain again rendering node.
Summary of the invention
The present invention is directed to above-mentioned prior art Problems existing to make improvements, namely the technical problem to be solved in the present invention is to provide a kind of many attribute judgement vertical handoff methods, this method improves efficiency of service and the quality of dimension pipe personnel substantially, and maintain the high efficiency of Cluster Rendering system works node, in order to solve the problems of the technologies described above, the invention provides following technical scheme:
A Cluster Rendering task dimension pipe method for intelligent equalization, in cluster computing system, arrange an intelligentized rendering task dimension pipe Coordination module, for dimension pipe, personnel distribute task of supervision intelligently automatically; Meanwhile, adopt the adaptively selected strategy based on dimension pipe worker tasks equilibrium, improve work efficiency and the quality of dimension pipe personnel substantially, and maintain the operation high efficiency of Cluster Rendering working node.
Specifically comprise the steps:
S1, in Cluster Rendering system, arrange the rendering task dimension pipe Coordination module of a pure software, it connects rendering task input, dimension pipe personnel task of supervision accepts, rendering task dispatches three aspects;
The rendering task dimension pipe Coordination module of S2, setting constantly accepts the request of user's rendering task, according to dimension pipe worker tasks homeostatic principle, new task is distributed to suitable dimension pipe personnel simultaneously, and then be distributed to rendering task scheduler module, be dispatched to concrete rendering node execution by it and play up;
The rendering task dimension pipe Coordination module of S3, setting is that each dimension pipe personnel set up a task of supervision dynamically arrangement table, and calculates its task of supervision according to task of supervision arrangement table and make a gesture of measuring;
When the rendering task dimension pipe Coordination module of S4, setting works, if when having new rendering task need supervise distribution, first its task of supervision amount is estimated according to rendering task amount, then task of supervision is selected to make a gesture of measuring minimum dimension pipe personnel, if its existing task of supervision amount adds that task amount to be regulated is less than maximum single task of supervision amount, then this new rendering task is distributed to this dimension pipe personnel and carry out supervision execution, otherwise new rendering task is waited for;
The rendering task dimension pipe Coordination module of S5, setting for the task of supervision situation of each dimension pipe personnel, to after the rendering task that enters, adopt random interval strategy backward to arrange its Starting Executing Time;
The rendering task dimension pipe Coordination module of S6, setting, according to rendering task implementation status, dynamically updates the current task of supervision dynamically arrangement table of each dimension pipe personnel.
The invention has the beneficial effects as follows: adopt the profile error realizing system dimension pipe personnel and rendering node in Cluster Rendering Operation system setting intelligentized rendering task dimension pipe Coordination module, farthest promote the actual operating efficiency of Cluster Rendering system; Dimension pipe Coordination module optimally distributes the supervision of new rendering task according to the current task of supervision arrangement table of each dimension pipe personnel, the workload making each dimension pipe personnel mutual is at equilibrium, make the working load of each dimension pipe personnel be in the normal state equilibrium level simultaneously, farthest eliminate system cloud gray model bottleneck, obtain the maximization of Cluster Rendering running efficiency.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, together with embodiments of the present invention for explaining the present invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the system construction drawing of the inventive method;
Fig. 2 is the fundamental diagram of the new rendering task T of the inventive method process;
Fig. 3 is the task arrangement list structure figure of the dimension pipe personnel described in the inventive method;
Fig. 4 is that in mode working method of the present invention, certain dimension pipe personnel accept the task arrangement list structure figure after new rendering task Tk+1.
Embodiment
The present invention discloses a kind of Cluster Rendering task dimension pipe method of intelligent equalization, and in cluster computing system, arrange an intelligentized rendering task dimension pipe Coordination module, for dimension pipe, personnel distribute task of supervision intelligently automatically; Crucially adopt the adaptively selected strategy based on dimension pipe worker tasks equilibrium, improve work efficiency and the quality of dimension pipe personnel substantially, and maintain the operation high efficiency of Cluster Rendering working node.
The rendering task dimension pipe Coordination module arranged continues to accept the request of user's rendering task, distribute dimension pipe task according to dimension pipe worker tasks homeostatic principle simultaneously, and the rendering task that preparation performs is distributed to task scheduling modules, be dispatched to concrete rendering node execution by it and play up.
The major function of rendering task dimension pipe Coordination module is: 1) for each dimension pipe personnel set up a task of supervision dynamically arrangement table, and calculates its task of supervision according to task of supervision arrangement table and make a gesture of measuring; 2) when there being new rendering task to enter, its task of supervision amount is estimated according to rendering task amount; Select task of supervision to make a gesture of measuring minimum dimension pipe personnel, if its existing task of supervision amount adds that task amount to be regulated is less than maximum single task of supervision amount, then new rendering task is distributed to this dimension pipe personnel and carry out supervision execution, otherwise new rendering task is waited for; 3) for the task of supervision situation of each dimension pipe personnel, to after the rendering task that enters, adopt random interval strategy backward to arrange its Starting Executing Time; 4) according to rendering task implementation status, the current task of supervision dynamically arrangement table of each dimension pipe personnel is dynamically updated.
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described further.
As shown in Figure 1, the inventive method structure for arranging task dimension pipe Coordination module in Cluster Rendering computing system, this module in charge receives client's rendering task, can allocating task situation according to dimension pipe personnel, distributed to suitable dimension pipe personnel, transferred to by rendering task task scheduling modules to dispatch simultaneously and play up.
As shown in Figure 2, in the principle of work of the inventive method: after new rendering task T generates, it is first by the process of dimension pipe Coordination module, perform dimension pipe task matching, the task that entered after being allocated successfully is played up scheduler module preparation and is played up, after playing up scheduler module successful dispatch, to the rendering node information feed back of this task be performed to dimension pipe Coordination module, rendering node information pushing is given dimension pipe personnel by dimension pipe Coordination module further, then tie up the implementation status of playing up of pipe personnel to this task to supervise in overall process, be successfully completed until play up.
As shown in Figure 3, in the principle of work of the inventive method, the task arrangement table representation of dimension pipe personnel is: the timeslice being divided into interval 1-3 minute the working time of dimension pipe personnel, and each timeslice points to a task of supervision; In a period of time interval, dimension pipe personnel supervise the different operating process of each task respectively in each timeslice, and realization is simultaneously responsible multiple plays up the supervision and maintenance work of executing the task.
As shown in Figure 4, in the working method of the inventive method, certain dimension pipe personnel are after accepting new rendering task Tk+1, the task of supervision arrangement table situation of change of this dimension pipe personnel, the supervision timeslice of new task Tk+1 is set to not conflict with task of supervision timeslice, and leaves random interval each other when task is made a gesture of measuring very not large.
To sum up, the present invention adopts the profile error realizing system dimension pipe personnel and rendering node in Cluster Rendering Operation system setting intelligentized rendering task dimension pipe Coordination module, farthest promotes the actual operating efficiency of Cluster Rendering system; Dimension pipe Coordination module optimally distributes the supervision of new rendering task according to the current task of supervision arrangement table of each dimension pipe personnel, the workload making each dimension pipe personnel mutual is at equilibrium, make the working load of each dimension pipe personnel be in the normal state equilibrium level simultaneously, farthest eliminate system cloud gray model bottleneck, obtain the maximization of Cluster Rendering running efficiency; In addition, the present invention can be applicable to Cluster Rendering system constructing and the optimization and upgrading of SaaS (namely software serve), the inner management of elevator system and overall operation efficiency.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, although with reference to previous embodiment to invention has been detailed description, for a person skilled in the art, it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (2)
1. a Cluster Rendering task dimension pipe method for intelligent equalization, is characterized in that, in cluster computing system, arrange an intelligentized rendering task dimension pipe Coordination module, for dimension pipe, personnel distribute task of supervision intelligently automatically; Meanwhile, the adaptively selected strategy based on dimension pipe worker tasks equilibrium is adopted.
2. the Cluster Rendering task dimension pipe method of intelligent equalization according to claim 1, is characterized in that, specifically comprise the steps:
S1, in Cluster Rendering system, arrange the rendering task dimension pipe Coordination module of a pure software, it connects rendering task input, dimension pipe personnel task of supervision accepts, rendering task dispatches three aspects;
The rendering task dimension pipe Coordination module of S2, setting constantly accepts the request of user's rendering task, according to dimension pipe worker tasks homeostatic principle, new task is distributed to suitable dimension pipe personnel simultaneously, and then be distributed to rendering task scheduler module, be dispatched to concrete rendering node execution by it and play up;
The rendering task dimension pipe Coordination module of S3, setting is that each dimension pipe personnel set up a task of supervision dynamically arrangement table, and calculates its task of supervision according to task of supervision arrangement table and make a gesture of measuring;
When the rendering task dimension pipe Coordination module of S4, setting works, if when having new rendering task need supervise distribution, first its task of supervision amount is estimated according to rendering task amount, then task of supervision is selected to make a gesture of measuring minimum dimension pipe personnel, if its existing task of supervision amount adds that task amount to be regulated is less than maximum single task of supervision amount, then this new rendering task is distributed to this dimension pipe personnel and carry out supervision execution, otherwise new rendering task is waited for;
The rendering task dimension pipe Coordination module of S5, setting for the task of supervision situation of each dimension pipe personnel, to after the rendering task that enters, adopt random interval strategy backward to arrange its Starting Executing Time;
The rendering task dimension pipe Coordination module of S6, setting, according to rendering task implementation status, dynamically updates the current task of supervision dynamically arrangement table of each dimension pipe personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039489.8A CN104572304B (en) | 2015-01-26 | 2015-01-26 | The Cluster Rendering task dimension pipe method of intelligent equalization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039489.8A CN104572304B (en) | 2015-01-26 | 2015-01-26 | The Cluster Rendering task dimension pipe method of intelligent equalization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104572304A true CN104572304A (en) | 2015-04-29 |
| CN104572304B CN104572304B (en) | 2018-02-02 |
Family
ID=53088449
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510039489.8A Active CN104572304B (en) | 2015-01-26 | 2015-01-26 | The Cluster Rendering task dimension pipe method of intelligent equalization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104572304B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127678A (en) * | 2016-06-23 | 2016-11-16 | 北京天文馆 | The cluster platform for making system of digital ball-screen cinema and method of work |
| CN107682220A (en) * | 2017-09-15 | 2018-02-09 | 宁波博恩电气有限公司 | Method for inspecting for power equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060095912A1 (en) * | 2004-11-04 | 2006-05-04 | Pixar | Renderfarm monitoring system |
| CN103208132A (en) * | 2012-08-10 | 2013-07-17 | 天津十彩动画科技有限公司 | Cluster animation rendering numerical control system |
| CN103268253A (en) * | 2012-02-24 | 2013-08-28 | 苏州蓝海彤翔系统科技有限公司 | Intelligent scheduling management method for multi-scale parallel rendering jobs |
| CN103546313A (en) * | 2013-09-26 | 2014-01-29 | 福州宙斯盾信息技术有限公司 | Cloud computing based IT (information technology) operation and maintenance management system |
| CN103617062A (en) * | 2013-12-05 | 2014-03-05 | 山东大学 | Flexible dynamic deployment system and method for render cluster |
-
2015
- 2015-01-26 CN CN201510039489.8A patent/CN104572304B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060095912A1 (en) * | 2004-11-04 | 2006-05-04 | Pixar | Renderfarm monitoring system |
| CN103268253A (en) * | 2012-02-24 | 2013-08-28 | 苏州蓝海彤翔系统科技有限公司 | Intelligent scheduling management method for multi-scale parallel rendering jobs |
| CN103208132A (en) * | 2012-08-10 | 2013-07-17 | 天津十彩动画科技有限公司 | Cluster animation rendering numerical control system |
| CN103546313A (en) * | 2013-09-26 | 2014-01-29 | 福州宙斯盾信息技术有限公司 | Cloud computing based IT (information technology) operation and maintenance management system |
| CN103617062A (en) * | 2013-12-05 | 2014-03-05 | 山东大学 | Flexible dynamic deployment system and method for render cluster |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127678A (en) * | 2016-06-23 | 2016-11-16 | 北京天文馆 | The cluster platform for making system of digital ball-screen cinema and method of work |
| CN107682220A (en) * | 2017-09-15 | 2018-02-09 | 宁波博恩电气有限公司 | Method for inspecting for power equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104572304B (en) | 2018-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104572305A (en) | Load-balanced cluster rendering task dispatching method | |
| CN105049536B (en) | SiteServer LBS and load-balancing method in IaaS cloud environment | |
| CN101237469B (en) | Method for optimizing multi-QoS grid workflow based on ant group algorithm | |
| CN105159762B (en) | Heuristic cloud computing method for scheduling task based on Greedy strategy | |
| CN104331321B (en) | Cloud computing task scheduling method based on tabu search and load balancing | |
| CN104731657B (en) | A kind of resource regulating method and system | |
| CN104065745A (en) | Cloud computing dynamic resource scheduling system and method | |
| CN104657220A (en) | Model and method for scheduling for mixed cloud based on deadline and cost constraints | |
| CN104572307A (en) | Method for flexibly scheduling virtual resources | |
| CN103679388B (en) | Production scheduling method and production scheduling system | |
| CN106020934A (en) | Optimized deploying method based on virtual cluster online migration | |
| CN104239144A (en) | Multilevel distributed task processing system | |
| CN103731372A (en) | Resource supply method for service supplier under hybrid cloud environment | |
| CN106371924B (en) | A kind of method for scheduling task minimizing MapReduce cluster energy consumption | |
| CN110109756A (en) | A kind of network target range construction method, system and storage medium | |
| CN108170530A (en) | A kind of Hadoop Load Balancing Task Scheduling methods based on mixing meta-heuristic algorithm | |
| CN103257900B (en) | Real-time task collection method for obligating resource on the multiprocessor that minimizing CPU takies | |
| CN106293933A (en) | A kind of cluster resource configuration supporting much data Computational frames and dispatching method | |
| CN102185726B (en) | Method for improving business resource management capability in information grid system | |
| CN103268253A (en) | Intelligent scheduling management method for multi-scale parallel rendering jobs | |
| CN112559122A (en) | Virtualization instance management and control method and system based on electric power special security and protection equipment | |
| CN106027318A (en) | Cloud computing-based two-level optimal scheduling management platform for virtual machine | |
| CN105005503A (en) | Cellular automaton based cloud computing load balancing task scheduling method | |
| CN104572304A (en) | Method for intelligently maintaining and managing cluster rendering tasks in balance manner | |
| CN105426247A (en) | HLA federate planning and scheduling method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 213022 Jiangsu city of Changzhou province Taihu New District Road No. 23 building 9-2 Applicant after: JIANGSU CUDATEC CO., LTD. Address before: 213022 Jiangsu city of Changzhou province Taihu New District Road No. 23 building 9-2 Applicant before: Cudatec Development Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |