CN107066338A - The computing environment method of automatic configuration of distributed computing system - Google Patents
The computing environment method of automatic configuration of distributed computing system Download PDFInfo
- Publication number
- CN107066338A CN107066338A CN201710240836.2A CN201710240836A CN107066338A CN 107066338 A CN107066338 A CN 107066338A CN 201710240836 A CN201710240836 A CN 201710240836A CN 107066338 A CN107066338 A CN 107066338A
- Authority
- CN
- China
- Prior art keywords
- task
- centroid
- terminal node
- configuration
- designated terminal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
Abstract
The present invention provides a kind of computing environment method of automatic configuration of distributed computing system.Technical scheme includes：The first step：Centroid is described and stored to task；Second step：Centroid collects each destination node information；3rd step：Centroid formulates task allocative decision；4th step：Centroid sends task bag and task configuration requirement；5th step：Designated terminal node performs task configuration；6th step：Centroid confirms task configuration result；7th step：Centroid, which is sent, starts task order；8th step：Terminal node starts task program process；9th step：Centroid confirms computing environment configuration result；Centroid collects the startup task result of designated terminal node, if task start is completed, computing environment configuration successful.The present invention is automatically configured suitable for the computing environment of all kinds of distributed computing systems, and distributed computing system rapid build.Test result shows that the present invention can greatly improve the service efficiency of system, and with stronger stability and applicability.
Description
Technical field
The invention belongs to Distributed Calculation field, it is related to the structure and administrative skill of distributed computing system, more particularly to
The computing environment method of automatic configuration of distributed computing system.
Background technology
With the development of science and technology, distributed computing technology is used widely in sorts of systems, is formed all kinds of
Distributed computing system.For example：Distributed satellites based on HLA (High Level Architecture, High Level Architecture)
Analogue system, distributed aviation Virtual Scene System etc..Distributed computing system is joined by many scattered computers via network
Connect and constitute, execution system management, the computer of distribution of computation tasks are referred to as Centroid in usual network, main to be responsible for and it
His computer sets up communication, collection and transmission message, transmission tasks resource；The computer for performing task computation is referred to as terminal section
Point, receives the task of Centroid distribution, and performs task computation, feedback result information.Distributed computing system running
In, information is exchanged between Centroid and terminal node, shared resource cooperates one common task of completion.
Due to the dispersing character of terminal node, it is to computing environment progress configuration in distributed computing system building process
The cumbersome work of one complexity.First, many people of the configuration requirement of computing environment carry out cooperating, example to scattered terminal node
Such as：Startup, setting, calculating task deployment, software merit rating of terminal node etc..But in practical application, in different spaces position
Communication has difficulties between the personnel put, it is difficult to ensure efficient cooperative cooperating.Secondly, the meter repeatedly tested to multisample
Calculate, operating personnel need for each pattern detection and configure computing environment, and the work of operating personnel is heavy.Above mentioned problem is deposited
So that the mode of human configuration computing environment can largely consume the structure time of distributed computing system, cause distributed meter
Calculation system service efficiency is low, and artificial cumbersome, fault-tolerance is poor.Therefore, it is necessary to propose a kind of distributed computing system
Computing environment method of automatic configuration, can reduce the operating personnel of distributed computing system, also can simplification computing environment configuration
Work, so as to improve the application efficiency and fault-tolerance of distributed computing system.
The content of the invention
The technical problem to be solved in the invention is：Overcome in the computing environment human configuration mode of distributed computing system
Artificial frequent operation, poor fault tolerance, the structure time is long, and the low problem of service efficiency proposes a kind of automatically configuring for computing environment
Method, reduces operating personnel and its workload, and the configuration work of simplification computing environment improves the service efficiency of distributed system.
The technical scheme is that, a kind of computing environment method of automatic configuration of distributed computing system, Centroid
The existence information transmission channel between each terminal node, it is characterised in that to any one tasks carrying following step：
The first step：Centroid is described and stored to task
Centroid forms mission bit stream and task bag according to the task data of reception, and mission bit stream and task bag are deposited
Store up database.Wherein, mission bit stream includes：The identification information of task, the function declaration of task, the configuration requirement of task, appoint
Business bag is the executable resource collection of task.
Second step：Centroid collects each destination node information
Centroid collects all destination node informations, includes identification information, computing capability, the task of each terminal node
Configuration record.All terminal nodes are set as available terminal node.
3rd step：Centroid formulates task allocative decision
Centroid is matched mission bit stream with all available terminal nodal informations, forms task allocative decision, is appointed
Allocative decision of being engaged in includes the designated terminal node of task and the configuration requirement of task.
4th step：Centroid sends task bag and task configuration requirement
Centroid sends out the mission bit stream in task configuration requirement and database, task bag according to task allocative decision
Give designated terminal node.
5th step：Designated terminal node performs task configuration
Designated terminal node receives mission bit stream, task bag and task configuration requirement, according to task configuration requirement to task
Performing environment is configured, and updates the task configuration record of designated terminal node.
6th step：Centroid confirms task configuration result
Whether task configuration record, which meets, on Centroid collection designated terminal nodal information, inspection designated terminal node is wanted
Ask
If meeting task configuration requirement, the 7th step is performed；
If being unsatisfactory for task configuration requirement, return the 4th step, i.e. Centroid resend task bag and task is matched somebody with somebody
Requirement is put, designated terminal node re-executes task configuration.If above-mentioned circulation is performed three times, task is matched somebody with somebody on designated terminal node
Put record to be still unsatisfactory for requiring, then designated terminal node sets are unavailable terminal node and return to the 3rd step, i.e., made again
Determine task allocative decision, and evade the terminal node of task configuration failure.
7th step：Centroid, which is sent, starts task order
Centroid is sent to designated terminal node starts task order.
8th step：Terminal node starts task program process
Designated terminal node, which is received, starts task order, starts task program process, and reports startup to Centroid and appoints
Business result.
9th step：Centroid confirms computing environment configuration result
Centroid collect designated terminal node startup task result, judge task whether start completion
If task start is completed, then computing environment configuration successful.
If task start is not completed, then return to the 4th step, i.e. Centroid and resend to designated terminal node and appointed
Business bag, re-starts task configuration.If above-mentioned circulation is performed three times, task start fails, then sets designated terminal node
It is set to unavailable terminal node and returns to the 3rd step, that is, reformulates task allocative decision, and evade the end of task start failure
End node.
Beneficial effects of the present invention
The present invention is for low, frequent, poor fault tolerance of manual intervention of computing environment allocative efficiency of distributed computing system etc.
Problem, it is proposed that the computing environment method of automatic configuration of distributed computing system, can carry out terminal section automatically according to configuring condition
The replacing of point, reduces operating personnel, improves System Error-tolerance Property, realizes the rapid build of distributed computing system.This method is applicable
Automatically configured in the computing environment of all kinds of distributed computing systems, and distributed computing system rapid build.Test result table
Bright, the computing environment method of automatic configuration of distributed computing system can greatly improve the service efficiency of system, and tool
There is stronger stability and applicability.
Brief description of the drawings
Fig. 1 is the schematic diagram of the computing environment method of automatic configuration of distributed computing system.
Embodiment
The present invention is described in detail below in conjunction with the accompanying drawings.
Fig. 1 is the schematic diagram of the computing environment method of automatic configuration of distributed computing system.Technical scheme
Thinking is：In distributed computing system, Centroid realizes that the communication and control to terminal node, and the concentration of task are deposited
Storage and management, are formulated and according to task allocative decision by task bag automatic deployment to designated terminal node, to task configuration surroundings
Confirmed.Terminal node communicates with Centroid, reports the hardware and software information of terminal node, receives task bag and task
Configuration requirement, performs task configuration, starts and control task program process.Pass through Centroid and the cooperative cooperating of terminal node
Complete automatically configuring for DCE, rapid build distributed computing system.
The detailed step of the present invention is described as follows：
The first step：Centroid is described and stored to task
Centroid forms mission bit stream and task bag according to the task data of reception, and mission bit stream and task bag are deposited
Store up database.Wherein, mission bit stream includes：The identification information of task, the function declaration of task, the configuration requirement of task, appoint
Business bag is the executable resource collection of task.Generally, task recognition information describe task names, mission number, task version number,
Task update date etc..Task function illustrates the algorithm and model of description task, input and output.The description of task configuration requirement is appointed
Data, file, operating system, store path that business needs etc..Task bag is stored in the executable of the task in same catalogue
The resource collections such as program, dynamic base, journal file.
Second step：Centroid collects each destination node information
Centroid collects all destination node informations, includes identification information, computing capability, the task of each terminal node
Configuration record.All terminal nodes are set as available terminal node.Node identification information describes nodename, node serial number, IP
Address.Computing capability illustrates to describe CPU frequency, memory size, operating system of node etc..Task configures record description terminal section
The mission bit stream being configured on point, including：Task recognition information, task function explanation, task store path, task program road
Footpath, the data configuration of task and file configuration.
3rd step：Centroid formulates task allocative decision
Centroid is matched mission bit stream with all available terminal nodal informations, forms task allocative decision, is appointed
Allocative decision of being engaged in includes the designated terminal node of task and the configuration requirement of task.According to application demand, to be calculated appoint is determined
Business.According to mission bit stream and destination node information, task is assigned to designated terminal node, the configuration data of task is set and matched somebody with somebody
Documentation requirements is put, task allocative decision is formed.
4th step：Centroid sends task bag and task configuration requirement
Centroid sends out the mission bit stream in task configuration requirement and database, task bag according to task allocative decision
Give designated terminal node.
5th step：Designated terminal node performs task configuration
Designated terminal node receives mission bit stream, task bag and task configuration requirement, and the storage of task bag is arrived into specified path,
Task execution environment is configured according to task configuration requirement, and updates the task configuration record of designated terminal node.
6th step：Centroid confirms task configuration result
Whether task configuration record, which meets, on Centroid collection designated terminal nodal information, inspection designated terminal node is wanted
Ask
If meeting task configuration requirement, the 7th step is performed；
If being unsatisfactory for task configuration requirement, return the 4th step, i.e. Centroid resend task bag and task is matched somebody with somebody
Requirement is put, designated terminal node re-executes task configuration.If above-mentioned circulation is performed three times, task is matched somebody with somebody on designated terminal node
Put record to be still unsatisfactory for requiring, then designated terminal node sets are unavailable node and return to the 3rd step, that is, reformulated and appoint
Business allocative decision, and evade the terminal node of task configuration failure.
7th step：Centroid, which is sent, starts task order
Centroid is sent to designated terminal node starts task order.
8th step：Terminal node starts task program process
Designated terminal node, which is received, starts task order, starts task program process, and reports startup to Centroid and appoints
Business result.
9th step：Centroid confirms computing environment configuration result
Centroid collect designated terminal node startup task result, judge task whether start completion
If task start is completed, then computing environment configuration successful.
If task start is not completed, then return to the 4th step, i.e. Centroid and resend to designated terminal node and appointed
Business bag, re-starts task configuration.If above-mentioned circulation is performed three times, task start fails, then sets designated terminal node
It is set to unavailable terminal node and returns to the 3rd step, that is, reformulates task allocative decision, and evade the end of task start failure
End node.
During the present invention is developed applied to the project that our unit undertakes, the calculating ring of distributed aviation Virtual Scene System is realized
Border is automatically configured.Distributed aviation Virtual Scene System is made up of many aviation virtual cockpits of network connection.Each aviation void
Intend cockpit to be made up of multiple stage computers terminal node and equipment, different terminals node performs different computing tasks, control virtual seat
What comes into a driver's projector equipment and human-computer interaction device in cabin.It is experimentally confirmed, the present invention can solve distributed aviation virtual views
The computing environment human configuration efficiency of system is low, operating personnel are professional the problems such as require height, poor fault tolerance, collaborative configuration difficulty.
It is of the invention effectively to realize automatically configuring for distributed aviation Virtual Scene System computing environment, reduce configuration operating personnel, enhancing
The fault-tolerance of computing environment configuration, improves the calculating task allocative efficiency of aviation virtual cockpit, and rapid build distribution aviation is empty
Intend visual system.
Claims (1)
1. a kind of computing environment method of automatic configuration of distributed computing system, exists between Centroid and each terminal node
Information transfer path, it is characterised in that to any one tasks carrying following step：
The first step：Centroid is described and stored to task：
Centroid forms mission bit stream and task bag according to the task data of reception, and mission bit stream and task bag storage are arrived
Database；Wherein, mission bit stream includes：The identification information of task, the function declaration of task, the configuration requirement of task；Task bag
That is the executable resource collection of task；
Second step：Centroid collects each destination node information：
Centroid collects all destination node informations, including identification information, computing capability, the task of each terminal node are configured
Record；All terminal nodes are set as available terminal node；
3rd step：Centroid formulates task allocative decision
Centroid is matched mission bit stream with all available terminal nodal informations, forms task allocative decision, task point
Include the designated terminal node of task and the configuration requirement of task with scheme；
4th step：Centroid sends task bag and task configuration requirement
Centroid sends the mission bit stream in the configuration requirement and database of task, task bag according to task allocative decision
Give designated terminal node；
5th step：Designated terminal node performs task configuration
Designated terminal node receives mission bit stream, task bag and task configuration requirement, according to task configuration requirement to tasks carrying
Environment is configured, and updates the task configuration record of designated terminal node；
6th step：Centroid confirms task configuration result
Centroid collects designated terminal nodal information, checks whether task configuration record meets requirement on designated terminal node：
If meeting task configuration requirement, the 7th step is performed；
If being unsatisfactory for task configuration requirement, the 4th step is returned to, if described return to circulation three times, times of designated terminal node
Business configuration record is still unsatisfactory for requiring, is then unavailable terminal node by designated terminal node sets and returns to the 3rd step；
7th step：Centroid, which is sent, starts task order：
Centroid is sent to designated terminal node starts task order；
8th step：Terminal node starts task program process：
Designated terminal node, which is received, starts task order, starts task program process, and report startup task knot to Centroid
Really；
9th step：Centroid confirms computing environment configuration result
Centroid collect designated terminal node startup task result, judge task whether start completion：
If task start is completed, then computing environment configuration successful；
If task start is not completed, then return to the 4th step, if the circulation for returning to the 4th step is performed three times, task start is equal
Failure, then be unavailable terminal node by designated terminal node sets and return to the 3rd step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710240836.2A CN107066338A (en) | 2017-04-13 | 2017-04-13 | The computing environment method of automatic configuration of distributed computing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710240836.2A CN107066338A (en) | 2017-04-13 | 2017-04-13 | The computing environment method of automatic configuration of distributed computing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107066338A true CN107066338A (en) | 2017-08-18 |
Family
ID=59600767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710240836.2A Pending CN107066338A (en) | 2017-04-13 | 2017-04-13 | The computing environment method of automatic configuration of distributed computing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107066338A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108958937A (en) * | 2018-06-29 | 2018-12-07 | 国家超级计算天津中心 | A kind of templating calculating environment configurations method |
| CN111586043A (en) * | 2020-05-07 | 2020-08-25 | 广东培正学院 | Operation and maintenance management system for computer equipment |
| WO2023231781A1 (en) * | 2022-06-02 | 2023-12-07 | 华为云计算技术有限公司 | Distributed collaborative ai task evaluation method, management apparatus, control apparatus and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073546A (en) * | 2010-12-13 | 2011-05-25 | 北京航空航天大学 | Task-dynamic dispatching method under distributed computation mode in cloud computing environment |
| CN102437938A (en) * | 2012-01-09 | 2012-05-02 | 北京邮电大学 | Large-scale network monitoring oriented virtual deployment system and method |
| CN103220163A (en) * | 2012-01-19 | 2013-07-24 | 阿里巴巴集团控股有限公司 | Updating method and system of configuring information |
| CN103810023A (en) * | 2014-03-06 | 2014-05-21 | 中国科学院信息工程研究所 | Distributed application intelligent deployment method and system in cloud platform |
| CN105743703A (en) * | 2016-03-10 | 2016-07-06 | 郑州悉知信息科技股份有限公司 | Node server configuration method and system in cluster and node server |
-
2017
- 2017-04-13 CN CN201710240836.2A patent/CN107066338A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073546A (en) * | 2010-12-13 | 2011-05-25 | 北京航空航天大学 | Task-dynamic dispatching method under distributed computation mode in cloud computing environment |
| CN102437938A (en) * | 2012-01-09 | 2012-05-02 | 北京邮电大学 | Large-scale network monitoring oriented virtual deployment system and method |
| CN103220163A (en) * | 2012-01-19 | 2013-07-24 | 阿里巴巴集团控股有限公司 | Updating method and system of configuring information |
| CN103810023A (en) * | 2014-03-06 | 2014-05-21 | 中国科学院信息工程研究所 | Distributed application intelligent deployment method and system in cloud platform |
| CN105743703A (en) * | 2016-03-10 | 2016-07-06 | 郑州悉知信息科技股份有限公司 | Node server configuration method and system in cluster and node server |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108958937A (en) * | 2018-06-29 | 2018-12-07 | 国家超级计算天津中心 | A kind of templating calculating environment configurations method |
| CN111586043A (en) * | 2020-05-07 | 2020-08-25 | 广东培正学院 | Operation and maintenance management system for computer equipment |
| WO2023231781A1 (en) * | 2022-06-02 | 2023-12-07 | 华为云计算技术有限公司 | Distributed collaborative ai task evaluation method, management apparatus, control apparatus and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Beeckman | CIM-OSA: computer integrated manufacturing—open system architecture | |
| CN111813570A (en) | Event-driven message interaction method for power Internet of things | |
| CN103002490B (en) | A kind of business simulating test macro and its implementation | |
| CN103095574A (en) | Management method for network system, network system, and management server | |
| CN107066338A (en) | The computing environment method of automatic configuration of distributed computing system | |
| CN112272234A (en) | Platform management system and method for realizing edge cloud collaborative intelligent service | |
| CN113055232B (en) | Network configuration deployment method, device and equipment | |
| CN113993214A (en) | Task-driven spatial information network resource management system and method | |
| Lin et al. | Key technologies and solutions of remote distributed virtual laboratory for E-learning and E-education | |
| CN114666335B (en) | Distributed system load balancing device based on data distribution service DDS | |
| CN114385126B (en) | K8 s-based multi-tenant deep learning model research and development system and method | |
| US6748419B2 (en) | System and method for solid modeling | |
| Boukerche et al. | Performance evaluation of data distribution management strategies | |
| CN112256386A (en) | Simulation operation support system | |
| CN109600251B (en) | Virtual network deployment method, system and device | |
| O'Brien et al. | Agents of change in business process management | |
| CN115981873B (en) | Distributed edge fusion calculation force deployment system | |
| Groenewegen et al. | Delegation modeling with paradigm | |
| Wang et al. | Integration of fogbus2 framework with container orchestration tools in cloud and edge computing environments | |
| CN109635377B (en) | Equipment maintenance resource allocation control method based on discrete event modeling | |
| CN115941504B (en) | Modeling and simulation method for operation and treatment of micro-service system | |
| EP1499065A2 (en) | Network planning method | |
| Schmidt | Adaptive and reflective middleware for distributed real-time and embedded systems | |
| Shah et al. | A real-time simulation framework for complex and large-scale optical transport networks based on the SDN paradigm | |
| CN117938636A (en) | Intelligent node management and control 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170818 |