CN102761570A - System and method for monitoring grid resources based on agents - Google Patents
System and method for monitoring grid resources based on agents Download PDFInfo
- Publication number
- CN102761570A CN102761570A CN2011101083991A CN201110108399A CN102761570A CN 102761570 A CN102761570 A CN 102761570A CN 2011101083991 A CN2011101083991 A CN 2011101083991A CN 201110108399 A CN201110108399 A CN 201110108399A CN 102761570 A CN102761570 A CN 102761570A
- Authority
- CN
- China
- Prior art keywords
- information
- resource
- transducer
- system resource
- management device
- 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
Images
Abstract
The invention provides a system and a method for monitoring grid resources based on agents. The method comprises the steps that first, sensors collect system resource information of hosts at respective node terminals, and pack and send the collected system resource information to sensor managers based on a predetermined protocol; after receiving the system resource information sent by the sensors in areas under respective jurisdiction, the sensor managers of all the domain agent layers carry out meta-information extraction and classification for the system resource information based on meta-data standards so as to save the information, and at the same time, send the state information of respective areas to an information center server; and finally, the information center server collects the information in all areas, and writes the information into a central database, thereby creating a multi-level monitoring environment based on the agents, and realizing the functions of collection, summarizing, arranging, storing and display for state information of the resource nodes in a 'virtual supermarket' environment.
Description
Technical field
The present invention relates to monitoring resource technical field in the grid resource, particularly a kind of gridding resource supervisory control system and method for supervising based on the agency.
Background technology
Grid computing be with distribute on the geography, the various resources of isomery, connect and integrate through high speed internet, form the transparent relatively virtual high-performance computing environment of user.The characteristics of gridding resource are: distributivity, isomerism, dynamic and autonomy.Resource management is the core of grid computing, and purpose is the isomerism and the complexity of shielding underlying resource, the various grid resources that disperse in the grid; Effective management resource; Improve the utilance of resource,, realize load balance for grid work distributes rational gridding resource.
The gridding resource supervisory control system is in charge of the static state and the multidate information of various resources in the grid computing environment (calculating, network, storage, instrument etc.); Each node current states in the grid for example; Like relevant informations such as the operating position of host IP address, operating system, cpu busy percentage, physical storage and virtual memory, network data transmission bandwidth, database, servers; Its target is to disperseing on the various geography in the grid environment; The resource that dynamically adds or leave Virtual Organization is visited pellucidly, and various running states of resources are carried out unified monitoring and management.
External representational monitoring tools has following several kinds at present:
1, MDS (Monitoring and Discovery Service): monitoring and discovery service are discovery, selection and the optimization that the Globus project is used for supporting resource information under the grid computing environment.It provides kit and application programming interfaces to be used for finding, issuing and visit the various resource informations of computing grid.
2, R-GMA (Resource-Grid Monitor Architecture): be gridding information service and the supervisory control system of developing in the Data Grid project.Its maximum characteristic be adopted traditional powerful, relational model is realized flexibly.This system is used to the gridding information service and uses two aspects of monitor service.
3, GRM (Globus Resource Monitor): be a semi-on-line watch-dog, it is collected in the application information that moves in the distributed system of isomery, and sends to the PROVE visualization tool to the information of collecting.Information can be the statistical information of event tracking data or application behavior.The semi-on-line monitoring is meant that the user can ask any available tracking data and watch-dog the enough rational time to collect data term of execution of application program.GRM is used for the monitoring application program performance, and it can handle mass data; GRM does not support sensor management.
Yet there is following shortcoming in above-mentioned each monitoring tools: based on " virtual supermarkets " resource-sharing and the collaboration services platform, calculation task all is to use XML to describe (1).Above-mentioned each monitoring tools can't be resolved the XML document in the platform.(2) on " virtual supermarkets " platform, above-mentioned monitoring tools is difficult to combine with the scheduler module and the information management model of system, and then task management functions and the gridding resource information that can't utilize platform to provide.What (3) " virtual supermarkets " the resource description agreement that adopted and above-mentioned monitoring tools were adopted is incompatible.
Summary of the invention
The object of the present invention is to provide a kind of gridding resource supervisory control system and method, to be implemented in the function of in " virtual supermarkets " environment each resource node being carried out status information capture, gather, puts in order, stores and showing based on the agency.
Reach other purposes in order to achieve the above object, the gridding resource supervisory control system based on the agency provided by the invention comprises: be arranged on the transducer of each node side, be respectively applied for the system resource information of the main frame of node side separately of gathering; Be in the sensor management device of territory Agent layer; Be used for the system resource information that transducer sent of each node side in its region within the jurisdiction being carried out metamessage extraction and classification, so that the main frame of each node side in the region within the jurisdiction is monitored and managed based on metadata standard; And the information centre's server that is in the global information layer, the metamessage that being used for the sensor management device with each territory Agent layer is provided is given record, with monitoring and manage each territory Agent layer.
A kind of gridding resource method for supervising provided by the invention based on the agency; Comprise step: 1) the system resource information of the main frame of each sensor acquisition node side separately, and the sensor management device under being sent to after based on predetermined agreement with the system resource information-package of being gathered; 2) after the sensor management device of each territory Agent layer receives the system resource information that transducer was sent to of region within the jurisdiction separately; Based on metadata standard system resource information is carried out that metamessage extracts and classification is preserved, simultaneously the state information in information centre's server sends territory separately; And 3) information centre's server is collected the information in each territory, and is written in the central database.
In sum; Gridding resource supervisory control system and method construct based on the agency of the present invention multi-level monitoring environment based on the agency, for " virtual supermarkets " resource-sharing provides one to overlap the instrument that the operating state of member node in " supermarket " each territory is monitored with the collaboration services platform administrative staff.
Description of drawings
Fig. 1 is the gridding resource supervisory control system sketch map based on the agency of the present invention.
Fig. 2 is the working sensor flow chart based on the gridding resource supervisory control system of acting on behalf of of the present invention.
Fig. 3 is the gridding resource method for supervising flow chart based on the agency of the present invention.
Embodiment
See also Fig. 1; Gridding resource supervisory control system based on the agency of the present invention is divided into 3 levels from structure; They are respectively global information layer, territory Agent layer; With the nodal information acquisition layer, it comprises: be arranged on the transducer of each node side, the sensor management device that is in the territory Agent layer and information centre's server.
In the present embodiment; Said gridding resource supervisory control system comprises the transducer 11 that is arranged on node side 11, the transducer 12 that is arranged on node side 12, the transducer 13 that is arranged on node side 13 and the transducer 21 that is arranged on node side 21, the transducer 22 that is arranged on node side 22, the transducer 23 that is arranged on node side 23, and each transducer is respectively applied for the system resource information of the main frame of node side separately of gathering.For example, transducer 11 is used for the system resource information of the main frame 11 of acquisition node end 11; Transducer 12 is used for the system resource information of the main frame 12 of acquisition node end 12; Transducer 13 is used for the system resource information of the main frame 13 of acquisition node end 13; Transducer 21 is used for the system resource information of the main frame 21 of acquisition node end 21; Transducer 22 is used for the system resource information of the main frame 22 of acquisition node end 22; Transducer 23 is used for the system resource information of the main frame 23 of acquisition node end 23.
Each transducer can obtain system data through the system parameters that regularly reads the main frame of node side separately, and the data owner of required monitoring will comprise: state of central processing unit operating position, Installed System Memory capacity and operating position, systems exchange district size and operating position, disk operating position and application program operation or the like.
As a kind of optimal way, each transducer can adopt the sigar kit of the community that increases income to come the acquisition system resource information, utilizes that this kit can make things convenient for, the various state informations of acquisition node end efficiently.Utilize its built-in a plurality of primary code library simultaneously, can be implemented in acquisition state information on the different operating systems, shielded the isomerism of system.
In addition, each transducer can comprise: acquisition module, protocol analysis are packed module, are reached sending module.Wherein, acquisition module is used to gather the system resource information of the main frame of respective nodes end; Protocol analysis packing module is used for the system resource packets of information of being gathered is dressed up the communication object that meets communications protocol; Sending module is used for the communication object that packing forms is periodically sent to corresponding sensor management device.
The acquisition module of each transducer, protocol analysis packing module, and the workflow of sending module as shown in Figure 2; Promptly carry out the image data thread by acquisition module earlier, and the data of gathering are sent into data buffer area, then; Again by protocol analysis packing module execution performance index encapsulation thread; Be about to packets of information and dress up the communication object that meets communications protocol, carry out the issue thread by sending module more at last, the communication object that is about to form periodically sends to corresponding sensor management device.
Said each sensor management device is in the territory Agent layer; Be used for the system resource information that transducer sent of each node side in its region within the jurisdiction being carried out metamessage extraction and classification, so that the main frame of each node side in the region within the jurisdiction is monitored and managed based on metadata standard.As shown in Figure 1; In the present embodiment; The sensor management device 1 that is in the territory Agent layer is monitored and is managed the transducer in its region within the jurisdiction 11, transducer 12 and transducer 13, and the sensor management device 2 that is in the territory Agent layer is monitored and managed the transducer in its region within the jurisdiction 21, transducer 22 and transducer 23.
Each sensor management management the state information of all node sides in the territory separately; Can adopt the Socket communication mode between each transducer and the corresponding sensor management device; Every at a distance from a collection period; Each sensor management device can receive the current state information of the main frame of the node side of gathering from each transducer in the territory separately, and thus, each sensor management device can show member's in the territory separately state information; Can upgrade the state information of respective nodes end in the local data base simultaneously, and pass to the information centre's server that is positioned at the global information layer after putting out current latest data set in order.
Need to prove that the local data base of each sensor management device can use the built-in JAVA DB database of JDK, it is a small-sized SQL database Derby who increases income, and advantage is that volume is little, can be embedded in the program.In all have installed the system of the above version of JDK6.0, can dispose easily and move, and have good performance.
Said information centre server is in the global information layer, and the metamessage that being used for the sensor management device with each territory Agent layer is provided is given record, monitors and manage each territory Agent layer simultaneously.For example; The metamessage that said information centre server provides sensor management device 1 and sensor management device 2 is put into central database and is preserved; Also write down the essential information of sensor management device 1 and sensor management device 2 each self-contained node side simultaneously; Like resource name, resource type, resource address etc., grasp global information thus.
Need to prove that the central database of said information centre server can use MySQL to handle the data and the quicker response of bigger rule film.
See also Fig. 3, the course of work of above-mentioned gridding resource supervisory control system is following:
At first, the system resource information of the main frame of each sensor acquisition node side separately, and the system resource information of being gathered is encapsulated, the protocol object that the encapsulation of redispatching subsequently the forms sensor management device under separately based on predetermined agreement.For example, transducer 11,12,13,21,22,23 gather separately main frame 11,12,13,21,22,23 state information, and tunneling object is redispatched subsequently.
Then; After the sensor management device of each territory Agent layer receives the system resource information that transducer was sent to of region within the jurisdiction separately; Based on metadata standard system resource information is carried out that metamessage extracts and classification is saved to local data base, simultaneously the state information in information centre's server sends territory separately.
At last, information centre's server is collected the information in each territory, and is written in the central database.For example, " virtual supermarkets " platform resource administration module is collected the information in each territory, and it is write in the central database.
Need to prove; Because " virtual supermarkets " has cross-domain cross-platform characteristics, therefore, the JAVA that selects to use SUN company based on agency's gridding resource supervisory control system of the present invention is as programming language; Because the professional platform independence that JAVA is good; No matter problem that can well the resolution system isomerism is Win32 or Linux, can move this supervisory control system.
In addition, the message transmission aspect, the mode that socket that has adopted and multicast UDP combine is utilized multicast can in local area network (LAN), transmit information easily, and needn't be removed to be provided with corresponding address and port as socket.But consider that simultaneously the territory of being made up of LAN need use the mode of socket and come to intercom mutually, need to realize corresponding socket communication function.So native system has very high flexibility, both can in same local area network (LAN), use, and also can cross-domainly monitor.
Communication protocol is used the method for objectification, and the performance index that are about to be collected all are encapsulated in the protocol object, and makes the territory agency can obtain the performance index in the protocol object through corresponding getter function is set.So, performance index data and data parsing have just all been put in the protocol object, and the generation of protocol data and the parsing of data all are responsible for by protocol object.The object sequence method among the JAVA of having used is simultaneously transmitted the protocol object of generation.
In sum; Gridding resource supervisory control system and method based on the agency of the present invention is the system and method that " virtual supermarkets " resource-sharing and collaboration services platform administrative staff provide a cover that the operating state of member node in " supermarket " each territory is monitored; Also constructed a multi-level monitoring environment simultaneously based on the agency; In " virtual supermarkets " environment, realized function that each resource node is carried out status information capture, gathers, puts in order, stores and shows; Owing to adopt this locality storage of state information; The state information of gridding resource is stored in the local database as far as possible, and can reduces the required performance cost of user capture monitoring information through multi-zone supervision to the monitoring resource data; The paralysis of the whole grid system of having avoided simultaneously causing because of individual domain failure.
With respect to prior art, advantage of the present invention comprises:
1) can run on based on " virtual supermarkets " resource-sharing and the collaboration services platform, realize based on scheduler module and information management module.
2) two kinds of different deployed environment of communication mode reply of socket and multicast have been adopted.
3) has professional platform independence, deployment, convenient configuration.
The foregoing description is just listed expressivity principle of the present invention and effect is described, but not is used to limit the present invention.Any personnel that are familiar with this technology all can make amendment to the foregoing description under spirit of the present invention and scope.Therefore, rights protection scope of the present invention should be listed like claims.
Claims (7)
- One kind based on agency the gridding resource supervisory control system, it is characterized in that comprising:Be arranged on the transducer of each node side, be respectively applied for the system resource information of the main frame of node side separately of gathering;Be in the sensor management device of territory Agent layer; Be used for the system resource information that transducer sent of each node side in its region within the jurisdiction being carried out metamessage extraction and classification, so that the main frame of each node side in the region within the jurisdiction is monitored and managed based on metadata standard;Be in information centre's server of global information layer, the metamessage that being used for the sensor management device with each territory Agent layer is provided is given record, with monitoring and manage each territory Agent layer.
- 2. the gridding resource supervisory control system based on the agency as claimed in claim 1 is characterized in that: each transducer adopts the sigar kit of the community that increases income to come the acquisition system resource information.
- 3. the gridding resource supervisory control system based on the agency as claimed in claim 1 is characterized in that: adopt the Socket communication mode between each transducer and the corresponding sensor management device.
- 4. like each described gridding resource supervisory control system based on the agency of claim 1 to 3, it is characterized in that: transducer comprises: acquisition module is used to gather the system resource information of the main frame of respective nodes end; Protocol analysis packing module is used for the system resource packets of information of being gathered is dressed up the communication object that meets communications protocol; And sending module, be used for the communication object that packing forms is periodically sent to corresponding sensor management device.
- 5. the gridding resource supervisory control system based on the agency as claimed in claim 1 is characterized in that: the local data base that the sensor management device adopts is the built-in JAVA DB database of JDK.
- 6. the gridding resource supervisory control system based on the agency as claimed in claim 1 is characterized in that: the central database that information centre's server adopts is a MySQL type database.
- One kind based on agency the gridding resource method for supervising, it is characterized in that comprising step:1) the system resource information of the main frame of each sensor acquisition node side separately, and the sensor management device under being sent to after based on predetermined agreement with the system resource information-package of being gathered;2) after the sensor management device of each territory Agent layer receives the system resource information that transducer was sent to of region within the jurisdiction separately; Based on metadata standard system resource information is carried out that metamessage extracts and classification is preserved, simultaneously the state information in information centre's server sends territory separately;3) information centre's server is collected the information in each territory, and is written in the central database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101083991A CN102761570A (en) | 2011-04-28 | 2011-04-28 | System and method for monitoring grid resources based on agents |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101083991A CN102761570A (en) | 2011-04-28 | 2011-04-28 | System and method for monitoring grid resources based on agents |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102761570A true CN102761570A (en) | 2012-10-31 |
Family
ID=47055891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011101083991A Pending CN102761570A (en) | 2011-04-28 | 2011-04-28 | System and method for monitoring grid resources based on agents |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102761570A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103701661A (en) * | 2013-12-23 | 2014-04-02 | 浪潮(北京)电子信息产业有限公司 | Method and system for realizing node monitoring |
| CN107506438A (en) * | 2017-08-23 | 2017-12-22 | 福建星瑞格软件有限公司 | A kind of data processing storage method and device for Internet of Things |
| CN111694854A (en) * | 2020-06-05 | 2020-09-22 | 北京百度网讯科技有限公司 | Method and device for monitoring entrance and exit of area, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1534949A (en) * | 2003-03-27 | 2004-10-06 | 国际商业机器公司 | Method and equipment used for obtaining state information in network |
| CN1547356A (en) * | 2003-12-10 | 2004-11-17 | 中国科学院计算技术研究所 | A grid-faced monitor system and method |
| CN101316280A (en) * | 2008-06-13 | 2008-12-03 | 南京邮电大学 | Gridding resource intelligent monitoring method based on feedback |
-
2011
- 2011-04-28 CN CN2011101083991A patent/CN102761570A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1534949A (en) * | 2003-03-27 | 2004-10-06 | 国际商业机器公司 | Method and equipment used for obtaining state information in network |
| CN1547356A (en) * | 2003-12-10 | 2004-11-17 | 中国科学院计算技术研究所 | A grid-faced monitor system and method |
| CN101316280A (en) * | 2008-06-13 | 2008-12-03 | 南京邮电大学 | Gridding resource intelligent monitoring method based on feedback |
Non-Patent Citations (1)
| Title |
|---|
| 王静宇等: "基于Agent的网格资源监控系统的设计与实现", 《微电子学与计算机》, vol. 23, 30 December 2006 (2006-12-30) * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103701661A (en) * | 2013-12-23 | 2014-04-02 | 浪潮(北京)电子信息产业有限公司 | Method and system for realizing node monitoring |
| CN103701661B (en) * | 2013-12-23 | 2017-08-25 | 浪潮(北京)电子信息产业有限公司 | A kind of method and system for realizing monitoring nodes |
| CN107506438A (en) * | 2017-08-23 | 2017-12-22 | 福建星瑞格软件有限公司 | A kind of data processing storage method and device for Internet of Things |
| CN111694854A (en) * | 2020-06-05 | 2020-09-22 | 北京百度网讯科技有限公司 | Method and device for monitoring entrance and exit of area, electronic equipment and storage medium |
| CN111694854B (en) * | 2020-06-05 | 2023-07-21 | 北京百度网讯科技有限公司 | Regional access monitoring method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Javed et al. | IoTEF: A federated edge-cloud architecture for fault-tolerant IoT applications | |
| Malik et al. | Cloud computing and internet of things integration: Architecture, applications, issues, and challenges | |
| CN103338135B (en) | A kind of method for real-time monitoring of cluster storage capacity | |
| Trihinas et al. | Monitoring elastically adaptive multi-cloud services | |
| CN110099099A (en) | For the spatial information micro services encapsulation at terrestrial information port and service integration method | |
| CN103390018B (en) | A kind of Web service data modeling and searching method based on SDD | |
| Brinkmann et al. | Scalable monitoring system for clouds | |
| CN101256599B (en) | System for gathering data of distributing simulation platform based on grid | |
| CN114143203A (en) | Kubernetes container network data packet index acquisition method and system based on dynamic service topological mapping | |
| CN107025222A (en) | A kind of distributed information log acquisition method and device | |
| CN107404417A (en) | A kind of processing method of monitoring data, processing unit and processing system | |
| Albrecht et al. | Making work queue cluster-friendly for data intensive scientific applications | |
| Weng et al. | Kmon: An in-kernel transparent monitoring system for microservice systems with ebpf | |
| CN102761570A (en) | System and method for monitoring grid resources based on agents | |
| Großmann et al. | SensIoT: an extensible and general Internet of Things monitoring framework | |
| CN108234622A (en) | Charging method and charge system | |
| Hetzel et al. | μactor: Stateful serverless at the edge | |
| Casola et al. | A reference architecture for sensor networks integration and management | |
| CN103067507A (en) | Internet cloud computing resource virtualization packaging system and method | |
| Sindhanaiselvan et al. | A survey on sensor cloud: architecture and applications | |
| WO2022193855A1 (en) | Task state updating method and apparatus, device, and medium | |
| CN115567526A (en) | Data monitoring method, device, equipment and medium | |
| CN113824801B (en) | Intelligent integration terminal unified access management component system | |
| Calderon et al. | Monitoring Framework for the Performance Evaluation of an IoT Platform with Elasticsearch and Apache Kafka | |
| CN114969199A (en) | Method, device and system for processing remote sensing data and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20121031 |