CN202565304U - Distributed computing task scheduling and execution system - Google Patents
Distributed computing task scheduling and execution system Download PDFInfo
- Publication number
- CN202565304U CN202565304U CN2012202282615U CN201220228261U CN202565304U CN 202565304 U CN202565304 U CN 202565304U CN 2012202282615 U CN2012202282615 U CN 2012202282615U CN 201220228261 U CN201220228261 U CN 201220228261U CN 202565304 U CN202565304 U CN 202565304U
- Authority
- CN
- China
- Prior art keywords
- cloud computing
- platform
- task scheduling
- server
- center
- 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.)
- Expired - Lifetime
Links
Images
Landscapes
- Computer And Data Communications (AREA)
Abstract
The utility model discloses a distributed computing task scheduling and execution system. The system comprises a cloud computing service application center, a cloud computing auxiliary supervision center, a cloud computing platform and a user side, wherein the cloud computing platform is composed of a firewall, a load equalizer, a platform management work station, an anti-virus server, a power equipment and environment server, a cloud computing service pool and a cloud computing resource pool which are connected with one another through network devices, the user side is connected with the cloud computing service application center, and the cloud computing service application center and the cloud computing auxiliary supervision center are connected with the firewall through special lines respectively. According to the system, all devices are hot spares for one another, loads are balanced, if the user side fails, a user can execute task scheduling and service requests by crossing sections, the system reliability is improved, all dynamic stability control (DSC) systems and computer loads which are born by DSC systems are changed dynamically, a three-level fault tolerance technology is used for storing a zero tolerance state, key information is prevented from being lost due to system failure, and heartbeat detection is not affected by system loads.
Description
Technical field
The utility model relates to a kind of distributed computing task scheduling and executive system.
Background technology
Along with science and technology development; Be painted into the research of the particle structure of microcosmic from the star chart of macroscopic view; From the have a nuclear test deciphering of post genome project of biological field of the simulation of military field; Arrive large scale network measurement, emulation and simulation or the like again; These significant challenge property computational problems all need be striden a large amount of distributed computing task scheduling of management domain utilization and carried out resource, and the demand that Along with people's is increasingly extensive to information data makes traditional distributed computing task scheduling and execution can't satisfy the demand that expands day by day.Traditional distributed computing task scheduling and executive system can't couple together many computers different location or that have difference in functionality or that have different pieces of information with communication network; Can't be under the unified management control of control system; Accomplish information handling task in phase; Can't dynamically be combined into the system towards different service objects at short notice, system can't provide respective service for the user according to user's demand.In traditional distributed task dispatching and executive system several kinds of states to task; Regulation " distribution just " and " accomplishing " two states are the zero tolerance state mostly; In the continuous state of DSC is reported; If because the transient fault of network (this is to allow to occur in the complicated distributed environment) causes the zero tolerance loss of state or is capped; Will cause chaotic or same the task of sub-distribution again of the distribution of task, if the object information of task is lost, the long-time achievement of accomplishing (such as the transcoding achievement) will become rubbish.
The utility model content
The purpose of the utility model is to solve the deficiency of prior art, and a kind of distributed computing task scheduling and executive system are provided, and each equipment room heat each other is equipped with; Load balancing, if user side breaks down, this user can stride column execute the task scheduling and services request; Promoted system reliability; And the load size of each DSC and carrying computer thereof is dynamic change, and three grades of fault-toleranr techniques are adopted in the preservation of zero tolerance state, guarantees can not lose key message because of the inefficacy of system; Heartbeat is surveyed and not influenced by system loading, thereby has avoided the mistake that heartbeat timeout caused under higher load condition to switch.
The purpose of the utility model realizes through following technical scheme: distributed computing task scheduling and executive system; It comprises cloud computing service application center, cloud computing auxiliary supervision center, cloud computing platform and one or more user side; Cloud computing platform is made up of fire compartment wall, one or more load equalizer, one or more platform management work station, one or more antivirus server, one or more rotating ring server, cloud computing service pool and cloud computing resource pool; User side is connected with cloud computing service application center; Cloud computing service application center, the auxiliary supervision center of cloud computing are connected with the fire compartment wall of cloud computing platform through special line respectively, and fire compartment wall, load equalizer, platform management work station, antivirus server, rotating ring server, cloud computing service pool and cloud computing resource pool are through network equipment interconnection.
The auxiliary supervision center of described cloud computing is made up of one or more rotating ring work stations and one or more network management workstation;
Described cloud computing service pool is by one or more WEB and application server cluster, one or more calculated datas exchange area cluster and one or more active and standby resource and task scheduling server through network equipment interconnection, and one or more active and standby cloud platform metadatabase and one or more active and standby cloud platform flow media server composition; Described cloud computing resource pool is by forming through one or more calculating virtual pools and one or more storing virtuals pond of network equipment interconnection.
The beneficial effect of the utility model is:
(1) the utility model provides a kind of distributed computing task scheduling and executive system; Employing based on the large platform Clustering of Big-Cluster as the aggregated structure between a plurality of distributed computing tasks scheduling and executive system (media services center); Soft or the hardware of the processing identical services in scheduling of single distributed computing task and executive system carries out the redundant heat of multimachine is equipped with; Can hotly each other be equipped with between distributed computing task scheduling and the executive system, load balancing is if intraplate distributed computing task scheduling of user and executive system break down; This user can stride column and carry out distributed task dispatching and service request processing; Its result is transparent fully concerning the user, shifts automatically, has promoted TV station's distributed computing task scheduling and executive system whole reliability;
(2) the utility model provides a kind of distributed computing task scheduling and executive system; Employing is based on Multi-Agent distributed task dispatching technology; Make that each DSC in distributed computing task scheduling and the executive system and the load size of carrying computer thereof are dynamic changes, the load state current according to this system takes effective dynamic task scheduling strategy to come the load of each node of balance (machine); Be sent to the task on the current heavily loaded computer on the underloading computer and carry out; Perhaps distribute to new task on the underloading computer and carry out, thereby make task executed in parallel as much as possible, improve the resource utilization and the efficient of whole system;
(3) the utility model provides a kind of distributed computing task scheduling and executive system, adopts the distributed parallel task treatment technology based on Intelli-DSC, makes DSC can carry out advanced technologies such as self-management, intelligence restoration, active clone.Before task is carried out; DSC is in resting state, and the moderate task section of particle can be resolved into to task according to the task handling demand in the control centre; The corresponding DSC process instance of each section; System is according to what of task section, wakes up and DSC (can be the DSC process instance in this column, also can be the DSC process instance of striding column) the instance quantity of (intelligence) clone's respective amount automatically; Guarantee that the processing quantity of DSC and current workload are complementary, and equally loaded;
(4) the utility model provides a kind of distributed computing task scheduling and executive system; In whole distributed computing task scheduling and executive system; In order to improve the fault-tolerant ability of whole system; Three grades of fault-toleranr techniques are adopted in preservation to zero tolerance state (it is the zero tolerance state that assignment of mission distributes success status, task completion status), guarantee that each link can not lose key state information because of the inefficacy of system's (such as network).These three grades fault-tolerant comprises that DSC is local, TSC is local, DPU is local.Can allow suitable losing for intermediateness information, such consideration can improve the performance of system and the degree of balance between the availability;
(5) the utility model provides a kind of distributed computing task scheduling and executive system, adopts the extremely fast fingerprint extraction technology based on microchannel memory-mapped and Smart-Slice, when media file moves, transmits in distributed computing task scheduling and executive system; In order to guarantee that file arrives efficient, the correct and reliability of purpose node; The extremely fast fingerprint extraction technology of creationary proposition under the prerequisite of correctness that guarantees file and reliability, has guaranteed the continuity of work simultaneously and has improved operating efficiency; Through test; Extract the fingerprint of a big small documents of 8GB, the time is 20 seconds, has shortened the several times time than conventional method.The microchannel memory-mapped is meant the skip operations system; The direct control disk file; Size utilization Smart-Slice (intelligence is cut apart) technology according to file becomes several big or small impartial data blocks of dividing to the file dynamic partition simultaneously, extracts the fingerprint of each data block at last with multi-threaded parallel;
(6) the utility model provides a kind of distributed computing task scheduling and executive system; Adopt multiple heartbeat inspecting technology; Support multiple heartbeat modes such as network interface card (private network/public network), com port, disk heartbeat, multiple guarantee heartbeat detection approach has further improved the reliability of system.The kernel spacing that utilizes that distributed computing task scheduling and executive system are unique carries out the technology that heartbeat is surveyed, and can guarantee that the heartbeat detection is not influenced by system loading, thereby avoid the mistake that heartbeat timeout caused under higher load condition to switch;
(7) the utility model provides a kind of distributed computing task scheduling and executive system; Cluster isomery gateway and white list authentication; Cluster UNIX gateway is under the prerequisite that guarantees reliability; Avoid not receiving between distributed computing task scheduling and each operational plate of executive system the attack of WINDOWS internet worm and wooden horse, guarantee to have only the complete reliable target approach plate of data ability in the white list simultaneously.
Description of drawings
Fig. 1 is the structural representation block diagram of the utility model.
Embodiment
Below in conjunction with accompanying drawing the technical scheme of the utility model is described in further detail, but the protection range of the utility model is not limited to the following stated.
As shown in Figure 1; Distributed computing task scheduling and executive system; It comprises cloud computing service application center, cloud computing auxiliary supervision center, cloud computing platform and one or more user side; Cloud computing platform is made up of fire compartment wall, one or more load equalizer, one or more platform management work station, one or more antivirus server, one or more rotating ring server, cloud computing service pool and cloud computing resource pool; User side is connected with cloud computing service application center; Cloud computing service application center, the auxiliary supervision center of cloud computing are connected with the fire compartment wall of cloud computing platform through special line respectively, and fire compartment wall, load equalizer, platform management work station, antivirus server, rotating ring server, cloud computing service pool and cloud computing resource pool are through network equipment interconnection.
The auxiliary supervision center of described cloud computing is made up of one or more rotating ring work stations and one or more network management workstation; Described cloud computing service pool is by one or more WEB and application server cluster, one or more calculated datas exchange area cluster and one or more active and standby resource and task scheduling server through network equipment interconnection, and one or more active and standby cloud platform metadatabase and one or more active and standby cloud platform flow media server composition; Described cloud computing resource pool is by forming through one or more calculating virtual pools and one or more storing virtuals pond of network equipment interconnection.
The utility model utilizes a plurality of functional parts or a plurality of processor to work simultaneously; Improve systematic function or reliability; Many the computers different location or that have difference in functionality or that have different pieces of information are coupled together with communication network, under the unified management control of control system, accomplish information handling task in phase; The utility model adopts many continuous computers to bear the different piece of same task separately; Under people's control, same task accomplished in operation jointly simultaneously.By a plurality of autonomous, interconnective information processing systems, the same task handling mode of common completion under a high level operating system is coordinated.The utility model can dynamically be combined into the system towards different service objects at short notice.Concerning the user, system is transparent, and the user only needs the appointing system What for and needn't point out that which parts just can provide this service.Each part of system is autonomous, but is not the state of anarchy, carries out co-ordination but follow certain Major program by high level operating system.In a computer network, have multiple host to differ establishing a capital is distributed treatment.If such system does not possess the ability that dynamic combined and task are assigned again, they remain centralized processing so.High level operating system is the key of distributed treatment.In distributed system, no longer use complete information, each part provides the state information of oneself, and high level operating system carries out task coordinate and redistribute resources according to these state informations, does not have hierarchical relationship between each part but autonomous.
All process instances of the utility model all are the dynamic self-adapting operations; The utility model is rejecting abnormalities process instance (such as critical processes such as TSC, DSC) automatically; At running; Can be according to corresponding task classification and the quantity of the concrete business demand dynamic-configuration DSC of each plate, the utility model incorporates automatically and monitors newly-increased DSC process instance.Accomplish the parameter configuration and the supervision of distributed cluster system in real time through network management workstation.
Claims (3)
1. distributed computing task is dispatched and executive system; It is characterized in that: it comprises cloud computing service application center, cloud computing auxiliary supervision center, cloud computing platform and one or more user side; Cloud computing platform is made up of fire compartment wall, one or more load equalizer, one or more platform management work station, one or more antivirus server, one or more rotating ring server, cloud computing service pool and cloud computing resource pool; User side is connected with cloud computing service application center; Cloud computing service application center, the auxiliary supervision center of cloud computing are connected with the fire compartment wall of cloud computing platform through special line respectively, and fire compartment wall, load equalizer, platform management work station, antivirus server, rotating ring server, cloud computing service pool and cloud computing resource pool are through network equipment interconnection.
2. distributed computing task scheduling according to claim 1 and executive system is characterized in that: the auxiliary supervision center of described cloud computing is made up of one or more rotating ring work stations and one or more network management workstation.
3. distributed computing task scheduling according to claim 1 and executive system; It is characterized in that: described cloud computing service pool is by one or more WEB and application server cluster, one or more calculated datas exchange area cluster and one or more active and standby resource and task scheduling server through network equipment interconnection, and one or more active and standby cloud platform metadatabase and one or more active and standby cloud platform flow media server composition; Described cloud computing resource pool is by forming through one or more calculating virtual pools and one or more storing virtuals pond of network equipment interconnection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012202282615U CN202565304U (en) | 2012-05-21 | 2012-05-21 | Distributed computing task scheduling and execution system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012202282615U CN202565304U (en) | 2012-05-21 | 2012-05-21 | Distributed computing task scheduling and execution system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN202565304U true CN202565304U (en) | 2012-11-28 |
Family
ID=47214906
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012202282615U Expired - Lifetime CN202565304U (en) | 2012-05-21 | 2012-05-21 | Distributed computing task scheduling and execution system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN202565304U (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103441904A (en) * | 2013-09-03 | 2013-12-11 | 北京邮电大学 | Communication scheduling simulation method based on time limit |
| CN103475672A (en) * | 2013-09-30 | 2013-12-25 | 南京大学 | Method for setting firewall with lowest cost of cloud computing platform |
| US20140047086A1 (en) * | 2012-08-10 | 2014-02-13 | Adobe Systems, Incorporated | Systems and Methods for Providing Hot Spare Nodes |
| CN104182283A (en) * | 2014-08-22 | 2014-12-03 | 北京京东尚科信息技术有限公司 | Task synchronization method |
| CN104462581A (en) * | 2014-12-30 | 2015-03-25 | 成都因纳伟盛科技股份有限公司 | Micro-channel memory mapping and Smart-Slice based ultrafast file fingerprint extraction system and method |
| CN104484228A (en) * | 2014-12-30 | 2015-04-01 | 成都因纳伟盛科技股份有限公司 | Distributed parallel task processing system based on Intelli-DSC (Intelligence-Data Service Center) |
| CN104486447A (en) * | 2014-12-30 | 2015-04-01 | 成都因纳伟盛科技股份有限公司 | Large platform cluster system based on Big-Cluster |
| US9292352B2 (en) | 2012-08-10 | 2016-03-22 | Adobe Systems Incorporated | Systems and methods for cloud management |
| CN106293893A (en) * | 2015-06-26 | 2017-01-04 | 阿里巴巴集团控股有限公司 | job scheduling method, device and distributed system |
| CN107391256A (en) * | 2017-07-03 | 2017-11-24 | 南京南瑞继保电气有限公司 | A kind of relay protection fixed value setting computing architecture and method based on cloud computing technology |
| CN111866096A (en) * | 2020-07-02 | 2020-10-30 | 广州市挖米科技有限责任公司 | Load balancing method and device for medical system |
-
2012
- 2012-05-21 CN CN2012202282615U patent/CN202565304U/en not_active Expired - Lifetime
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140047086A1 (en) * | 2012-08-10 | 2014-02-13 | Adobe Systems, Incorporated | Systems and Methods for Providing Hot Spare Nodes |
| US10963420B2 (en) * | 2012-08-10 | 2021-03-30 | Adobe Inc. | Systems and methods for providing hot spare nodes |
| US9292352B2 (en) | 2012-08-10 | 2016-03-22 | Adobe Systems Incorporated | Systems and methods for cloud management |
| CN103441904B (en) * | 2013-09-03 | 2017-04-12 | 北京邮电大学 | Communication scheduling simulation method based on time limit |
| CN103441904A (en) * | 2013-09-03 | 2013-12-11 | 北京邮电大学 | Communication scheduling simulation method based on time limit |
| CN103475672A (en) * | 2013-09-30 | 2013-12-25 | 南京大学 | Method for setting firewall with lowest cost of cloud computing platform |
| CN103475672B (en) * | 2013-09-30 | 2016-08-17 | 南京大学 | The fire wall setting method of cost minimization in a kind of cloud computing platform |
| CN104182283A (en) * | 2014-08-22 | 2014-12-03 | 北京京东尚科信息技术有限公司 | Task synchronization method |
| CN104182283B (en) * | 2014-08-22 | 2018-07-10 | 北京京东尚科信息技术有限公司 | A kind of task synchronization method |
| CN104462581A (en) * | 2014-12-30 | 2015-03-25 | 成都因纳伟盛科技股份有限公司 | Micro-channel memory mapping and Smart-Slice based ultrafast file fingerprint extraction system and method |
| CN104484228B (en) * | 2014-12-30 | 2017-12-29 | 成都因纳伟盛科技股份有限公司 | Distributed parallel task processing system based on Intelli DSC |
| CN104462581B (en) * | 2014-12-30 | 2018-03-06 | 成都因纳伟盛科技股份有限公司 | Very fast file fingerprint extraction system and method based on the mapping of microchannel internal memory and Smart Slice |
| CN104486447A (en) * | 2014-12-30 | 2015-04-01 | 成都因纳伟盛科技股份有限公司 | Large platform cluster system based on Big-Cluster |
| CN104484228A (en) * | 2014-12-30 | 2015-04-01 | 成都因纳伟盛科技股份有限公司 | Distributed parallel task processing system based on Intelli-DSC (Intelligence-Data Service Center) |
| CN106293893A (en) * | 2015-06-26 | 2017-01-04 | 阿里巴巴集团控股有限公司 | job scheduling method, device and distributed system |
| US10521268B2 (en) | 2015-06-26 | 2019-12-31 | Alibaba Group Holding Limited | Job scheduling method, device, and distributed system |
| CN107391256A (en) * | 2017-07-03 | 2017-11-24 | 南京南瑞继保电气有限公司 | A kind of relay protection fixed value setting computing architecture and method based on cloud computing technology |
| CN111866096A (en) * | 2020-07-02 | 2020-10-30 | 广州市挖米科技有限责任公司 | Load balancing method and device for medical system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN202565304U (en) | Distributed computing task scheduling and execution system | |
| CN103514043B (en) | The data processing method of multicomputer system and the system | |
| CN103164287B (en) | The distributed paralleling calculation platform system that sing on web dynamically participates in | |
| CN102254016B (en) | Cloud-computing-environment-oriented fault-tolerant parallel Skyline inquiry method | |
| CN104618693A (en) | Cloud computing based online processing task management method and system for monitoring video | |
| CN104463492A (en) | Operation management method of electric power system cloud simulation platform | |
| CN102724313B (en) | Clustering bridge operation safety monitoring system based on cloud computation | |
| CN105183554A (en) | Hybrid computing system of high-performance computing and cloud computing, and resource management method therefor | |
| CN103873321A (en) | Distributed file system-based simulation distributed parallel computing platform and method | |
| CN102271145A (en) | Virtual computer cluster and enforcement method thereof | |
| CN112579288A (en) | Cloud computing-based intelligent security data management system | |
| CN104484228B (en) | Distributed parallel task processing system based on Intelli DSC | |
| Lin et al. | Novel algorithms and equivalence optimisation for resource allocation in cloud computing | |
| CN102932399A (en) | Cloud disaster recovery system for power grid dispatching | |
| CN106020969A (en) | High-performance cloud computing hybrid computing system and method | |
| CN106412094A (en) | A method for organizing and managing scattered resources in a public cloud mode | |
| CN104468710A (en) | Mixed big data processing system and method | |
| Huang | On study of building smart campus under conditions of cloud computing and internet of things | |
| CN104298537A (en) | Electric power information communication virtual environment resource model | |
| CN104486447A (en) | Large platform cluster system based on Big-Cluster | |
| CN106250210A (en) | Dispatching method of virtual machine under cloud environment | |
| CN105825332A (en) | Cloud computing based electric energy metering monitoring platform and method thereof | |
| CN103763399A (en) | Cloud server operation supporting system based on XEN virtualization framework | |
| CN104462581B (en) | Very fast file fingerprint extraction system and method based on the mapping of microchannel internal memory and Smart Slice | |
| Bloch et al. | Analysis and survey of issues in live virtual machine migration interferences |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term |
Granted publication date: 20121128 |
|
| CX01 | Expiry of patent term |