CN105760240A - Distributed task processing method and device - Google Patents
Distributed task processing method and device Download PDFInfo
- Publication number
- CN105760240A CN105760240A CN201410784173.7A CN201410784173A CN105760240A CN 105760240 A CN105760240 A CN 105760240A CN 201410784173 A CN201410784173 A CN 201410784173A CN 105760240 A CN105760240 A CN 105760240A
- Authority
- CN
- China
- Prior art keywords
- node
- delaying
- machine
- waiting task
- cluster
- 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
Landscapes
- Multi Processors (AREA)
Abstract
The invention discloses a distributed task processing method. The distributed task processing method comprises following steps: when a to-be-processed task from an acceptance terminal is received, a target node is selected from a node cluster comprising multiple nodes according to scheduled rules, and the to-be-processed task is distributed to the target node for processing; a running state of each node in the node cluster is monitored, and a shutdown node with a shutdown incident occurring is captured immediately; the to-be-processed task corresponding to the shutdown node is migrated to a normal node without the shutdown incident for processing. The invention further discloses a distributed task processing device. With the adoption of the distributed task processing method and the distributed task processing device, disaster tolerance capacity and business processing performance of an application system can be improved.
Description
Technical field
The present invention relates to technical field of data processing, particularly relate to a kind of distributed task scheduling processing method and a kind of distributed task scheduling processes device.
Background technology
In existing various application systems, major applications system is required for certain business of Timing Processing.Commonly used system can adopt the mode that single-point is disposed or multiple spot is disposed to process business.If application system adopts the mode that single-point is disposed, namely rely solely on individual node and process corresponding task, the problem that then there is Single Point of Faliure on the one hand, namely when the individual node processing this task delays machine event, whole application system all will be unable to normal use.On the other hand, owing to individual node cannot realize the concurrent processing to task, therefore single-point is disposed and be there is also the problem that performance is relatively low.If application system adopts the mode that multiple spot is disposed, namely it is jointly processed by corresponding task by multiple nodes, although the problem that Single Point of Faliure can be solved, but needs multinode to access certain public resource simultaneously, cause the concurrently access of resource, it is possible to can cause that partial task processes unsuccessfully.
Summary of the invention
For solving the problems referred to above, the present invention provides a kind of distributed task scheduling processing method and device, it is possible not only to solve Single Point of Faliure problem during application system process task, the concurrent processing of task can also be realized, thus being effectively improved redundancy ability and the service process performance of application system, there is High Availabitity, high-performance and feature easy to use.
Specifically, a kind of distributed task scheduling processing method, including: whenever receiving from when accepting a waiting task of end, from the node cluster including multiple node, select a destination node according to pre-defined rule, this waiting task is distributed to this destination node and processes；Monitor the running status of each node in this node cluster, immediately catch machine of the delaying node of machine event of delaying；Waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying process.
In one embodiment of the invention, above-mentioned distributed task scheduling processing method also includes: safeguard a node tasks relation table in data base, and this node tasks relation table is for recording the corresponding relation of each node and waiting task in node cluster.
In one embodiment of the invention, described the step of a destination node is selected to include from the node cluster including multiple node according to pre-defined rule: the quantity of the waiting task that each node is corresponding in query node cluster in described node tasks relation table；Quantity according to waiting task corresponding to node each in node cluster judges the busy extent of each node；Selecting the node that busy extent is minimum from node cluster is this destination node.
In one embodiment of the invention, the running status of each node in this node cluster of described monitoring, the step of machine of the delaying node immediately catching machine event of delaying includes: with each node in node cluster for node undetermined, detect whether this node undetermined network failure occurs, if so, this node undetermined machine of delaying node as machine event of delaying then is judged；If network failure does not occur this node undetermined, then whether inquire about the quantity of waiting task corresponding to this node undetermined further more than the first predetermined threshold value, if so, then judge this node undetermined machine of delaying node as machine event of delaying.
In one embodiment of the invention, described waiting task corresponding for this machine node of delaying is migrated the step that the normal node to machine event of delaying carries out processing include: be some groups of subtasks by waiting task cutting corresponding for this machine node of delaying, wherein often these some groups of subtasks, less than the second predetermined threshold value, are then migrated and process to described normal node by the quantity of the waiting task that group subtask includes in batches.
A kind of distributed task scheduling processes device, including: distribution module, for whenever receiving from when accepting a waiting task of end, selecting a destination node from the node cluster including multiple node according to pre-defined rule, this waiting task distributed to this destination node and processes；Monitoring module, for monitoring the running status of each node in this node cluster, catches machine of the delaying node of machine event of delaying immediately；Transferring module, processes for waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying.
In one embodiment of the invention, described distribution module is additionally operable in data base to safeguard a node tasks relation table, and this node tasks relation table is for recording the corresponding relation of each node and waiting task in node cluster.
In one embodiment of the invention, described a destination node is selected to include from the node cluster including multiple node according to pre-defined rule: the quantity of the waiting task that each node is corresponding in query node cluster in described node tasks relation table；Quantity according to waiting task corresponding to node each in node cluster judges the busy extent of each node；Selecting the node that busy extent is minimum from node cluster is this destination node.
In one embodiment of the invention, described monitoring module is used for: with each node in node cluster for node undetermined, detects whether this node undetermined network failure occurs, and if so, then judges this node undetermined machine of delaying node as machine event of delaying；If network failure does not occur this node undetermined, then whether inquire about the quantity of waiting task corresponding to this node undetermined further more than the first predetermined threshold value, if so, then judge this node undetermined machine of delaying node as machine event of delaying.
In one embodiment of the invention, described transferring module is used for: be some groups of subtasks by waiting task cutting corresponding for this machine node of delaying, wherein often these some groups of subtasks, less than the second predetermined threshold value, are then migrated and process to described normal node by the quantity of the waiting task that group subtask includes in batches.
The invention has the beneficial effects as follows: distributed task scheduling processing method provided by the present invention and device, use the multiple nodes in node cluster to process and accept the waiting task that end sends, the running status of node each in node cluster is monitored simultaneously, when capturing machine of the delaying node of machine event of delaying, waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying process, thus Single Point of Faliure problem when being possible not only to solve application system process task, the concurrent processing of task can also be realized, thus being effectively improved redundancy ability and the service process performance of application system, there is High Availabitity, high-performance and feature easy to use.
Accompanying drawing explanation
Fig. 1 is environment schematic during a kind of distributed task scheduling processing method application of the embodiment of the present invention.
Fig. 2 is the steps flow chart schematic diagram of a kind of distributed task scheduling processing method of the embodiment of the present invention.
Fig. 3 is the detail flowchart of step S2 in Fig. 2.
Fig. 4 is the functional block diagram of a kind of distributed task scheduling process device of the embodiment of the present invention.
In accompanying drawing, the labelling of each parts is as follows: 1: server end；2: network；3: accept end；10: distributed task scheduling processes device；101: distribution module；102: monitoring module；103: transferring module；S1, S2, S3, S21, S22, S23, S24: step.
Detailed description of the invention
Below in conjunction with accompanying drawing, presently preferred embodiments of the present invention is described in detail, so that advantages and features of the invention can be easier to be readily appreciated by one skilled in the art, thus protection scope of the present invention being made apparent clear and definite defining.
Consult shown in Fig. 1, for the embodiment of the present invention provide distributed task scheduling processing method application time environment schematic.In the present embodiment, this distributed task scheduling processing method is applied in server end 1, and this server end 1 carries out network service by network 2 with accepting end 3.Wherein, this server end 1 can include one or more server, and this server end 1 can also be that virtual cloud calculates module certainly.This accepts end 3 and installs and run in terminal, and this terminal can be such as smart mobile phone, panel computer, desk computer, portable computer or other similar arithmetic units.This network 2 can be arbitrary internetwork connection mode, for instance the Internet (Internet), mobile Internet (such as the 2G of telecom operators' offer, 3G network), LAN (wired or wireless) etc..
Being deployed with a node cluster in server end 1, this node cluster includes multiple node, and wherein each node can be used in processing the waiting task accepting end 3 submission.This waiting task can be such as open invoice, statistics invoiced amount etc..In the present embodiment, server end 1 can store the configuration file of this node cluster, specifically, it is possible to adopts the mode of data base or file to store.In this configuration file, record has the unique logical name of each node and reference address in node cluster, for identifying and accessing different nodes.Additionally, this node cluster is also provided with a web service interface, by this web service interface, it is possible to the running status of each node in monitor node cluster, including normal condition or machine state of delaying.
Consult shown in Fig. 2, for the steps flow chart schematic diagram of a kind of distributed task scheduling processing method of the embodiment of the present invention.This distributed task scheduling processing method comprises the following steps:
Step S1, whenever receiving from when accepting the waiting task holding 3, selects a destination node from the described node cluster including multiple node according to pre-defined rule, this waiting task is distributed to this destination node and processes.
Whenever receiving from, when accepting the waiting task holding 3, first this waiting task being preserved in the database.This waiting task is with unique mark, for instance with unique No. ID in specific field.Described from node cluster, select a destination node according to pre-defined rule, in the present embodiment, for selecting node that busy extent is minimum as this destination node from node cluster.Specifically, in node cluster, the busy extent of each node can judge according to the quantity of waiting task corresponding to each node.The waiting task that node is corresponding refers to the task of being already allocated to this node processing, and this node is presently processing or is about to process this task.The described data base of server end 1 safeguards there is a node tasks relation table.This node tasks relation table is for recording the corresponding relation of each node and waiting task in node cluster, and wherein each node refers to respective logical name, and each waiting task refers to each unique mark.Such as corresponding for node a waiting task is A and B, and waiting task corresponding for node b is C, D and E.
Whenever receiving from when accepting the waiting task holding 3, step S1 quantity of the waiting task that each node is corresponding in query node cluster in this node tasks relation table, quantity according to waiting task corresponding to each node judges the busy extent of each node, then selects the node that busy extent is minimum as described destination node from node cluster.If the quantity of the waiting task that node is corresponding is more many, then the busy extent of this node is more high, if the quantity of waiting task corresponding to node is more low, then the busy extent of this node is more low.
This waiting task received is distributed to this destination node and is processed after selecting described destination node by step S1.Specifically, step S1 according to the described node tasks relation table in destination node and this waiting task more new database, adds in this node tasks relation table by the corresponding relation of this destination node Yu this waiting task.In node cluster, each node will this node tasks relation table of timing scan, determining according to this node tasks relation table need to by the waiting task of this node processing, after obtaining this waiting task from data base, according to this waiting task of the type of process of this waiting task.Certainly, when corresponding waiting task is disposed by node, also sending the instruction that process terminates, step S1 will update described node tasks relation table according to this instruction, such as in node tasks relation table, specified sign position is carried out labelling, thus it is complete to indicate this waiting task to be processed.But, the record of still this node in store and the corresponding relation of the waiting task being disposed by it in node tasks relation table, in order to the follow-up process reviewing this waiting task is passed through.
Step S2, monitors the running status of each node in this node cluster, immediately catches machine of the delaying node of machine event of delaying.Step S2 can monitor the running status of each node in this node cluster by described web service interface, and this running status includes delay machine event and machine event of delaying.Wherein, to include node generation network failure, node busy or without response for this machine event of delaying.
Specifically, step S2 comprises the following steps:
Step S21, at interval of a period of time, with each node in node cluster for node undetermined, detects whether this node undetermined network failure occurs, and if so, then performs step S22, it is determined that this node undetermined is machine of the delaying node of machine event of delaying.By ping order, step S21 can detect whether this node undetermined network failure occurs.Additionally, step S22 is after judging that this node undetermined is as machine node of delaying, also needing to monitor whether this machine node of delaying recovers normal further, whether the network failure namely monitoring this machine node of delaying eliminates.Such as, after these maintained personnel of machine node that delay are restarted, this network failure delaying machine node is likely to be eliminated.Recovering after normally when monitoring this machine node of delaying, step S1 can select to have recovered to delay machine node normally as described destination node, and distributes waiting task to this destination node.If detecting there is not network failure in this node undetermined, then perform step S23.
Whether step S23, inquire about the quantity of waiting task corresponding to this node undetermined more than the first predetermined threshold value, for instance 20 from described node tasks relation table.If the quantity of the waiting task that this node undetermined is corresponding exceedes this first predetermined threshold value, illustrate that this node undetermined is in busy or without response state, then perform described step S22, it is determined that this node undetermined is machine of the delaying node of machine event of delaying.If the quantity of the waiting task that this node undetermined is corresponding is not less than this first predetermined threshold value, then perform step S24, it is determined that this node undetermined is the normal node of machine event of delaying.It should be noted that, load capacity for node each in node cluster is different, described first predetermined threshold value that each node is arranged can differ, this first predetermined threshold value of node that load capacity is stronger can be set to higher, this first predetermined threshold value of node that load capacity is more weak can be set to relatively low.
Additionally, step S2 can also safeguard a monitoring nodes table and a machine event table of delaying in data base.This monitoring nodes table for recording the monitoring period to each node of node cluster, whether each node delays machine event and recovers the information such as properly functioning time after machine event of delaying.This machine event table of delaying may be used for record and delays the information such as time of each node of machine event and machine event of delaying.Whenever catching machine of the delaying node of a machine event of delaying, step S2 can send the information attendant to server end 1 according to preset configuration, and prompting maintenance personnel pay close attention to or maintenance.In this preset configuration, record has when a certain node delays machine event, if need to send information to this attendant, the name of attendant and receiving party's formula, for instance by information such as short message mode receptions.This information can generate according to described monitoring nodes table and machine event table of delaying.When using this node cluster to process waiting task after considering to dispose described node cluster first, the too much situation of waiting task number to the distribution of this node cluster, in order to avoid the problem that the prompting to attendant is too much, system can also be supported to close prompt facility in advance.Return after normally until system application and can be then turned on this prompt facility.
Step S3, migrates the normal node to machine event of delaying by waiting task corresponding for this machine node of delaying and processes.
Specifically, step S3 judges the busy extent of each normal node according to the quantity of waiting task corresponding to each normal node of machine event of delaying in node cluster, then select one or more normal node according to the busy extent of each normal node, waiting task corresponding for this machine node of delaying is migrated and processes to selected normal node.Step S3 can inquire about the quantity of waiting task corresponding to each normal node in described node tasks relation table, then judges the busy extent of each normal node according to the quantity of waiting task corresponding to each normal node.If the quantity of the waiting task that node is corresponding is more many, then the busy extent of this node is more high, if the quantity of waiting task corresponding to node is more low, then the busy extent of this node is more low.
Then, step S3 selects one or more normal node according to the busy extent of each normal node, such as step S3 can select the normal node that busy extent is minimum, after each normal node can also being sorted from low to high according to busy extent, select to come multiple normal node of predetermined number above.Waiting task corresponding for machine node of delaying is migrated and processes to selected normal node by step S3, delete by the corresponding relation of this delay machine node and this waiting task in this node tasks relation table, and record the corresponding relation of selected normal node and this waiting task.Selected normal node is by this node tasks relation table of timing scan, determining from this node tasks relation table need to by the waiting task of this node processing, after obtaining this waiting task from data base, according to this waiting task of the type of process of this waiting task.
Owing to the quantity of waiting task corresponding to machine node of delaying is likely to more, if all waiting tasks corresponding for machine node of delaying all are migrated to selected normal node process disposable, then this normal node is likely to the machine event of delaying because the quantity of corresponding waiting task exceedes described first predetermined threshold value.Therefore, step S3 can first by waiting task cutting corresponding for machine node of delaying be several group subtask, wherein often these some groups of subtasks, less than the second predetermined threshold value, are then migrated and process to described normal node by the quantity of the waiting task that group subtask includes in batches.Such as, the waiting task that machine node is corresponding if delaying is 1000, then these 1000 waiting tasks can be divided into one group with every 50 by step S3, thus being 20 groups of subtasks by these 1000 waiting task cuttings, then these 20 groups of subtasks being migrated in batches and processing to described normal node.
In addition, in order to avoid selected normal node because the quantity of corresponding waiting task exceedes described first predetermined threshold value machine event of delaying, automatically this first predetermined threshold value is heightened, by this first predetermined threshold value plus preset increments, the first predetermined threshold value after being heightened.After the waiting task migrated by machine node of delaying and come all is disposed by normal node, then the first predetermined threshold value after heightening is renewed back to initial value.
Consult shown in Fig. 4, for the functional block diagram of a kind of distributed task scheduling process device 10 of the embodiment of the present invention.This distributed task scheduling processes device 10 and includes distribution module 101, monitoring module 102 and transferring module 103..It is appreciated that above-mentioned each module refers to computer program or program segment, is used for performing certain one or more specific function.Additionally, the program code that the differentiation of above-mentioned each module does not represent reality also must separate.
Distribution module 101, for whenever receiving from when accepting a waiting task of end, selecting a destination node according to pre-defined rule, this waiting task distributed to this destination node and processes from the node cluster including multiple node；
Monitoring module 102, for monitoring the running status of each node in this node cluster, catches machine of the delaying node of machine event of delaying immediately；
Transferring module 103, processes for waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying.
For the specific works process of above each module, the distributed task scheduling processing method that can provide with further reference to the embodiment of the present invention, do not repeat them here.
In sum, the distributed task scheduling processing method of the present embodiment offer and device, use the multiple nodes in node cluster to process and accept the waiting task that end sends, the running status of node each in node cluster is monitored simultaneously, when capturing machine of the delaying node of machine event of delaying, waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying process, thus Single Point of Faliure problem when being possible not only to solve application system process task, the concurrent processing of task can also be realized, thus being effectively improved redundancy ability and the service process performance of application system, there is High Availabitity, high-performance and feature easy to use.
The foregoing is only embodiments of the invention; not thereby the scope of the claims of the present invention is limited; every equivalent structure utilizing description of the present invention and accompanying drawing content to make or equivalence flow process conversion; or directly or indirectly it is used in other relevant technical fields, all in like manner include in the scope of patent protection of the present invention.
Claims (10)
1. a distributed task scheduling processing method, it is characterised in that the method comprises the following steps:
Whenever receiving from when accepting a waiting task of end, from the node cluster including multiple node, select a destination node according to pre-defined rule, this waiting task is distributed to this destination node and processes；
Monitor the running status of each node in this node cluster, immediately catch machine of the delaying node of machine event of delaying；
Waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying process.
2. distributed task scheduling processing method as claimed in claim 1, it is characterised in that the method also includes:
Safeguarding a node tasks relation table in data base, this node tasks relation table is for recording the corresponding relation of each node and waiting task in node cluster.
3. distributed task scheduling processing method as claimed in claim 2, it is characterised in that described select the step of a destination node to include from the node cluster including multiple node according to pre-defined rule:
The quantity of the waiting task that each node is corresponding in query node cluster in described node tasks relation table；
Quantity according to waiting task corresponding to node each in node cluster judges the busy extent of each node；
Selecting the node that busy extent is minimum from node cluster is this destination node.
4. distributed task scheduling processing method as claimed in claim 1, it is characterised in that the running status of each node in this node cluster of described monitoring, the step of machine of the delaying node immediately catching machine event of delaying includes:
With each node in node cluster for node undetermined, detect whether this node undetermined network failure occurs, if so, then judge this node undetermined machine of delaying node as machine event of delaying；
If there is not network failure in this node undetermined, whether then inquire about the quantity of waiting task corresponding to this node undetermined further more than the first predetermined threshold value, if, then judge this node undetermined machine of delaying node as machine event of delaying, preferably distinguish to some extent with machine of the delaying node of network failure node, so can be clearer and more definite.
5. distributed task scheduling processing method as claimed in claim 1, it is characterised in that described waiting task corresponding for this machine node of delaying is migrated the step that the normal node to machine event of delaying carries out processing include:
Being some groups of subtasks by waiting task cutting corresponding for this machine node of delaying, wherein often these some groups of subtasks, less than the second predetermined threshold value, are then migrated and process to described normal node by the quantity of the waiting task that group subtask includes in batches.
6. a distributed task scheduling processes device, it is characterised in that this device includes:
Distribution module, for whenever receiving from when accepting a waiting task of end, selecting a destination node according to pre-defined rule, this waiting task distributed to this destination node and processes from the node cluster including multiple node；
Monitoring module, for monitoring the running status of each node in this node cluster, immediately catches machine of the delaying node of machine event of delaying, and triggers task immigration activity；
Transferring module, processes for waiting task corresponding for this machine node of delaying is migrated the normal node to machine event of delaying.
7. distributed task scheduling as claimed in claim 6 processes device, it is characterized in that, described distribution module is additionally operable in data base to safeguard a node tasks relation table, and this node tasks relation table is for recording the corresponding relation of each node and waiting task in node cluster.
8. distributed task scheduling as claimed in claim 7 processes device, it is characterised in that described select a destination node to include from the node cluster including multiple node according to pre-defined rule:
The quantity of the waiting task that each node is corresponding in query node cluster in described node tasks relation table；
Quantity according to waiting task corresponding to node each in node cluster judges the busy extent of each node；
Selecting the node that busy extent is minimum from node cluster is this destination node.
9. distributed task scheduling as claimed in claim 6 processes device, it is characterised in that described monitoring module is used for:
With each node in node cluster for node undetermined, detect whether this node undetermined network failure occurs, if so, then judge this node undetermined machine of delaying node as machine event of delaying；
If network failure does not occur this node undetermined, then whether inquire about the quantity of waiting task corresponding to this node undetermined further more than the first predetermined threshold value, if so, then judge this node undetermined machine of delaying node as machine event of delaying.
10. distributed task scheduling as claimed in claim 6 processes device, it is characterised in that described transferring module is used for:
Being some groups of subtasks by waiting task cutting corresponding for this machine node of delaying, wherein often these some groups of subtasks, less than the second predetermined threshold value, are then migrated and process to described normal node by the quantity of the waiting task that group subtask includes in batches.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410784173.7A CN105760240A (en) | 2014-12-16 | 2014-12-16 | Distributed task processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410784173.7A CN105760240A (en) | 2014-12-16 | 2014-12-16 | Distributed task processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105760240A true CN105760240A (en) | 2016-07-13 |
Family
ID=56339215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410784173.7A Pending CN105760240A (en) | 2014-12-16 | 2014-12-16 | Distributed task processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105760240A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106375373A (en) * | 2016-08-24 | 2017-02-01 | 广西小草信息产业有限责任公司 | Task decomposition method and system based on dynamic cloud nodes |
| CN107526648A (en) * | 2017-08-10 | 2017-12-29 | 天脉聚源（北京）科技有限公司 | A kind of node device that handles is delayed the method and device of machine |
| CN108446177A (en) * | 2018-02-07 | 2018-08-24 | 平安普惠企业管理有限公司 | A kind of task processing method, computer readable storage medium and terminal device |
| CN108665235A (en) * | 2018-04-27 | 2018-10-16 | 平安科技（深圳）有限公司 | Operation method, terminal device and the medium of resource handling node |
| CN109343930A (en) * | 2018-09-11 | 2019-02-15 | 郑州云海信息技术有限公司 | The management method and device of virtual machine in cloud computing system |
| CN109656706A (en) * | 2018-12-25 | 2019-04-19 | 江苏满运软件科技有限公司 | Distributed task dispatching method, system, equipment and medium |
| CN109656701A (en) * | 2018-12-19 | 2019-04-19 | 北京达佳互联信息技术有限公司 | A kind of dispatching method, device, electronic equipment and readable storage medium storing program for executing |
| CN110096358A (en) * | 2019-04-11 | 2019-08-06 | 上海交通大学 | Chain drive remote center distributed storage and distributed computing method |
| CN110275762A (en) * | 2019-04-30 | 2019-09-24 | 北京媒球信息科技有限公司 | Calculation method, server, mobile terminal and device |
| CN110365809A (en) * | 2019-07-23 | 2019-10-22 | 中南民族大学 | Distributed server address configuration system and method |
| CN110968420A (en) * | 2018-09-30 | 2020-04-07 | 北京国双科技有限公司 | Scheduling method and device for multi-crawler platform, storage medium and processor |
| CN111385134A (en) * | 2018-12-30 | 2020-07-07 | 浙江宇视科技有限公司 | Access device dynamic migration method and device access platform |
| CN111753997A (en) * | 2020-06-28 | 2020-10-09 | 北京百度网讯科技有限公司 | Distributed training method, system, device and storage medium |
| CN111897494A (en) * | 2020-07-27 | 2020-11-06 | 星辰天合（北京）数据科技有限公司 | Target file processing method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521044A (en) * | 2011-12-30 | 2012-06-27 | 北京拓明科技有限公司 | Distributed task scheduling method and system based on messaging middleware |
| US20120180050A1 (en) * | 2007-09-28 | 2012-07-12 | Oracle America, Inc. | Method and system for coordinating hypervisor scheduling |
| CN102694868A (en) * | 2012-06-07 | 2012-09-26 | 浪潮电子信息产业股份有限公司 | Cluster system implementation and task dynamic distribution method |
| CN103957231A (en) * | 2014-03-18 | 2014-07-30 | 成都盛思睿信息技术有限公司 | Virtual machine distributed task scheduling method under cloud calculating platform |
-
2014
- 2014-12-16 CN CN201410784173.7A patent/CN105760240A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120180050A1 (en) * | 2007-09-28 | 2012-07-12 | Oracle America, Inc. | Method and system for coordinating hypervisor scheduling |
| CN102521044A (en) * | 2011-12-30 | 2012-06-27 | 北京拓明科技有限公司 | Distributed task scheduling method and system based on messaging middleware |
| CN102694868A (en) * | 2012-06-07 | 2012-09-26 | 浪潮电子信息产业股份有限公司 | Cluster system implementation and task dynamic distribution method |
| CN103957231A (en) * | 2014-03-18 | 2014-07-30 | 成都盛思睿信息技术有限公司 | Virtual machine distributed task scheduling method under cloud calculating platform |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106375373A (en) * | 2016-08-24 | 2017-02-01 | 广西小草信息产业有限责任公司 | Task decomposition method and system based on dynamic cloud nodes |
| CN107526648A (en) * | 2017-08-10 | 2017-12-29 | 天脉聚源（北京）科技有限公司 | A kind of node device that handles is delayed the method and device of machine |
| CN108446177A (en) * | 2018-02-07 | 2018-08-24 | 平安普惠企业管理有限公司 | A kind of task processing method, computer readable storage medium and terminal device |
| CN108446177B (en) * | 2018-02-07 | 2020-05-15 | 平安普惠企业管理有限公司 | Task processing method, computer readable storage medium and terminal device |
| WO2019205307A1 (en) * | 2018-04-27 | 2019-10-31 | 平安科技（深圳）有限公司 | Resource processing node operation method and device, and terminal device and medium |
| CN108665235A (en) * | 2018-04-27 | 2018-10-16 | 平安科技（深圳）有限公司 | Operation method, terminal device and the medium of resource handling node |
| CN108665235B (en) * | 2018-04-27 | 2023-04-18 | 平安科技（深圳）有限公司 | Resource processing node operation method, terminal device and medium |
| CN109343930A (en) * | 2018-09-11 | 2019-02-15 | 郑州云海信息技术有限公司 | The management method and device of virtual machine in cloud computing system |
| CN110968420A (en) * | 2018-09-30 | 2020-04-07 | 北京国双科技有限公司 | Scheduling method and device for multi-crawler platform, storage medium and processor |
| CN109656701A (en) * | 2018-12-19 | 2019-04-19 | 北京达佳互联信息技术有限公司 | A kind of dispatching method, device, electronic equipment and readable storage medium storing program for executing |
| CN109656706A (en) * | 2018-12-25 | 2019-04-19 | 江苏满运软件科技有限公司 | Distributed task dispatching method, system, equipment and medium |
| CN111385134A (en) * | 2018-12-30 | 2020-07-07 | 浙江宇视科技有限公司 | Access device dynamic migration method and device access platform |
| CN111385134B (en) * | 2018-12-30 | 2022-12-13 | 浙江宇视科技有限公司 | Access device dynamic migration method and device access platform |
| CN110096358A (en) * | 2019-04-11 | 2019-08-06 | 上海交通大学 | Chain drive remote center distributed storage and distributed computing method |
| CN110275762A (en) * | 2019-04-30 | 2019-09-24 | 北京媒球信息科技有限公司 | Calculation method, server, mobile terminal and device |
| CN110365809A (en) * | 2019-07-23 | 2019-10-22 | 中南民族大学 | Distributed server address configuration system and method |
| CN111753997A (en) * | 2020-06-28 | 2020-10-09 | 北京百度网讯科技有限公司 | Distributed training method, system, device and storage medium |
| CN111753997B (en) * | 2020-06-28 | 2021-08-27 | 北京百度网讯科技有限公司 | Distributed training method, system, device and storage medium |
| CN111897494A (en) * | 2020-07-27 | 2020-11-06 | 星辰天合（北京）数据科技有限公司 | Target file processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105760240A (en) | Distributed task processing method and device | |
| CN111950988B (en) | Distributed workflow scheduling method and device, storage medium and electronic equipment | |
| CN109634730B (en) | Task scheduling method, device, computer equipment and storage medium | |
| EP3489825A1 (en) | Method, apparatus and computer readable storage medium for processing service | |
| JP2012521607A (en) | Distributed application monitoring | |
| JP2014528126A (en) | Distributing multi-source push notifications to multiple targets | |
| CN105516266B (en) | A kind of business monitoring method, system and relevant device | |
| US20240039984A1 (en) | Rule-based action triggering in a provider network | |
| CN109361525B (en) | Method, device, control terminal and medium for restarting distributed deployment of multiple services | |
| EP3126961A1 (en) | Monitoring of node.js applications | |
| CN110750426A (en) | Service state monitoring method and device, electronic equipment and readable storage medium | |
| US9948502B2 (en) | Method and system for self-healing of communication network | |
| WO2016175850A1 (en) | Detection of node.js memory leaks | |
| CN105096014A (en) | Method and system for recording work operation condition remotely | |
| CN113238849A (en) | Timed task processing method and device, storage medium and electronic equipment | |
| CN111376255B (en) | Robot data acquisition method and device and terminal equipment | |
| CN108154343B (en) | Emergency processing method and system for enterprise-level information system | |
| CN116431290A (en) | Job scheduling method, apparatus, device, medium and program product | |
| CN110764838A (en) | Service model loading method and system, electronic equipment and storage medium | |
| CN103618643B (en) | A kind of message queue dynamic alert monitoring method and device | |
| CN103514044A (en) | Resource optimization method, device and system of dynamic behavior analysis system | |
| US20160261476A1 (en) | Message system for avoiding processing-performance decline | |
| CN110087232A (en) | A kind of call processing method based on smart machine, device and server | |
| US10228958B1 (en) | Systems and methods for archiving time-series data during high-demand intervals | |
| US9240968B1 (en) | Autogenerated email summarization process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160713 |
|
| RJ01 | Rejection of invention patent application after publication |