CN104850394B - The management method and distributed system of distributed application program - Google Patents

The management method and distributed system of distributed application program Download PDF

Info

Publication number
CN104850394B
CN104850394B CN201510184249.7A CN201510184249A CN104850394B CN 104850394 B CN104850394 B CN 104850394B CN 201510184249 A CN201510184249 A CN 201510184249A CN 104850394 B CN104850394 B CN 104850394B
Authority
CN
China
Prior art keywords
parent process
subprocess
application program
distributed
strategy
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.)
Active
Application number
CN201510184249.7A
Other languages
Chinese (zh)
Other versions
CN104850394A (en
Inventor
曹东刚
刘澜涛
梅宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Peking University
Original Assignee
Peking University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Peking University filed Critical Peking University
Priority to CN201510184249.7A priority Critical patent/CN104850394B/en
Publication of CN104850394A publication Critical patent/CN104850394A/en
Application granted granted Critical
Publication of CN104850394B publication Critical patent/CN104850394B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Stored Programmes (AREA)

Abstract

The present invention provides a kind of management method and distributed system of distributed application program, including：According to the logical relation of each process in distributed application program, the process tree of distributed application program is established；Determine the subprocess of each parent process and each parent process in process tree；The Distribution Strategy of the subprocess of parent process is added in each parent process；Each parent process is created according to the order of process tree from top to bottom successively, and according to the Distribution Strategy of subprocess and the resource information of distributed system of each parent process, determines to be distributed the target device of the subprocess of each parent process；The subprocess of each parent process is created in the target device of subprocess of each parent process is distributed.The management method and distributed system of distributed application program provided by the invention, when enabling distributed system to be distributed distributed application program automatically, it is possible to increase the efficiency of distributed application program exploitation.

Description

The management method and distributed system of distributed application program
Technical field
The present invention relates to the management method and distribution of field of computer technology, more particularly to a kind of distributed application program System.
Background technology
Distributed Application is to be distributed application program on a different computer, is accomplished a task jointly by network Working method.Since Distributed Application can improve the speed of service of application program, so can be lifted user perception and Satisfaction, therefore, more and more application programs have all been developed to distributed application program.
At present, there are many general Distributed Architecture on the market, user can be opened by any general Distributed Architecture Send out distributed application program various types of, and it is distributed to pass through the distributed system operation corresponding to general Distributed Architecture Application program.In order to make distributed system automatically can be distributed distributed application program when running distributed application program On computer in a network, user is needed in development DDB application program, is set with reference to the resource information of distributed system Determine distributing position of the distributed application program in distributed system.
But when realizing that distributed system is distributed distributed application program automatically using aforesaid way, user needs to consume Taking the substantial amounts of time removes setting distributed application program distributing position in a distributed system.
The content of the invention
The present invention provides a kind of management method and distributed system of distributed application program, to solve in the prior art In order to enable distributed system to be distributed distributed application program automatically, user is developing distribution using general Distributed Architecture , it is necessary to take a substantial amount of time the distributing position of setting distributed application program in a distributed system during formula application program Problem.
To reach above-mentioned purpose, the embodiment of the present invention adopts the following technical scheme that：
In a first aspect, the present invention provides a kind of management method of distributed application program, including：
According to the logical relation of each process in distributed application program, the process tree of the distributed application program is established；
Determine the subprocess of each parent process and each parent process in the process tree；
The Distribution Strategy of the subprocess of the parent process is added in each parent process；Wherein, the parent process The Distribution Strategy of subprocess is for indicating the distribution mode of the subprocess of the parent process in a distributed system；
Each parent process is created according to the order of the process tree from top to bottom successively, and according to each father into The resource information of the Distribution Strategy of the subprocess of journey and the distributed system, determines the subprocess of each parent process of distribution Target device；
The subprocess of each parent process is created in the target device of the subprocess of each parent process of distribution.
With reference to first aspect, in the first possible embodiment of first aspect, the method further includes：
The abnormality processing strategy of the subprocess of the parent process is added in each parent process；Wherein, the father into The abnormality processing strategy of the subprocess of journey is used to indicate when the subprocess of the parent process is abnormal, to the son of the parent process into The processing mode of journey.
The possible embodiment of with reference to first aspect the first, in second of possible embodiment of first aspect In, the abnormality processing strategy of the subprocess of the parent process includes at least following one kind：Ignore abnormality processing strategy, reboot process Strategy, exits processing strategy；
Wherein, the abnormality processing strategy of ignoring is the place for ignoring the unexpected message that the subprocess of the parent process reports Reason strategy；
The reboot process strategy is the processing strategy for restarting the subprocess for the parent process for reporting unexpected message；
The processing strategy that exits is that will report the subprocess of the parent process of unexpected message processing plan out of service Slightly.
Second of possible embodiment of the possible embodiment of with reference to first aspect the first or first aspect, In the third possible embodiment of first aspect, the abnormality processing strategy of the subprocess of the parent process is additionally operable to instruction and works as When the subprocess of the parent process is abnormal, to the processing rule of other subprocess of the parent process.
The third possible embodiment with reference to first aspect, in the 4th kind of possible embodiment of first aspect In, the abnormality processing strategy of the subprocess of the parent process also includes at least following one kind：Individually processing rule, all handles rule Then；
Wherein, the individually processing rule is the processing rule not processed to other subprocess of the parent process；Institute It is that the processing handled using same abnormality processing strategy whole subprocess of the parent process is advised to state all processing rules Then.
Any one of with reference to first aspect to second of possible embodiment of first aspect, the of first aspect In five kinds of possible embodiments, the Distribution Strategy of the subprocess of the parent process includes at least following one kind：Local distribution plan Slightly, integrated distribution strategy, Arbitrary distribution strategy；
Wherein, the local Distribution Strategy is that the subprocess of the parent process and the parent process is distributed in same target Distribution Strategy in equipment；
The integrated distribution strategy is that all subprocess of the parent process are distributed in the distribution plan in same target device Slightly；
The Arbitrary distribution strategy is that the subprocess of the parent process and the parent process is according to the distributed system The Distribution Strategy of resource information Arbitrary distribution.
Any one of with reference to first aspect to second of possible embodiment of first aspect, the of first aspect In six kinds of possible embodiments, the method further includes：
Monitoring strategies are added respectively in the subprocess of each parent process and each parent process；The monitoring plan Slightly it is used to indicate the monitoring relation between the subprocess of each parent process and each parent process.
The 6th kind of possible embodiment with reference to first aspect, in the 7th kind of possible embodiment of first aspect In, it is described when creating the subprocess of each parent process in the target device of the subprocess of each parent process of distribution Method further includes：
According to the monitoring strategies, the monitoring for establishing the subprocess of each parent process and each parent process connects Connect, so that the heartbeat that the subprocess of the parent process and the parent process connects mutual sending cycle by the monitoring is believed Breath.
Second aspect, the present invention provide a kind of distributed system, including：Multiple resource apparatus and distributed platform；Wherein, The distributed platform is used for the management method for performing the distributed application program as described in claim 1-8 is any；It is described more A resource apparatus is used to run the distributed application program.
The management method and distributed system of distributed application program provided by the invention, by default multiple general Distribution Strategy so that user, can be with when developing various types of distributed application programs using general Distributed Architecture According to the programmed logic of each distributed application program, in each parent process in the process tree of each distributed application program Be incorporated as each parent process subprocess selection Distribution Strategy so that distributed system according to distributed application program into , can be according in each parent process when the orders of Cheng Shu from top to bottom create each parent process of distributed application program successively The Distribution Strategy of the subprocess of parent process and the resource information of distributed system, select suitable target device create each father into The subprocess of journey, makes distributed system realize automatic distribution distributed application program.With user in the prior art using logical When Distributed Architecture develops each distributed application program, it is necessary to reference to distributed system resource information, expend a large amount of Time go to set the distributing position of each distributed application program in a distributed system, i.e. needs take a substantial amount of time Code is write for each distributed application program, to realize the automatic distribution phase of distributed application program in a distributed system Than the management method and distributed system of the distributed application program that the embodiment of the present invention is provided, can save User Exploitation The time of distributed application program, improves the efficiency of distributed application program exploitation.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this Some embodiments of invention, for those of ordinary skill in the art, without creative efforts, can be with Other attached drawings are obtained according to these attached drawings.
The flow chart of the management method embodiment one for the distributed application program that Fig. 1 is provided by the embodiment of the present invention；
The structure diagram of the process tree for the distributed application program that Fig. 2 is provided by the embodiment of the present invention；
The flow chart of the management method embodiment two for the distributed application program that Fig. 3 is provided by the embodiment of the present invention；
The flow chart of the management method embodiment three for the distributed application program that Fig. 4 is provided by the embodiment of the present invention；
The flow chart of the management method example IV for the distributed application program that Fig. 5 is provided by the embodiment of the present invention；
The parent process and the subprocess of parent process that Fig. 6 is provided by the embodiment of the present invention use the distribution of local Distribution Strategy Schematic diagram；
The parent process and the subprocess of parent process that Fig. 7 is provided by the embodiment of the present invention use the distribution of integrated distribution strategy Schematic diagram；
The parent process and the subprocess of parent process that Fig. 8 is provided by the embodiment of the present invention use the distribution of Arbitrary distribution strategy Schematic diagram；
Fig. 9 adds relaxation by the parent process and the subprocess of parent process that the embodiment of the present invention provides using integrated distribution strategy The distribution schematic diagram of Distribution Strategy；
The flow chart of the management method embodiment five for the distributed application program that Figure 10 is provided by the embodiment of the present invention；
The flow chart of the management method embodiment six for the distributed application program that Figure 11 is provided by the embodiment of the present invention；
The structure diagram for the distributed system that Figure 12 is provided by the embodiment of the present invention.
Embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, those of ordinary skill in the art are obtained every other without creative efforts Embodiment, belongs to the scope of protection of the invention.
The embodiment of the present invention can be adapted in the distributed system corresponding to general Distributed Architecture, it can be realized The automatic distributed application program of management operating in a distributed system.Therefore, Distributed Application provided in an embodiment of the present invention The management method and distributed system of program, it is intended to solve in the prior art to enable distributed system to be distributed distribution automatically Formula application program, user is when using general Distributed Architecture development DDB application program, it is necessary to take a substantial amount of time The problem of going to set the distributing position of distributed application program in a distributed system.
It should be noted that when distributed application program is run in a distributed system, process is Distributed Application journey The execution entity of sequence in a distributed system, it can apply for and possess the resource of distributed system.Due to Distributed Application journey Sequence is performed parallel by multiple resource apparatus in distributed system, thus distributed application program can correspond to it is multiple into Journey.Therefore, run on the multiple resource apparatus of distributed application program in a distributed system it can be understood that being answered to be distributed Run with the multiple resource apparatus of multiple processes of program in a distributed system, between process by multiple resource apparatus it Between network connection communicate with each other.
It should be noted that for the ease of describing, in the following each embodiments of the present invention, " User Exploitation Distributed Application journey Sequence " acute pyogenic infection of finger tip " user uses general Distributed Architecture development DDB application program ", " distributed system " acute pyogenic infection of finger tip are " general The corresponding distributed system of Distributed Architecture ".
Technical scheme is described in detail with specifically embodiment below.These specific implementations below Example can be combined with each other, and may be repeated no more for the same or similar concept or process in some embodiments.
The flow chart of the management method embodiment one for the distributed application program that Fig. 1 is provided by the embodiment of the present invention.Its In, the executive agent of this method can be to have the function of the communication equipment of management of process or communication module in distributed system, As shown in Figure 1, this method may include steps of：
Step S101：According to the logical relation of each process in distributed application program, establish distributed application program into Cheng Shu.
Specifically, the logical relation of each process can be understood as each process with being created by the process in distributed application program Establishment relation between the process built, and each process and create the process process between be created relation, by upper The tree-like process creation relation of distributed application program can be established out by stating establishment relation and being created relation, i.e. distributed The process tree of application program.
The structure diagram of the process tree for the distributed application program that Fig. 2 is provided by the embodiment of the present invention, such as Fig. 2 institutes Showing, it is assumed that certain distributed application program has process A, B, C and D, wherein, process A creates process B and process C, process B establishment Process D, then, the process tree of the distributed application program can be as shown in Figure 2.
Step S102：Determine the subprocess of each parent process and each parent process in process tree.
Specifically, in the process tree of a distributed application program, each process is by the process of the process creation Parent process, then by the process of the process creation be the parent process subprocess.For example, with continued reference to Fig. 2, according to Fig. 2 The process tree of distributed application program, by taking process B and process D as an example, since process B creates process D, then process B for father into Journey, process D are the subprocess of parent process B.
In the process tree of a distributed application program, process can be the parent process of a subprocess, can be with For the subprocess of another parent process, that is to say, that according to the order of the process tree of distributed application program from top to bottom, except most Outside the process (that is, the process of last layer) of end, all processes can be some other processes or certain it is multiple into The parent process of journey.Such as：With continued reference to Fig. 2, the process tree of the distributed application program according to Fig. 2, the Distributed Application In the process tree of program in addition to the process C and process D of end, including two parent processes, it is respectively：Process A and process B. Process A is the parent process of subprocess B and subprocess C, and process B is the parent process of subprocess D；Wherein, process B is subprocess D's Parent process, process B are also the subprocess of parent process A at the same time.
In the process tree of a distributed application program, the quantity of the subprocess of the corresponding parent process of different parent processes Can be different.For example, with continued reference to Fig. 2, the process tree of the distributed application program according to Fig. 2, the subprocess of parent process A There are two, be respectively：Process B and process C；The subprocess of parent process B only has one, i.e. process D.
Step S103：The Distribution Strategy of the subprocess of parent process is added in each parent process；Wherein, parent process son into The Distribution Strategy of journey is for indicating the distribution mode of the subprocess of parent process in a distributed system.
Specifically, the Distribution Strategy of the subprocess of parent process described in the embodiment of the present invention can have multiple, each distribution A kind of distribution mode of the corresponding subprocess of parent process of strategy in a distributed system, the plurality of Distribution Strategy is general distribution Strategy, that is to say, that the subprocess of each parent process in the process tree of any type of distributed application program, can be A Distribution Strategy for being adapted to oneself distribution mode in a distributed system is found in the plurality of Distribution Strategy.During specific implementation, User can be in development DDB application program, according to the programmed logic of distributed application program, i.e. distributed application program Process tree in each parent process subprocess and other of parent process and parent process subprocess between communication frequency, The Distribution Strategy for the subprocess for being adapted to parent process is added in each parent process.For example, if the subprocess of parent process leads to parent process Letter is frequent, then can be that the subprocess of the parent process selects a kind of subprocess by parent process and parent process to be distributed in same resource Distribution Strategy in equipment, to improve the operational efficiency of distributed application program, reduces the communication overhead in distributed system.
The embodiment of the present invention to the representation of the Distribution Strategy of the subprocess of above-mentioned parent process without limiting, such as can Think the parameter that distributed system can identify, code block corresponding to parameter is set in a distributed system, during specific implementation, User can be each in the process tree of each distributed application program when developing different types of distributed application program The subprocess of parent process selects suitable Distribution Strategy, and the corresponding parameter of the Distribution Strategy of the subprocess of each parent process is added Enter each parent process, without taking a substantial amount of time the resource information with reference to distributed system again, for each distribution The distributing position of formula application setting in a distributed system.
Optionally, when a certain parent process in the process tree of distributed application program has multiple subprocess, the parent process The Distribution Strategy of each subprocess may be the same or different.Such as：With continued reference to Fig. 2, the distributed application program The subprocess of parent process A in process tree has two, is respectively：The distribution of process B and process C, then subprocess B and subprocess C Strategy can be same Distribution Strategy, or different Distribution Strategies.
Step S104：Each parent process is created according to the order of process tree from top to bottom successively, and according to each parent process The Distribution Strategy of subprocess and the resource information of distributed system, determine that the target of the subprocess of each parent process of distribution is set It is standby.
Specifically, the resource information of above-mentioned distributed system can be the available appearance of each resource apparatus in distributed system Amount, available bandwidth, CPU occupation rates etc..Above-mentioned target device is that the resource that distributed application program is run in distributed system is set It is standby, such as：Computer, server, virtual machine etc..After above-mentioned distributed application program is submitted to distributed system, due to The Distribution Strategy of the subprocess of parent process is added in each parent process in the process tree of distributed application program, then distributed system System can be automatically created according to the order of the process tree of distributed application program from top to bottom and start distributed application program First parent process in process tree, i.e., a process of top in process tree, and while the parent process is created, according to The Distribution Strategy of the subprocess of the added parent process in the parent process, it is automatic true with reference to the resource information of distributed system Surely it is distributed the target device of the subprocess of the parent process.In this manner, it may be determined that in the process tree of distributed application program All parent processes subprocess target device.In this way, various types of distributions of the distributed system in operation User Exploitation , can be with by the above-mentioned means, each father in automatically determining the process tree of each distributed application program during formula application program The target device of the subprocess of process.
By taking the parameter that the Distribution Strategy of the subprocess of above-mentioned parent process can identify for distributed system as an example, wherein, ginseng The corresponding code blocks of number are set in a distributed system, then when above-mentioned distributed application program is submitted to distributed system Afterwards, the Distribution Strategy (i.e. parameter) of the subprocess for the parent process that distributed system adds in reading or getting each parent process When, it is possible to stored corresponding code block of Distribution Strategy (i.e. parameter) in a distributed system is called, and combines distribution The resource information of formula system, determines the target device of the subprocess of all parent processes in the process tree of distributed application program.
Optionally, when a certain parent process in the process tree of distributed application program has multiple subprocess, if the plurality of The Distribution Strategy of subprocess is identical, then the target device for being distributed the plurality of subprocess can be same.
Step S105：The subprocess of each parent process is created in the target device of subprocess of each parent process is distributed.
Specifically, the son of each parent process when above-mentioned distributed system in the process tree of distributed application program into The Distribution Strategy of journey and the resource information of distributed system, determine to be distributed after the target device of the subprocess of each parent process, Distributed system can create the subprocess of parent process in corresponding target device automatically, and in this manner, until creating All processes in complete process tree.So far, distributed system realizes automatic distribution distributed application program.
The distributed application program using universal distributed Development of Framework is directed in the prior art, in order to make distributed system When running distributed application program, distributed application program can be distributed on the resource apparatus of distributed system automatically, User is needed in development DDB application program, is being divided with reference to the resource information setting distributed application program of distributed system The distributing position of cloth system, causes user when developing each distributed application program using general Distributed Architecture, Need to take a substantial amount of time and set the distributing position of each distributed application program in a distributed system, i.e. need to consume To take the substantial amounts of time write code for each distributed application program, to realize distributed application program in a distributed system Automatic distribution.And in the embodiment of the present invention, user, can be according to distribution when using general framework development DDB application program The programmed logic of formula application program, divides for the subprocess of each parent process in the process tree of distributed application program in general Selection one is adapted to the Distribution Strategy of oneself in cloth strategy, and adds in parent process.In this way, should according to distribution in distributed system When creating each parent process with the order from top to bottom of the process tree of program, it is possible to according to the parent process in each parent process The Distribution Strategy of subprocess and the resource information of distributed system, automatically select suitable resource apparatus and create subprocess, make Distributed system realizes automatic distribution distributed application program.Compared with prior art, user is realizing the embodiment of the present invention When distributed system is distributed distributed application program automatically, it is not necessary to take a substantial amount of time write code again, save use The time at family, improves the efficiency of distributed application program exploitation.
The management method for the distributed application program that the embodiment of the present invention is provided, passes through default multiple general distributions Strategy so that user, can basis when developing various types of distributed application programs using general Distributed Architecture The programmed logic of each distributed application program, adds in each parent process in the process tree of each distributed application program The Distribution Strategy selected for the subprocess of each parent process so that distributed system is in the process tree according to distributed application program When order from top to bottom creates each parent process of distributed application program successively, can according to the father in each parent process into The Distribution Strategy of the subprocess of journey and the resource information of distributed system, select suitable target device to create each parent process Subprocess, makes distributed system realize automatic distribution distributed application program.With user in the prior art using general Distributed Architecture develop each distributed application program when, it is necessary to reference to distributed system resource information, when expending substantial amounts of Between go to set the distributing position of each distributed application program in a distributed system, i.e. needs take a substantial amount of time to be every A distributed application program writes code, to realize that the automatic distribution of distributed application program in a distributed system is compared, this The management method for the distributed application program that inventive embodiments are provided, can save User Exploitation distributed application program when Between, improve the efficiency that distributed application program is developed.
Further, on the basis of above-mentioned embodiment illustrated in fig. 1, distribution that Fig. 3 is provided by the embodiment of the present invention The flow chart of the management method embodiment two of application program.Wherein, the executive agent of this method can be to have in distributed system Have the communication equipment or communication module of management of process function, as shown in figure 3, after step s 102 and step S104 it Before, this method can also include the following steps：
Step S103A：The abnormality processing strategy of the subprocess of parent process is added in each parent process；Wherein, parent process The abnormality processing strategy of subprocess be used to indicate when the subprocess of parent process is abnormal, to the processing side of the subprocess of parent process Formula.
Specifically, the abnormality processing strategy of the subprocess of parent process described in the embodiment of the present invention can have it is multiple, each Abnormality processing strategy corresponds to a kind of processing mode of the subprocess to parent process, and the plurality of abnormality processing strategy is general distribution Strategy, that is to say, that to the exception of the subprocess of each parent process in the process tree of any type of distributed application program Processing mode, can be found in the plurality of abnormality processing strategy one be adapted to the abnormality processing strategy of oneself.It is specific real Now, user can be in development DDB application program, according to the programmed logic of distributed application program, i.e. distribution should With the communication frequency between the subprocess and other of parent process and parent process subprocess of each parent process in the process tree of program Rate, adds the abnormality processing strategy for the subprocess for being adapted to parent process in each parent process.If for example, the subprocess of parent process with Run independently of each other between parent process, then can be that the subprocess selection of the parent process is a kind of when the subprocess appearance of parent process is different Chang Shi, the abnormality processing strategy that the subprocess of parent process is restarted.
The embodiment of the present invention is to the representation of the abnormality processing strategy of the subprocess of above-mentioned parent process without restriction, example Such as can be the parameter that distributed system can identify, the code block corresponding to parameter is set in a distributed system, specific real Now, user can be each distribution when developing different types of distributed application program using general Distributed Architecture The subprocess of each parent process in the process tree of formula application program selects suitable abnormality processing strategy, and by each parent process The corresponding parameter of abnormality processing strategy of subprocess add each parent process, without be individually for it is each it is distributed should Abnormality processing mode is set with each process of program, i.e. need not be taken a substantial amount of time as each distributed application program Each process writes the code of processing anomalous mode, and the different of distributed application program is automatically processed to realize in distributed system Often.In this way, when distributed application program is run in a distributed system, if exception occurs in a certain subprocess, subprocess is automatic Parent process of the unexpected message to subprocess is reported, the exception by parent process according to the subprocess of the parent process added in parent process Reason strategy, determines the processing mode to the subprocess of parent process, and by distributed system according to the processing mode to parent process Subprocess is handled so that distributed system can automatically process the exception of distributed application program.
Further, when parent process has multiple subprocess, then the exception of the subprocess of the parent process described in step S103A Processing strategy is additionally operable to instruction when the subprocess of parent process is abnormal, to the processing rule of other subprocess of parent process.
Specifically, user can in development DDB application program, according to the programmed logic of distributed application program, That is, the communication between the subprocess of each parent process in the process tree of distributed application program and other subprocess of parent process Frequency, sets the processing rule of other subprocess of parent process in the abnormality processing strategy of the subprocess of each parent process.Example Such as, if being run independently of each other between the subprocess of parent process and other processes of parent process, at the subprocess of parent process Reason has no effect on the operation of other subprocess of parent process, then can select to work as in the exception policy of the subprocess of parent process and be somebody's turn to do When the subprocess of parent process is abnormal, only the subprocess of the parent process is handled, other subprocess of parent process are not done The processing rule of processing.In this way, when distributed application program is run in a distributed system, if subprocess is by the different of subprocess When being often reported to the parent process of subprocess, parent process then can be according to the exception of the subprocess of the parent process added in parent process Processing in reason strategy to other subprocess of parent process is regular, is determining to the same of the processing mode of the subprocess of parent process When, determine the processing mode to other subprocess of parent process, and by distributed system according to the processing mode to parent process Other subprocess are handled so that distributed system can automatically process the exception of distributed application program, make distributed answer Use normal program operation.
Likewise, the representation of above-mentioned processing rule may be referred to the above-mentioned table to Distribution Strategy and abnormality processing strategy Show the description of form, details are not described herein for the embodiment of the present invention.
It should be noted that the embodiment of the present invention to the execution sequence of step S103 and step S103A without limiting, can Step S103A is performed again to first carry out step S103, can also be first carried out step S103A and be performed step S103 again.
The management method for the distributed application program that the embodiment of the present invention is provided, passes through default multiple general exceptions Processing strategy so that user, can be with when developing various types of distributed application programs using general Distributed Architecture According to the programmed logic of each distributed application program, in each parent process in the process tree of each distributed application program It is incorporated as the abnormality processing strategy of the subprocess selection of each parent process so that distributed system is in operation distributed application program When, when the subprocess of distributed application program occurs abnormal, the parent process of the subprocess can be according to the subprocess of parent process Abnormality processing strategy, determine the processing rules of the processing mode of the subprocess of parent process and other subprocess of parent process, and Distributed system is transferred to be handled automatically.The general each distribution of Distributed Architecture exploitation is being used with user in the prior art During formula application program, in order to allow distributed system to automatically process the exception of distributed application program, it is necessary to expend substantial amounts of The code that time writes abnormality processing for each process of each distributed application program is compared, what the embodiment of the present invention was provided The management method of distributed application program, can further save the time of User Exploitation distributed application program, improve into The efficiency of one step distributed application program exploitation.
Further, on the basis of above-mentioned embodiment illustrated in fig. 3, distribution that Fig. 4 is provided by the embodiment of the present invention The flow chart of the management method embodiment three of application program.Wherein, the executive agent of this method can be to have in distributed system Have the communication equipment or communication module of management of process function, as shown in figure 4, after step s 102 and step S104 it Before, this method can also include the following steps：
Step S103B：Monitoring strategies are added respectively in the subprocess of each parent process and each parent process；Monitoring strategies For indicating the monitoring relation between each parent process and the subprocess of each parent process.
Specifically, when the resource apparatus residing for a certain process of distributed application program is abnormal, such as：Resource is set Standby power down etc., then process can automatically exit from running, in order to allow the parent process of distributed application program quickly to find and handle This exception of the subprocess of parent process, the embodiment of the present invention is in development DDB application program, in distributed application program Process tree in each parent process and each parent process subprocess in add monitoring strategies so that distribution application system When running the distributed application program, according to the monitoring strategies be assured that the son of each parent process and each parent process into Monitoring relation between journey.It should be noted that " in development DDB application program, in the process of distributed application program Each parent process in tree and add monitoring strategies in the subprocess of each parent process " mode can use in the prior art this The implementation that field technology personnel use, the embodiment of the present invention is to this without limiting.
Then further, after above-mentioned steps S105, this method can also include the following steps：
Step S106：According to monitoring strategies, the monitoring for establishing each parent process and the subprocess of each parent process connects, with So that the subprocess of parent process and parent process connects the heartbeat message of mutual sending cycle by monitoring.
Specifically, when distributed application program is run in a distributed system, distributed system is being distributed each father When the subprocess of each parent process is created in the target device of the subprocess of process, according to the prison added in parent process and subprocess Control strategy, establishes the monitoring connection between the subprocess of each parent process and each parent process automatically, and is establishing monitoring connection Afterwards, parent process and subprocess is notified to connect the heartbeat message of mutual sending cycle by monitoring.When parent process because exception When exiting, the periodic heartbeat that the subprocess of parent process can also be sent by not receiving parent process in preset time is believed Breath, it is voluntarily out of service after determining that parent process exits.Subprocess is caused to exit when the resource apparatus residing for subprocess is abnormal When, the periodic heartbeat information that parent process can be sent by not receiving the subprocess of parent process in preset time is and active It was found that the subprocess of parent process exits.When parent process determines that the subprocess of parent process exits, parent process can be according to parent process The abnormality processing strategy of the subprocess of the parent process of middle addition, determines the processing mode to the subprocess of parent process, and by being distributed Formula system is handled the subprocess of parent process according to the processing mode.Above-mentioned monitoring connection can be TCP connections, above-mentioned father The mode of the heartbeat message of the mutual sending cycle of the subprocess of process and parent process can be by parent process and parent process The IP address of subprocess sends mutually heartbeat message.The sending cycle of above-mentioned heartbeat message and above-mentioned preset time can bases The demand setting of user.
Optionally, first parent process in the process tree of above-mentioned distributed application program, i.e., top in process tree One process can set the monitoring strategies that monitoring connection is established with distributed system so that distributed system can monitor first The exception of a parent process, so that when any process in ensure that the process tree of distributed application program occurs any abnormal, all It can be found rapidly by parent process, and the abnormality processing strategy of the subprocess according to the parent process added in parent process, determine father The subprocess processing mode of process, and the subprocess of parent process is handled according to the processing mode by distributed system.
By the above-mentioned means, so that each parent process in the process tree of distributed application program can quickly find father The exception of the subprocess of process, and the abnormality processing strategy of the subprocess according to the parent process added in parent process, determine to father The processing mode of the subprocess of process, and the subprocess of parent process is handled according to the processing mode by distributed system, Distributed system allow automatically and timely handles the exception of distributed application program, ensure that distributed application program Normal operation.
It should be noted that the embodiment of the present invention to the execution sequence of step S103, step S103A and step S103B not It is defined.
The management method for the distributed application program that the embodiment of the present invention is provided, can further save User Exploitation The time of distributed application program, on the basis of improving the efficiency that further distributed application program is developed, makes distributed system System can automatically and timely handle the exception of distributed application program, ensure that the normal operation of distributed application program, carry The high result of calculation of distributed application program operation or the accuracy for the service for being supplied to client.
Several specific embodiments are used below, and the technical solution of embodiment of the method shown in Fig. 1, Fig. 3 and Fig. 4 is carried out in detail Describe in detail bright.
The flow chart of the management method example IV for the distributed application program that Fig. 5 is provided by the embodiment of the present invention.Its In, the executive agent of this method can be to have the function of the communication equipment of management of process or communication module in distributed system. As shown in figure 5, this method may include steps of：
Step S501：According to the logical relation of each process in distributed application program, establish distributed application program into Cheng Shu.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S101, and details are not described herein.
Step S502：Determine the son of each parent process in the process tree of distributed application program and each parent process into Journey.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S102, and details are not described herein.
Step S503：The subprocess of parent process is added in each parent process in the process tree of distributed application program Distribution Strategy.
The management method embodiment one of the distributed application program provided such as the embodiments of the present invention is described, father The Distribution Strategy of the subprocess of process can have it is multiple, each Distribution Strategy correspond to a kind of subprocess of parent process distribution be Distribution mode in system.Optionally, the Distribution Strategy of the subprocess of above-mentioned parent process can include at least following one kind：Local point Cloth strategy, integrated distribution strategy, Arbitrary distribution strategy.
Wherein, the parent process and the subprocess of parent process that Fig. 6 is provided by the embodiment of the present invention are using local Distribution Strategy Distribution schematic diagram, as shown in fig. 6, local Distribution Strategy is distributed in same target device for the subprocess of parent process and parent process In Distribution Strategy.Can be that the subprocess of the parent process selects when communicating frequent between parent process and the subprocess of parent process This Distribution Strategy is selected, so that the operational efficiency of distributed application program is higher.Exemplary, Web service application in a distributed manner Exemplified by program, since such distributed application program provides different types of service to the user, due to these different types of clothes There are data dependence, (that is, a process needs what is produced using another process in the process of running between corresponding process of being engaged in Data), then can be the local distribution of subprocess selection of each parent process in the process tree of such distributed application program Strategy.
The parent process and the subprocess of parent process that Fig. 7 is provided by the embodiment of the present invention use the distribution of integrated distribution strategy Schematic diagram, as shown in fig. 7, integrated distribution strategy is, all subprocess of parent process are distributed in the distribution in same target device Strategy.Can be that the subprocess of the parent process selects this to be distributed plan when the communication between all subprocess of parent process is frequent Slightly, so that the operational efficiency of distributed application program is higher.Exemplary, in a distributed manner exemplified by web game program, due to It is different user service processes that such distributed application program has multiple in the Same Scene of web game, and this multiple process Between communicate it is frequent, then can be such distributed application program process tree in each parent process subprocess choice set Middle Distribution Strategy.
The parent process and the subprocess of parent process that Fig. 8 is provided by the embodiment of the present invention use the distribution of Arbitrary distribution strategy Schematic diagram, as shown in figure 8, Arbitrary distribution strategy is, the subprocess of parent process and parent process is believed according to the resource of distributed system Cease the Distribution Strategy of Arbitrary distribution.When the operation between the subprocess of parent process of parent process is separate, when communicating less, Can be that the subprocess of the parent process selects when communicating less and the operation between each subprocess of parent process is also separate This Distribution Strategy is selected, so that the utilization that the resource of distributed system can be balanced.It is exemplary, data processing journey in a distributed manner Exemplified by sequence, since such distributed application program needs to be carried out at the same time numerical computations using a large amount of processes run parallel, to carry The calculating speed of high distributed application program, wherein, each equal independent operating of process, then can be such Distributed Application journey The subprocess selection Arbitrary distribution strategy of each parent process in the process tree of sequence, in terms of maximumlly improving distributed system The utilization rate of resource is calculated, shortens the run time of distributed application program.
When for parent process subprocess selection Distribution Strategy for local Distribution Strategy or integrated distribution strategy when, further Optionally, relaxation Distribution Strategy can also be added on the basis of local Distribution Strategy or integrated distribution strategy.The slackness point Cloth strategy is used to indicate when being distributed the target device of subprocess of parent process overload, will can not create on the target device also The process built is distributed on other resource apparatus, to balance the load of resource apparatus.Exemplary, pine is added with integrated distribution strategy Exemplified by relaxation Distribution Strategy, parent process and the subprocess of parent process that Fig. 9 is provided by the embodiment of the present invention use integrated distribution plan The distribution schematic diagram for the Distribution Strategy that relaxes slightly, as shown in figure 9, when all subprocess should be distributed in same resource apparatus, If the resource apparatus load too high, the subprocess not being created also during above-mentioned subprocess is created can be distributed on same machine On other resource apparatus of frame；If all resource apparatus in the corresponding rack of the resource apparatus overload, it is not created also Subprocess can be distributed in the resource apparatus in other racks.
It is further alternative, the embodiment of the present invention to the representation of the Distribution Strategy of the subprocess of above-mentioned parent process not into Row limits, such as can be the parameter that distributed system can identify, the code block corresponding to parameter is arranged on distributed system In.When the parameter that Distribution Strategy can be identified using distributed system, following example is specifically may be referred to, wherein, local point The corresponding parameter of cloth strategy is：Strategy is distributed to be local, and the corresponding parameter of integrated distribution strategy is：Strategy is concentration Distribution, the corresponding parameter of Arbitrary distribution strategy are：Strategy is Arbitrary distribution, then the parameter set in distributed system corresponds to Code block specifically can be as follows：
Step S504：The subprocess of parent process is added in each parent process in the process tree of distributed application program Abnormality processing strategy.
The management method embodiment two of the distributed application program provided such as the embodiments of the present invention is described, father The abnormality processing strategy of the subprocess of process can have multiple, and each abnormality processing strategy corresponds to a kind of subprocess to parent process Processing mode.Optionally, the abnormality processing strategy of the subprocess of above-mentioned parent process can include at least following one kind：Ignore different Often processing strategy, reboot process strategy, exit processing strategy.
Wherein, it is the processing strategy for ignoring the unexpected message that the subprocess of parent process reports to ignore abnormality processing strategy.When Can be the parent process when influence of the abnormal operation to the subprocess of parent process of the subprocess of parent process can be ignored Subprocess select this abnormality processing strategy.It is exemplary, with calculate it is abnormal (such as：Divisor is zero, and computing is overflowed etc.) exemplified by, Due to calculate it is abnormal usually caused in itself by distributed application program, even if the subprocess of parent process, which is restarted, can not also avoid out Now calculate abnormal, and the calculating can't cause the subprocess of parent process not run extremely, then can be parent process son into Abnormality processing strategy is ignored in journey selection.
Reboot process strategy is the processing strategy for restarting the subprocess for the parent process for reporting unexpected message.Work as parent process The abnormal operation to the subprocess of the parent process of subprocess have an impact, but can be recovered by restarting the son of the parent process into Can be that the subprocess of the parent process selects this abnormality processing strategy during the operation of journey.Wherein, it is above-mentioned restart parent process son into Journey for distributed system again according to the Distribution Strategy of the subprocess of the parent process added in parent process, with reference to distributed system Resource information, redefines the target device of the subprocess of distribution parent process, and creates and be somebody's turn to do on the target device redefined The subprocess of parent process.Exemplary, there is hardware fault (for example, resource with the resource apparatus residing for the subprocess of parent process The low memory of equipment, the disk failures of resource apparatus etc.), due to the hardware that the exception of the subprocess of parent process is resource apparatus Caused by failure, can select to re-create the subprocess on new resource apparatus can be to avoid there is the exception again When, then can be that the subprocess of parent process selects reboot process strategy.
It is the processing strategy that the subprocess of the parent process of unexpected message will be reported out of service to exit processing strategy.Work as father The abnormal influence caused by the subprocess of parent process of the subprocess of process, can not also be recovered by the subprocess for restarting parent process When, can be that the subprocess of the parent process selects this abnormality processing strategy.Exemplary, it is distributed with the subprocess of parent process Exemplified by system compulsory withdrawal, caused by such is distributed system extremely, even if restarting the subprocess of the parent process, It can not avoid again by distributed system compulsory withdrawal, then can be that processing strategy is exited in the subprocess selection of the parent process.
It is further alternative, when parent process can not handle the unexpected message of subprocess active reporting of parent process, father into Journey can be with operation suspension, and the unexpected message is reported to the parent process of parent process, is handled by the parent process of the parent process.
It is when parent process has multiple subprocess, then further alternative, the abnormality processing plan of the subprocess of above-mentioned parent process Slightly also include at least following one kind：Individually processing rule, all handles rule.
Wherein, individually processing rule is the processing rule not processed to other subprocess of parent process.When to parent process Some subprocess being abnormal when being handled, during the operation for other subprocess for having no effect on parent process, Ke Yiwei Other subprocess of parent process select this processing rule.
All processing rule is that the processing handled using same abnormality processing strategy whole subprocess of parent process is advised Then.When some subprocess being abnormal to parent process is handled, other subprocess of parent process can not be just Can be that other subprocess of parent process select this processing rule often during operation.
Step S505：Each parent process in the process tree of distributed application program and in the subprocess of each parent process Monitoring strategies are added respectively.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S103B, and details are not described herein.
It should be noted that the embodiment of the present invention to the execution sequence of step S503, step S504 and step S505 not into Row limits.
Step S506：Order of the distributed system according to the process tree of distributed application program from top to bottom, creates first A parent process.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S104, and details are not described herein.
Step S507：Distributed system is according to the Distribution Strategy of the subprocess of first parent process and the money of distributed system Source information, determines the target device of the subprocess of first parent process of distribution.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S104, and details are not described herein.
Step S508：Distributed system creates first father in the target device of subprocess of first parent process is distributed The subprocess of process.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S105, and details are not described herein.
Step S509：Distributed system is established first parent process and is connected with the monitoring of the subprocess of first parent process.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S106, and details are not described herein.
Step S510：Distributed system notifies the subprocess of first parent process and first parent process to be connected by monitoring The heartbeat message of mutual sending cycle.
Wherein, the associated description of this step may refer to the description of above-mentioned steps S106, and details are not described herein.
Step S511：Order of the distributed system according to the process tree of distributed application program from top to bottom, continues to use step The mode of S507 to step S510, creates each parent process in the process tree of distributed application program and each parent process successively Subprocess, until all processes in the process tree of distributed application program have all been created in a distributed system Finish.
So far, distributed system realizes automatic distribution distributed application program.
The management method for the distributed application program that the embodiment of the present invention is provided, passes through default multiple general distributions Strategy so that user, can basis when developing various types of distributed application programs using general Distributed Architecture The programmed logic of each distributed application program, adds in each parent process in the process tree of each distributed application program The Distribution Strategy selected for the subprocess of each parent process so that distributed system is in the process tree according to distributed application program When order from top to bottom creates each parent process of distributed application program successively, can according to the father in each parent process into The Distribution Strategy of the subprocess of journey and the resource information of distributed system, select suitable target device to create each parent process Subprocess, makes distributed system realize automatic distribution distributed application program.With user in the prior art using general Distributed Architecture develop each distributed application program when, it is necessary to reference to distributed system resource information, when expending substantial amounts of Between go to set the distributing position of each distributed application program in a distributed system, i.e. needs take a substantial amount of time to be every A distributed application program writes code, to realize that the automatic distribution of distributed application program in a distributed system is compared, this The management method for the distributed application program that inventive embodiments are provided, can save User Exploitation distributed application program when Between, improve the efficiency that distributed application program is developed.
Further, on the basis of above-mentioned embodiment illustrated in fig. 5, distribution that Figure 10 is provided by the embodiment of the present invention The flow chart of the management method embodiment five of application program, as shown in Figure 10, the application scenarios of this method are to work as distributed system The subprocess of a certain parent process in the process tree of a certain distributed application program of middle operation is abnormal, and active reporting is to father The management method of distributed application program during process, wherein, the executive agent of this method can be to have in distributed system The communication equipment or communication module of management of process function, then can also include the following steps after step S512：
Step S1001：Subprocess actively sends unexpected message by monitoring connection to the parent process of the subprocess.
Step S1002：Parent process adds after the unexpected message that the subprocess for receiving parent process is sent according in parent process The abnormality processing strategy of the subprocess of the parent process entered, determines the processing mode to the subprocess of parent process, and to parent process Other subprocess processing rule.
Step S1003：Parent process by the processing mode of the definite subprocess to parent process, and to parent process other The processing rule of subprocess is sent to distributed system.
Step S1004：Distributed system according to the processing mode of the subprocess of the parent process of reception to the son of parent process into Cheng Jinhang processing, while other subprocess of parent process are carried out according to the processing rule of other subprocess of the parent process of reception Processing.
So far, distributed system realizes any exception for automatically processing distributed application program active reporting.
The management method for the distributed application program that the embodiment of the present invention is provided, passes through default multiple general exceptions Processing strategy so that user, can be with when developing various types of distributed application programs using general Distributed Architecture According to the programmed logic of each distributed application program, in each parent process in the process tree of each distributed application program It is incorporated as the abnormality processing strategy of the subprocess selection of each parent process so that distributed system is in operation distributed application program When, when the subprocess of distributed application program occurs abnormal, the parent process of the subprocess can be according to the subprocess of parent process Abnormality processing strategy, determine the processing rules of the processing mode of the subprocess of parent process and other subprocess of parent process, and Distributed system is transferred to be handled automatically.The general each distribution of Distributed Architecture exploitation is being used with user in the prior art During formula application program, in order to allow distributed system to automatically process the exception of distributed application program, it is necessary to expend substantial amounts of The code that time writes abnormality processing for each process of each distributed application program is compared, what the embodiment of the present invention was provided The management method of distributed application program, can further save the time of User Exploitation distributed application program, improve into The efficiency of one step distributed application program exploitation.
Further, on the basis of above-mentioned Fig. 5 and embodiment illustrated in fig. 10, Figure 11 is provided by the embodiment of the present invention The flow chart of the management method embodiment six of distributed application program, as shown in figure 11, the application scenarios of this method are when distribution A certain parent process in the process tree of a certain distributed application program run in formula system does not receive this in preset time The management method of distributed application program during the periodic heartbeat message that the subprocess of parent process is sent, wherein, this method Executive agent can be distributed system in there is the communication equipment of management of process or communication module, then in step It can also include the following steps after S512：
Step S1101：The periodic heart that parent process is sent by not receiving the subprocess of parent process in preset time Hop-information, confirms that the subprocess of parent process exits.
Step S1102：Parent process is after the subprocess for confirming parent process exits, according to the parent process added in parent process The abnormality processing strategy of subprocess, determines the processing mode to the subprocess of parent process, and other subprocess to parent process Processing rule.
Step S1103：Parent process by the processing mode of the definite subprocess to parent process, and to parent process other The processing rule of subprocess is sent to distributed system.
Step S1104：Distributed system according to the processing mode of the subprocess of the parent process of reception to the son of parent process into Cheng Jinhang processing, while other subprocess of parent process are carried out according to the processing rule of other subprocess of the parent process of reception Processing.
So far, distributed system realizes any exception for automatically processing distributed application program.
The management method for the distributed application program that the embodiment of the present invention is provided, can further save User Exploitation The time of distributed application program, on the basis of improving the efficiency that further distributed application program is developed, makes distributed system System can automatically and timely handle the exception of distributed application program, ensure that the normal operation of distributed application program, carry The high result of calculation of distributed application program operation or the accuracy for the service for being supplied to client.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above-mentioned each method embodiment can lead to The relevant hardware of programmed instruction is crossed to complete.Foregoing program can be stored in a computer read/write memory medium.The journey Sequence upon execution, execution the step of including above-mentioned each method embodiment；And foregoing storage medium includes：ROM, RAM, magnetic disc or Person's CD etc. is various can be with the medium of store program codes.
The structure diagram for the distributed system that Figure 12 is provided by the embodiment of the present invention.As shown in figure 12, the distribution System includes：Multiple resource apparatus 1201 and distributed platform 1202.
Wherein, distributed platform 1202 is used for the pipe for performing the distributed application program as described in claim 1-8 is any Reason method；Multiple resource apparatus 1201 are used to run distributed application program.
Specifically, above-mentioned distributed platform can be distributed system in have the function of the communication equipment of management of process or Communication module, above-mentioned resource apparatus can be computer, server, virtual machine etc..Distributed system provided in an embodiment of the present invention System, can be used for the technical solution for performing either method embodiment shown in Fig. 1, Fig. 3, Fig. 4, Fig. 5, Figure 10 and Figure 11, it is realized Principle is similar with technique effect, and details are not described herein again.
Finally it should be noted that：The above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent Pipe is described in detail the present invention with reference to foregoing embodiments, it will be understood by those of ordinary skill in the art that：Its according to Can so modify to the technical solution described in foregoing embodiments, either to which part or all technical characteristic into Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology The scope of scheme.

Claims (8)

  1. A kind of 1. management method of distributed application program, it is characterised in that including：
    According to the logical relation of each process in distributed application program, the process tree of the distributed application program is established；
    Determine the subprocess of each parent process and each parent process in the process tree；
    The Distribution Strategy of the subprocess of the parent process is added in each parent process；Wherein, the parent process son into The Distribution Strategy of journey is for indicating the distribution mode of the subprocess of the parent process in a distributed system；
    Wherein, the Distribution Strategy of the subprocess of the parent process includes at least following one kind：Local Distribution Strategy, integrated distribution plan Slightly, Arbitrary distribution strategy；
    The local Distribution Strategy is that the subprocess of the parent process and the parent process is distributed in point in same target device Cloth strategy；
    The integrated distribution strategy is that all subprocess of the parent process are distributed in the Distribution Strategy in same target device；
    The Arbitrary distribution strategy is, the subprocess of the parent process and the parent process is according to the resource of the distributed system The Distribution Strategy of information Arbitrary distribution；
    Each parent process is created according to the order of the process tree from top to bottom successively, and according to each parent process The resource information of the Distribution Strategy of subprocess and the distributed system, determines the mesh of the subprocess of each parent process of distribution Marking device；
    The subprocess of each parent process is created in the target device of the subprocess of each parent process of distribution.
  2. 2. the management method of distributed application program according to claim 1, it is characterised in that the method further includes：
    The abnormality processing strategy of the subprocess of the parent process is added in each parent process；Wherein, the parent process The abnormality processing strategy of subprocess is used to indicate when the subprocess of the parent process is abnormal, to the subprocess of the parent process Processing mode.
  3. 3. the management method of distributed application program according to claim 2, it is characterised in that the parent process son into The abnormality processing strategy of journey includes at least following one kind：Ignore abnormality processing strategy, reboot process strategy, exits processing strategy；
    Wherein, the abnormality processing strategy of ignoring is the processing plan for ignoring the unexpected message that the subprocess of the parent process reports Slightly；
    The reboot process strategy is the processing strategy for restarting the subprocess for the parent process for reporting unexpected message；
    The processing strategy that exits is that will report the subprocess of the parent process of unexpected message processing strategy out of service.
  4. 4. the management method of the distributed application program according to Claims 2 or 3, it is characterised in that the parent process The abnormality processing strategy of subprocess is additionally operable to instruction when the subprocess of the parent process is abnormal, to other sons of the parent process The processing rule of process.
  5. 5. the management method of distributed application program according to claim 4, it is characterised in that the parent process son into The abnormality processing strategy of journey also includes at least following one kind：Individually processing rule, all handles rule；
    Wherein, the individually processing rule is the processing rule not processed to other subprocess of the parent process；It is described complete Portion's processing rule is the processing rule handled using same abnormality processing strategy whole subprocess of the parent process.
  6. 6. according to the management method of claim 1-3 any one of them distributed application programs, it is characterised in that the method Further include：
    Monitoring strategies are added respectively in the subprocess of each parent process and each parent process；The monitoring strategies are used Monitoring relation between each parent process of instruction and the subprocess of each parent process.
  7. 7. the management method of distributed application program according to claim 6, it is characterised in that in each father of distribution When the subprocess of each parent process is created in the target device of the subprocess of process, the method further includes：
    According to the monitoring strategies, the monitoring for establishing each parent process and the subprocess of each parent process connects, with So that the subprocess of the parent process and the parent process connects the heartbeat message of mutual sending cycle by the monitoring.
  8. A kind of 8. distributed system, it is characterised in that including：Multiple resource apparatus and distributed platform；Wherein, the distribution Platform is used for the management method for performing the distributed application program as described in claim 1-7 is any；The multiple resource apparatus For running the distributed application program.
CN201510184249.7A 2015-04-17 2015-04-17 The management method and distributed system of distributed application program Active CN104850394B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510184249.7A CN104850394B (en) 2015-04-17 2015-04-17 The management method and distributed system of distributed application program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510184249.7A CN104850394B (en) 2015-04-17 2015-04-17 The management method and distributed system of distributed application program

Publications (2)

Publication Number Publication Date
CN104850394A CN104850394A (en) 2015-08-19
CN104850394B true CN104850394B (en) 2018-04-17

Family

ID=53850057

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510184249.7A Active CN104850394B (en) 2015-04-17 2015-04-17 The management method and distributed system of distributed application program

Country Status (1)

Country Link
CN (1) CN104850394B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105653371A (en) * 2015-12-29 2016-06-08 曙光云计算技术有限公司 Resource pool management method and device under PaaS environment
CN106020984B (en) * 2016-05-20 2020-01-31 青岛海信移动通信技术股份有限公司 Method and device for creating process in electronic equipment
CN107229504A (en) * 2017-05-12 2017-10-03 广州接入信息科技有限公司 Program distribution operation method, apparatus and system
CN109388489A (en) * 2017-08-03 2019-02-26 成都蓝盾网信科技有限公司 A kind of technological frame of the fault-tolerant high stable of height of more subprocess and process signals processing based on single guiding systems
CN109408158B (en) * 2018-11-06 2022-11-18 恒生电子股份有限公司 Method and device for quitting child process along with parent process, storage medium and electronic equipment
CN113656175B (en) * 2021-08-18 2022-07-08 北京百度网讯科技有限公司 Method and equipment for training model based on distributed system
CN114996013B (en) * 2022-06-09 2023-06-02 河南省肿瘤医院 Microwave treatment equipment based on it is embedded

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101788915A (en) * 2010-02-05 2010-07-28 北京工业大学 White list updating method based on trusted process tree
CN101944167A (en) * 2010-09-29 2011-01-12 中国科学院计算技术研究所 Method and system for identifying malicious program

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5640432B2 (en) * 2010-03-31 2014-12-17 富士通株式会社 Distributed processing apparatus, distributed processing program, and distributed processing method
JP5899144B2 (en) * 2013-03-22 2016-04-06 ヤフー株式会社 Advertisement distribution apparatus, advertisement distribution method, and advertisement distribution program
JP6364727B2 (en) * 2013-09-24 2018-08-01 日本電気株式会社 Information processing system, distributed processing method, and program

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101788915A (en) * 2010-02-05 2010-07-28 北京工业大学 White list updating method based on trusted process tree
CN101944167A (en) * 2010-09-29 2011-01-12 中国科学院计算技术研究所 Method and system for identifying malicious program

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于分布式环境的子进程监控软件设计与实现;张虎等;《软件天地》;20130415;第32卷(第7期);第1-4页 *

Also Published As

Publication number Publication date
CN104850394A (en) 2015-08-19

Similar Documents

Publication Publication Date Title
CN104850394B (en) The management method and distributed system of distributed application program
AU2011299337B2 (en) Controlled automatic healing of data-center services
US11169854B2 (en) Node eligibility determinations
CN105005532B (en) A kind of system and method for automatic test application programming interfaces stability
CN108810115B (en) Load balancing method and device suitable for distributed database and server
US9154366B1 (en) Server maintenance modeling in cloud computing
Huang et al. Cap3: A cloud auto-provisioning framework for parallel processing using on-demand and spot instances
Konstantinou et al. Tiramola: elastic nosql provisioning through a cloud management platform
CN107526645A (en) A kind of communication optimization method and system
CN109614227A (en) Task resource concocting method, device, electronic equipment and computer-readable medium
CN107153573A (en) Distributed task scheduling treating method and apparatus
CN106101213A (en) Information-distribution type storage method
CN113157379A (en) Cluster node resource scheduling method and device
Emeakaroha et al. Managing and optimizing bioinformatics workflows for data analysis in clouds
CN106059940A (en) Flow control method and device
CN115225642A (en) Elastic load balancing method and system of super-fusion system
Li et al. MapReduce task scheduling in heterogeneous geo-distributed data centers
CN105827744A (en) Data processing method of cloud storage platform
CN106254452A (en) The big data access method of medical treatment under cloud platform
CN109885384A (en) Task concurrency optimization method, apparatus, computer equipment and storage medium
CN109257256A (en) Apparatus monitoring method, device, computer equipment and storage medium
CN105471986B (en) A kind of Constructing data center Scale Revenue Ratio method and device
CN109240914A (en) The method for managing and monitoring and terminal device of safety test task
CN106302656A (en) The Medical Data processing method of cloud storage platform
CN107872480A (en) Big data cluster data balancing method and apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
EXSB Decision made by sipo to initiate substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant