CN107783843A - A kind of processing method and processing device of cloud data - Google Patents

A kind of processing method and processing device of cloud data Download PDF

Info

Publication number
CN107783843A
CN107783843A CN201610799726.5A CN201610799726A CN107783843A CN 107783843 A CN107783843 A CN 107783843A CN 201610799726 A CN201610799726 A CN 201610799726A CN 107783843 A CN107783843 A CN 107783843A
Authority
CN
China
Prior art keywords
cloud
data
processing
alarming
task
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.)
Granted
Application number
CN201610799726.5A
Other languages
Chinese (zh)
Other versions
CN107783843B (en
Inventor
孙翠锋
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.)
China Mobile Communications Group Co Ltd
China Mobile Communications Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
China Mobile Communications Co Ltd
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 China Mobile Communications Group Co Ltd, China Mobile Communications Co Ltd filed Critical China Mobile Communications Group Co Ltd
Priority to CN201610799726.5A priority Critical patent/CN107783843B/en
Publication of CN107783843A publication Critical patent/CN107783843A/en
Application granted granted Critical
Publication of CN107783843B publication Critical patent/CN107783843B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/52—Program synchronisation; Mutual exclusion, e.g. by means of semaphores
    • G06F9/524—Deadlock detection or avoidance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5083—Techniques for rebalancing the load in a distributed system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/546—Message passing systems or structures, e.g. queues
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/54—Indexing scheme relating to G06F9/54
    • G06F2209/548—Queue

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Debugging And Monitoring (AREA)
  • Alarm Systems (AREA)

Abstract

Embodiments of the invention provide a kind of processing method and processing device of cloud data, and method includes：Obtain the cloud associated data of waiting task；Waiting task cloud associated data first lock processing it is unsuccessful, and the cloud associated data be cloud conflict confirm data when, the waiting task is dispatched to clash handle queue；Otherwise, locking again is carried out to the cloud associated data to handle；When the cloud associated data locks the unsuccessful number of processing more than a limit value again, cloud collision alert data are exported.Conflict Tasks are dispatched to single clash handle queue and are uniformly processed by the solution of the present invention, not being re-used as under cloud environment for task, concurrent scheduling are carried out, so as to avoid conflicting.

Description

A kind of processing method and processing device of cloud data
Technical field
The present invention relates to cloud technical field of data processing, particularly relates to a kind of processing method and processing device of cloud data.
Background technology
In the prior art, in cloud data processing circumstance, scheduling mechanism typically has following several：
Preemption scheduling mechanism：In cloud environment, in the treatment progress free time of each task processing node, active inquiry is appointed Waiting task in business processing queue, the disposal right of certain waiting task is obtained according to certain rule, and the task to seizing is entered Row is locked from the mechanism reprocessed by other nodes.
Distribution formula scheduling mechanism：In cloud environment, waiting task in queue is pressed busy degree by task scheduling process Each processing node is assigned to certain rule, each processing node is handled the task of distribution.
Autonomous task：The data of task processing are not related to the change of other outer data of task, delta data and task Data itself is relevant, and this task is referred to as autonomous task, and conflict is not present in autonomous task.
The task of Non-Self-Governing：The data of task processing except in itself mutually outside the Pass, further related to task change task it is outer other Data, this task be Non-Self-Governing task.
Cloud association process data：Related to Non-Self-Governing task does not include the data for needing to change in data in task, this The intermediate data or result data of incidence relation, generally task processing be present in the data of kind data and task itself.Such as telecommunications Subscriber's account in industry in ticket processing task is exactly cloud associated data.
Cloud collision alert data (CWD)：Occur locking the alarm exported after failure when cloud computing handles the task of Non-Self-Governing Data.
Cloud colliding data (CD)：Collision alert data caused by cloud computing, cloud is generated after the processing of alarm process confirms Colliding data records, and cloud colliding data includes data major key, data type, recently nearest conflict number, conflict discovery time, shape The information such as state (cloud colliding data state is " cloud conflict ", and state can update when being changed into cloud conflict and confirming data).
Cloud colliding data effective time (CDVT)：Cloud colliding data is only effective within a period of time that conflict occurs, when super Conflict may disappear after a period of time, and this section of cloud conflict occurs to the time to disappear being referred to as cloud colliding data effective time, The time is assessed and can be set as a constants VT based on experience value in system.In the cloud of cloud colliding data effective time Colliding data is referred to as effective cloud colliding data, is otherwise referred to as failure cloud colliding data.
Cloud conflict confirms data (CCD)：When conflict number of the cloud colliding data in CDVT exceedes setting value MC, the cloud Colliding data is referred to as cloud conflict and confirms data, and MC, which is referred to as conflicting, confirms threshold values.Cloud conflict confirms that data are a kind of cloud colliding datas, It is stored in cloud colliding data, is simply identified with single state.
In existing task processor system, based on the cloud processing of message, file, data base table recording as process object In system, in order to reach the equilibrium assignment that task throughout manages node, mainly there are two kinds of task scheduling machines of preemptive type and distributed System.
Both scheduling mechanisms, distribute processing task when only can be according to task rule for distribution, the self-contained number of task Task processing is carried out according to other data (data of other data without other tasks outside non-task) of, task association, It will not go to associate the data that other tasks include and be handled.
For example：The ticket wholesale price processing of telecommunications industry, be when handling CDR file according to the record data of CDR file, Subscriber data, the history consumption of user, rate situation carry out wholesale price processing, and the different task between two CDR files exists Incidence relation is not accounted for during task scheduling.
Under existing cloud environment in task handling process, when exporting cloud collision alert data, multiple rear outputting alarm Afterwards, generally rely on and manually extract the task of association, and task is diverted to by hand individually non-concurrent Conflict Tasks team Column processing, avoid occurring lock conflict in processing procedure again.
Moreover, in order to ensure to the correctness of cloud association process data manipulation under cloud computing mode, it is all to these data Adjustment must first lock.But when multiple nodes all lock to same data, conflict will be produced and lock waits, same data are few The lock of amount is waited to overall process capacity very little, but works as the processing for being related to some cloud association process data manipulation with for the moment In section in multiple tasks when there are a lot, due to a large amount of lock waits treatment effeciency will be caused drastically to decline, or even obstruction is whole The operation of individual system.
The content of the invention
The invention provides a kind of processing method and processing device of cloud data, Conflict Tasks are dispatched to single clash handle Queue is uniformly processed, not being re-used as under cloud environment for task, concurrent scheduling is carried out, so as to avoid conflicting.
In order to solve the above technical problems, embodiments of the invention provide following scheme：
A kind of processing method of cloud data, including：
Obtain the cloud associated data of waiting task；
It is unsuccessful that processing is locked first in the cloud associated data of waiting task, and the cloud associated data is that cloud conflict is true When recognizing data, the waiting task is dispatched to clash handle queue；Otherwise, the cloud associated data is carried out again at locking Reason；
When the cloud associated data locks the unsuccessful number of processing more than a limit value again, cloud collision alert number is exported According to.
Wherein, the cloud associated data is carried out including the step of locking processing again：
After waiting a prefixed time interval, locking again is carried out to the cloud associated data and handled.
Wherein, the processing method of above-mentioned cloud data also includes：Handled in the cloud associated data locking first of waiting task During success, the waiting task is normally handled.
Wherein, the processing method of above-mentioned cloud data also includes：When the cloud associated data locks handle successfully again, to institute Waiting task is stated normally to be handled.
Wherein, the waiting task is carried out also including after normally handling：Discharge the lock.
Wherein, also include after exporting cloud collision alert data：The cloud collision alert data of output are handled, led The cloud conflict of processing obstruction is caused to confirm data.
Wherein, the cloud collision alert data of output are handled, the cloud conflict for obtaining causing processing to be blocked confirms data The step of include：
The alarming assignment included in the cloud collision alert data of output is put into an alarming processing task queue；
An alarming assignment is taken out from alarming processing task queue, if getting, the alarming assignment for judging currently to get is It is no cloud colliding data to be present, if so, updating the nearest conflict number of the cloud colliding data, conflict discovery time recently；
Judge whether the cloud colliding data is that cloud conflict confirms data, if so, the alarming assignment currently got is dispatched Deleted to independent clash handle queue, and by the alarming assignment currently got from alarming processing task queue；If it is not, judge most Whether nearly conflict number, which exceedes conflict, confirms threshold value, if so, the state of renewal cloud colliding data confirms data for cloud conflict；If It is no, alarming assignment is turned back to common task scheduling queue again and handled.
Wherein, the processing method of cloud data also includes：
If not getting alarming assignment from alarming processing task queue, after being spaced a preset time period, continue from announcement Alarming assignment is taken in alert processing task queue.
Wherein, if after getting alarming assignment from alarming processing task queue, the alarming assignment for judging currently to get is It is no exist also include before cloud colliding data：
Present system time is read, and deletes failure cloud colliding data corresponding to the alarming assignment got.
Wherein, the processing method of cloud data also includes：
Cloud colliding data is not present in the alarming assignment for judging currently to get, and generates a cloud colliding data, and alarm is appointed Business turns back to common task scheduling queue and handled again.
Wherein, the state for updating cloud colliding data is also to include after cloud conflict confirms data：
The alarming assignment currently got is dispatched to independent clash handle queue, and by the alarming assignment currently got from announcement Deleted in alert processing task queue.
Wherein, the alarming assignment currently got is dispatched to independent clash handle queue, and the alarm currently got is appointed Business also includes after being deleted from alarming processing task queue：
Judge whether also have alarming assignment in alarming processing task queue, if so, then being taken from alarming processing task queue Go out alarming assignment to be judged, if not having, terminate.
Embodiments of the invention also provide a kind of processing unit of cloud data, including：
Module is obtained, for obtaining the cloud associated data of waiting task；
First processing module, for unsuccessful, and the cloud that locks processing first in the cloud associated data of waiting task When associated data is that cloud conflict confirms data, the waiting task is dispatched to clash handle queue；Otherwise, to the cloud incidence number Handled according to locking again is carried out；
Output module, it is defeated for when the cloud associated data locks the unsuccessful number of processing more than a limit value again Go out cloud collision alert data.
Wherein, the processing unit of cloud data also includes：
Second processing module, for handling the cloud collision alert data of output, obtain the cloud for causing processing to be blocked Conflict confirms data.
Wherein, the Second processing module is specifically used for：The alarming assignment that will be included in the cloud collision alert data of output It is put into an alarming processing task queue；An alarming assignment is taken out from alarming processing task queue, if getting, judgement currently takes The alarming assignment arrived whether there is cloud colliding data, if so, updating the nearest conflict number of the cloud colliding data, recently conflict Discovery time；
Judge whether the cloud colliding data is that cloud conflict confirms data, if so, the alarming assignment currently got is dispatched Deleted to independent clash handle queue, and by the alarming assignment currently got from alarming processing task queue；If it is not, judge most Whether nearly conflict number, which exceedes conflict, confirms threshold value, if so, the state of renewal cloud colliding data confirms data for cloud conflict；If It is no, alarming assignment is turned back to common task scheduling queue again and handled.
The such scheme of the present invention comprises at least following beneficial effect：
The such scheme of the present invention, by the cloud associated data for obtaining waiting task；Associated in the cloud of waiting task Data lock first processing it is unsuccessful, and the cloud associated data be cloud conflict confirm data when, the waiting task is dispatched To clash handle queue；Otherwise, locking again is carried out to the cloud associated data to handle；In the cloud associated data, locking is handled again When unsuccessful number is more than a limit value, cloud collision alert data are exported.Conflict Tasks are dispatched to single clash handle Queue is uniformly processed, and not being re-used as under cloud environment for task carries out concurrent scheduling so as to avoid conflicting.
Brief description of the drawings
Fig. 1 is the flow chart of the processing method of the cloud data of the present invention；
Fig. 2 is a particular flow sheet of the processing method of the cloud data of the present invention；
Fig. 3 is the specific implementation flow chart of step 301 in the processing method of the cloud data of the invention shown in Fig. 2.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although the disclosure is shown in accompanying drawing Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here Limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure Completely it is communicated to those skilled in the art.
For existing task handling process, embodiments of the invention are carried out for the first time in processing task to cloud associated data When locking failure, judge whether cloud associated data is that cloud conflict confirms data, if it is by the task scheduling to single punching Prominent processing queue is handled；Do not determine whether to confirm data for cloud conflict in data processing entrance, avoid reducing entirely The disposal ability of system, because accounting is considerably less on the whole for colliding data, sentence again after only locking failure in first time Disconnected neither to influence overall treatment efficiency, and can is found in time when real colliding data occurs.
As shown in figure 1, embodiments of the invention provide a kind of processing method of cloud data, including：
Step 11, the cloud associated data of waiting task is obtained；
Step 12, lock that processing is unsuccessful, and the cloud associated data is first in the cloud associated data of waiting task When cloud conflict confirms data, the waiting task is dispatched to clash handle queue；Otherwise, the cloud associated data is carried out again Locking is handled；
Step 13, when the cloud associated data locks the unsuccessful number of processing more than a limit value again, output cloud punching Prominent alarm data.
The above embodiment of the present invention, by the cloud associated data for obtaining waiting task；Closed in the cloud of waiting task Connection data lock first processing it is unsuccessful, and the cloud associated data be cloud conflict confirm data when, the waiting task is adjusted Spend clash handle queue；Otherwise, locking again is carried out to the cloud associated data to handle；In the cloud associated data again at locking When managing unsuccessful number more than a limit value, cloud collision alert data are exported.Conflict Tasks are dispatched at single conflict Reason queue is uniformly processed, and not being re-used as under cloud environment for task carries out concurrent scheduling so as to avoid conflicting.
In the specific embodiment of the present invention, including on the basis of above-mentioned steps 11-13, in step 12, the cloud is associated Data include the step of locking processing again：After waiting a prefixed time interval, the cloud associated data is carried out again to add Lock processing.
Further, above-described embodiment also includes：When the cloud associated data of waiting task locks handle successfully first, The waiting task is normally handled, discharges the lock.
Further, the processing method of above-mentioned cloud data also includes：When the cloud associated data locks handle successfully again, The waiting task is normally handled, discharges the lock.
In the specific embodiment of the present invention, including on the basis of above-mentioned steps 11-13, after exporting cloud collision alert data Also include：
Step 14, the cloud collision alert data of output are handled, the cloud conflict for obtaining causing processing to be blocked confirms number According to.
Specifically, the step 14 includes：
Step 141, the alarming assignment included in the cloud collision alert data of output is put into an alarming processing task queue；
Step 142, an alarming assignment is taken out from alarming processing task queue, if getting, judges the announcement currently got Alert task whether there is cloud colliding data, if so, updating the nearest conflict number of cloud colliding data, conflict discovery time recently；
Step 143, judge whether the cloud colliding data is that cloud conflict confirms data, if so, the alarm that will currently get Task scheduling is deleted to independent clash handle queue, and by the alarming assignment currently got from alarming processing task queue；If It is no, judge whether nearest conflict number exceedes conflict and confirm threshold value, if so, the state of renewal cloud colliding data confirms for cloud conflict Data；Handled if it is not, alarming assignment to be turned back to common task scheduling queue again.
Further, the step 14 also includes：If not getting alarming assignment from alarming processing task queue, it is spaced After one preset time period, continue to take alarming assignment from alarming processing task queue.
If after getting alarming assignment from alarming processing task queue, the alarming assignment for judging currently to get whether there is Also include before cloud colliding data：Present system time is read, and deletes failure cloud number of collisions corresponding to the alarming assignment got According to.
Further, the step 14 also includes：Cloud colliding data, generation one is not present in the alarming assignment for judging currently to get Bar cloud colliding data, and alarming assignment is turned back into common task scheduling queue again and handled.
Wherein, in step 143, the state of renewal cloud colliding data is also to include after cloud conflict confirms data：
The alarming assignment currently got is dispatched to independent clash handle queue, and by the alarming assignment currently got from announcement Deleted in alert processing task queue；
And determine whether also have alarming assignment in alarming processing task queue, if so, then from alarming processing task Alarming assignment is taken out in queue to be judged, if not having, is terminated.
Shown in Fig. 2, the idiographic flow of above method embodiment is as follows：
Step 21:Obtain waiting task；
Step 22:The cloud associated data of waiting task is locked；
Step 23:Judge to lock and whether succeed, if the determination result is YES, then perform step 24, otherwise perform step 25；
Step 24:Waiting task is normally handled, release lock after having handled；
Step 25:Judge whether cloud associated data is that cloud conflict confirms data, if the determination result is YES, then performs step 26, otherwise perform step 27；
Step 26:By task scheduling to independent clash handle queue；
Step 27:Dormancy X milliseconds, re-start locking；
Step 28:Judge to lock and whether succeed, if the determination result is YES, then return and perform step 24, otherwise perform step 29；
Step 29:Judge to lock whether number exceedes limit value, if the determination result is YES, then perform step 30, otherwise return Receipt row step 27；
Step 30:Cloud collision alert data are exported, the cloud collision alert data include：Alarming assignment and with alarm appoint Specific processing data corresponding to business；
Step 301:The cloud collision alert data of output are handled, therefrom have found that it is likely that the cloud punching for causing processing to be blocked It is prominent to confirm data.
The above embodiment of the present invention is when there are a large amount of lock conflicts, punching that system can send out cloud environment existing in real time automatically It is prominent, and by adjusting task handling process, Conflict Tasks are dispatched to individually processing task queue and are uniformly processed, are not re-used as Task under cloud environment carries out concurrent scheduling so as to avoid conflicting, and independent task processing queue is automatically provided to lock conflict task Avoid influenceing other tasks processing node.
In addition, the above embodiment of the present invention establishes automatic discovery and the affirmation mechanism of conflict simultaneously, therefrom finding can The cloud association process data that processing can be caused to block, in order to the during subsequent treatment pairs of task related to the cloud association process data Specially treated is carried out not determine whether to confirm that data are in order to avoid reducing whole system for cloud conflict in data processing entrance The disposal ability of system, because accounting is considerably less on the whole for colliding data, judge neither to influence again after locking failure in first time Overall treatment efficiency, and can are found in time when real colliding data occurs.
As shown in figure 3, the detailed process of above-mentioned steps 301 (cloud conflict discovery and confirmation processing), is described in detail as follows：
Step 31:The alarming assignment included in all cloud collision alert data is put into single alarming processing task team Row；
Step 32:An alarming assignment is taken out from alarming processing task queue；
Step 33:Judge whether to get alarming assignment, if the determination result is YES, then perform step 34, otherwise perform step 35；
Step 34:Read current time in system T2；
Step 35:Process dormancy is for a period of time (such as x milliseconds)；
Step 36:Delete failure cloud colliding data corresponding to current alarm task；
Step 37:Judge that current alarm task whether there is cloud colliding data, if the determination result is YES, then perform step 38, otherwise perform step 39；
Step 38:The nearest conflict number of renewal cloud colliding data, conflict discovery time recently, i.e., conflict number is added into 1, Conflict discovery time is set to T2；
Step 39:A corresponding cloud colliding data is generated, it is 1 to set conflict number, and conflict discovery time is T2, then Perform step 43；
Step 40:Judge whether current cloud colliding data is that cloud conflict confirms data, if the determination result is YES, then performs step Rapid 44, otherwise perform step 41；
Step 41:Judge whether nearest conflict number exceedes conflict and confirm threshold value, if the determination result is YES, then perform step 42, otherwise perform step 43；
Step 42:The state for updating cloud colliding data record confirms data for cloud conflict；
Step 43:Alarming assignment is turned back to common task scheduling queue again to be handled；
Step 44:By current alarm task scheduling to independent clash handle queue, and by current alarm task at alarm Deleted in reason task queue；
Step 45:Judge whether also have alarming assignment in alarming processing task queue, if the determination result is YES, then return and hold Row step 32, otherwise ends task.
Embodiments of the invention are by being analyzed and being handled to the alarm data of output, by the automatic hair for establishing conflict Now and affirmation mechanism, the cloud association process data for causing processing to be blocked therefrom are had found that it is likely that, in order to during subsequent treatment pairs with should The related task of cloud association process data carries out specially treated.
Especially, during system processing task, using the colliding data of discovery, it can just carry out conflict in locking failure first and sentence It is disconnected, effectively Conflict Tasks can be identified, and can avoids finding the treatment effeciency that conflict process reduces system；The conflict of confirmation is appointed Business will be dispatched to individually processing node and carry out non-concurrent processing, both avoid manual intervention and ensure that treatment effeciency turn avoid The appearance again of conflict；, just can be in time when conflict starts to occur on a small quantity by handling the alarm data in task processes It was found that effective conflict.
Embodiments of the invention also provide a kind of processing unit of cloud data, including：
Module is obtained, for obtaining the cloud associated data of waiting task；
First processing module, for unsuccessful, and the cloud that locks processing first in the cloud associated data of waiting task When associated data is that cloud conflict confirms data, the waiting task is dispatched to clash handle queue；Otherwise, to the cloud incidence number Handled according to locking again is carried out；
Output module, it is defeated for when the cloud associated data locks the unsuccessful number of processing more than a limit value again Go out cloud collision alert data.
When the first processing module carries out locking processing again to the cloud associated data, it is specifically used for：One is waited to preset After time interval, locking again is carried out to the cloud associated data and handled.
The first processing module is additionally operable to, right when the cloud associated data of waiting task locks handle successfully first The waiting task is normally handled, and discharges the lock.
The first processing module is additionally operable to, when the cloud associated data locks handle successfully again, to described pending Task is normally handled, and discharges the lock.
Wherein, the processing unit of cloud data also includes：
Second processing module, for handling the cloud collision alert data of output, obtain the cloud for causing processing to be blocked Conflict confirms data.
Wherein, the Second processing module is specifically used for：The alarming assignment that will be included in the cloud collision alert data of output It is put into an alarming processing task queue；An alarming assignment is taken out from alarming processing task queue, if getting, judgement currently takes The alarming assignment arrived whether there is cloud colliding data, if so, updating the nearest conflict number of the cloud colliding data, recently conflict Discovery time；
Judge whether the cloud colliding data is that cloud conflict confirms data, if so, the alarming assignment currently got is dispatched Deleted to independent clash handle queue, and by the alarming assignment currently got from alarming processing task queue；If it is not, judge most Whether nearly conflict number, which exceedes conflict, confirms threshold value, if so, the state of renewal cloud colliding data confirms data for cloud conflict；If It is no, alarming assignment is turned back to common task scheduling queue again and handled.
The Second processing module is additionally operable to, if not getting alarming assignment from alarming processing task queue, is spaced After one preset time period, continue to take alarming assignment from alarming processing task queue.
The Second processing module judges what is currently got after alarming assignment is got from alarming processing task queue Alarming assignment is additionally operable to before with the presence or absence of cloud colliding data：Present system time is read, and deletes the alarming assignment pair got The failure cloud colliding data answered.
The Second processing module is additionally operable to, and cloud colliding data, generation one is not present in the alarming assignment for judging currently to get Bar cloud colliding data, and alarming assignment is turned back into common task scheduling queue again and handled.
The state of the Second processing module renewal cloud colliding data is additionally operable to after confirming data for cloud conflict：To currently it take To alarming assignment be dispatched to independent clash handle queue, and by the alarming assignment currently got from alarming processing task queue Delete.
The alarming assignment currently got is dispatched to independent clash handle queue by the Second processing module, and will currently be taken To alarming assignment deleted from alarming processing task queue after be additionally operable to：
Judge whether also have alarming assignment in alarming processing task queue, if so, then being taken from alarming processing task queue Go out alarming assignment to be judged, if not having, terminate.
It should be noted that the device is device corresponding with above method embodiment, own in above method embodiment Implementation can also reach identical technique effect suitable for the embodiment of the device.
Described above is the preferred embodiment of the present invention, it is noted that for those skilled in the art For, on the premise of principle of the present invention is not departed from, some improvements and modifications can also be made, these improvements and modifications It should be regarded as protection scope of the present invention.

Claims (15)

  1. A kind of 1. processing method of cloud data, it is characterised in that including：
    Obtain the cloud associated data of waiting task；
    Waiting task cloud associated data first lock processing it is unsuccessful, and the cloud associated data be cloud conflict confirm number According to when, the waiting task is dispatched to clash handle queue；Otherwise, locking again is carried out to the cloud associated data to handle；
    When the cloud associated data locks the unsuccessful number of processing more than a limit value again, cloud collision alert data are exported.
  2. 2. the processing method of cloud data according to claim 1, it is characterised in that carry out adding again to the cloud associated data The step of lock processing, includes：
    After waiting a prefixed time interval, locking again is carried out to the cloud associated data and handled.
  3. 3. the processing method of cloud data according to claim 1, it is characterised in that also include：In the cloud of waiting task Associated data is locked first when handling successfully, and the waiting task is normally handled.
  4. 4. the processing method of cloud data according to claim 1, it is characterised in that also include：In the cloud associated data again When secondary locking is handled successfully, the waiting task is normally handled.
  5. 5. the processing method of the cloud data according to claim 3 or 4, it is characterised in that carried out to the waiting task Also include after normal processing：Discharge the lock.
  6. 6. the processing method of cloud data according to claim 1, it is characterised in that also wrapped after output cloud collision alert data Include：The cloud collision alert data of output are handled, the cloud conflict for obtaining causing processing to be blocked confirms data.
  7. 7. the processing method of cloud data according to claim 6, it is characterised in that the cloud collision alert data of output are entered Row processing, obtain including the step of causing the cloud conflict of processing obstruction to confirm data：
    The alarming assignment included in the cloud collision alert data of output is put into an alarming processing task queue；
    An alarming assignment is taken out from alarming processing task queue, if getting, whether the alarming assignment for judging currently to get is deposited In cloud colliding data, if so, updating the nearest conflict number of the cloud colliding data, conflict discovery time recently；
    Judge whether the cloud colliding data is that cloud conflict confirms data, if so, the alarming assignment currently got is dispatched into list Only clash handle queue, and the alarming assignment currently got is deleted from alarming processing task queue；If it is not, judge punching recently Whether prominent number, which exceedes conflict, confirms threshold value, if so, the state of renewal cloud colliding data confirms data for cloud conflict；If it is not, will Alarming assignment turns back to common task scheduling queue and handled again.
  8. 8. the processing method of cloud data according to claim 7, it is characterised in that also include：
    If not getting alarming assignment from alarming processing task queue, after being spaced a preset time period, continue at alarm Alarming assignment is taken in reason task queue.
  9. 9. the processing method of cloud data according to claim 7, it is characterised in that if being taken from alarming processing task queue To after alarming assignment, the alarming assignment for judging currently to get also includes before with the presence or absence of cloud colliding data：
    Present system time is read, and deletes failure cloud colliding data corresponding to the alarming assignment got.
  10. 10. the processing method of cloud data according to claim 7, it is characterised in that also include：
    Judge that the alarming assignment currently got is not present cloud colliding data, one cloud colliding data of generation, and by alarming assignment weight Newly common task scheduling queue is turned back to be handled.
  11. 11. the processing method of cloud data according to claim 7, it is characterised in that renewal cloud colliding data state be Cloud conflict also includes after confirming data：
    The alarming assignment currently got is dispatched to independent clash handle queue, and by the alarming assignment currently got at alarm Deleted in reason task queue.
  12. 12. the processing method of the cloud data according to claim 7 or 11, it is characterised in that appoint the alarm currently got Business is dispatched to independent clash handle queue, and is also wrapped after the alarming assignment currently got is deleted from alarming processing task queue Include：
    Judge whether also have alarming assignment in alarming processing task queue, accused if so, then being taken out from alarming processing task queue Alert task is judged, if not having, is terminated.
  13. A kind of 13. processing unit of cloud data, it is characterised in that including：
    Module is obtained, for obtaining the cloud associated data of waiting task；
    First processing module, for waiting task cloud associated data first lock processing it is unsuccessful, and the cloud associate When data are that cloud conflict confirms data, the waiting task is dispatched to clash handle queue；Otherwise, the cloud associated data is entered Locking is handled row again；
    Output module, for when the cloud associated data locks the unsuccessful number of processing more than a limit value again, exporting cloud Collision alert data.
  14. 14. the processing unit of cloud data according to claim 13, it is characterised in that also include：
    Second processing module, for handling the cloud collision alert data of output, obtain the cloud conflict for causing processing to be blocked Confirm data.
  15. 15. the processing unit of cloud data according to claim 14, it is characterised in that the Second processing module is specifically used In：The alarming assignment included in the cloud collision alert data of output is put into an alarming processing task queue；Appoint from alarming processing It is engaged in taking out an alarming assignment in queue, if getting, the alarming assignment for judging currently to get whether there is cloud colliding data, if Be update the nearest conflict number of the cloud colliding data, conflict discovery time recently；
    Judge whether the cloud colliding data is that cloud conflict confirms data, if so, the alarming assignment currently got is dispatched into list Only clash handle queue, and the alarming assignment currently got is deleted from alarming processing task queue；If it is not, judge punching recently Whether prominent number, which exceedes conflict, confirms threshold value, if so, the state of renewal cloud colliding data confirms data for cloud conflict；If it is not, will Alarming assignment turns back to common task scheduling queue and handled again.
CN201610799726.5A 2016-08-31 2016-08-31 Cloud data processing method and device Active CN107783843B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610799726.5A CN107783843B (en) 2016-08-31 2016-08-31 Cloud data processing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610799726.5A CN107783843B (en) 2016-08-31 2016-08-31 Cloud data processing method and device

Publications (2)

Publication Number Publication Date
CN107783843A true CN107783843A (en) 2018-03-09
CN107783843B CN107783843B (en) 2021-08-06

Family

ID=61451731

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610799726.5A Active CN107783843B (en) 2016-08-31 2016-08-31 Cloud data processing method and device

Country Status (1)

Country Link
CN (1) CN107783843B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111444199A (en) * 2019-01-17 2020-07-24 阿里巴巴集团控股有限公司 Data processing method and device, storage medium and processor
CN111857995A (en) * 2020-06-30 2020-10-30 海尔优家智能科技（北京）有限公司 Process scheduling method and device, storage medium and electronic device

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101702176A (en) * 2009-11-25 2010-05-05 南开大学 XML data concurrency control method based on local path lock
CN102981904A (en) * 2011-09-02 2013-03-20 阿里巴巴集团控股有限公司 Task scheduling method and system
CN103294728A (en) * 2012-03-05 2013-09-11 阿里巴巴集团控股有限公司 Data processing method and system
CN103336726A (en) * 2013-07-10 2013-10-02 北京百度网讯科技有限公司 Method and device detecting multitasking conflicts in Linux system
US20130346946A1 (en) * 2012-06-21 2013-12-26 Microsoft Corporation System for hosted, shared, source control build
CN103634347A (en) * 2012-08-24 2014-03-12 腾讯科技（深圳）有限公司 Parallel service processing method, device and system
US20140324925A1 (en) * 2013-04-24 2014-10-30 Dell Products, Lp Efficient Rename in a Lock-Coupled Traversal of B+Tree
CN104461747A (en) * 2014-12-31 2015-03-25 合一网络技术(北京)有限公司 Distributed type task scheduling system
CN104932933A (en) * 2014-03-21 2015-09-23 华为技术有限公司 Spin lock acquisition method and apparatus
CN105468450A (en) * 2015-12-29 2016-04-06 华为技术有限公司 Task scheduling method and system
CN105573850A (en) * 2015-11-09 2016-05-11 广州多益网络科技有限公司 Multi-process interactive method and system and server

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101702176A (en) * 2009-11-25 2010-05-05 南开大学 XML data concurrency control method based on local path lock
CN102981904A (en) * 2011-09-02 2013-03-20 阿里巴巴集团控股有限公司 Task scheduling method and system
CN103294728A (en) * 2012-03-05 2013-09-11 阿里巴巴集团控股有限公司 Data processing method and system
US20130346946A1 (en) * 2012-06-21 2013-12-26 Microsoft Corporation System for hosted, shared, source control build
CN103634347A (en) * 2012-08-24 2014-03-12 腾讯科技（深圳）有限公司 Parallel service processing method, device and system
US20140324925A1 (en) * 2013-04-24 2014-10-30 Dell Products, Lp Efficient Rename in a Lock-Coupled Traversal of B+Tree
CN103336726A (en) * 2013-07-10 2013-10-02 北京百度网讯科技有限公司 Method and device detecting multitasking conflicts in Linux system
CN104932933A (en) * 2014-03-21 2015-09-23 华为技术有限公司 Spin lock acquisition method and apparatus
CN104461747A (en) * 2014-12-31 2015-03-25 合一网络技术(北京)有限公司 Distributed type task scheduling system
CN105573850A (en) * 2015-11-09 2016-05-11 广州多益网络科技有限公司 Multi-process interactive method and system and server
CN105468450A (en) * 2015-12-29 2016-04-06 华为技术有限公司 Task scheduling method and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HOWTOS: ""Real-Time Linux Kernel Scheduler"", 《HTTPS://WWW.LINUXJOURNAL.COM/ARTICLE/10165》 *
陈锐忠: ""非对称多核处理器的若干调度问题研究"", 《中国博士学位论文全文数据库 信息科技辑》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111444199A (en) * 2019-01-17 2020-07-24 阿里巴巴集团控股有限公司 Data processing method and device, storage medium and processor
CN111444199B (en) * 2019-01-17 2023-11-14 阿里巴巴集团控股有限公司 Data processing method and device, storage medium and processor
CN111857995A (en) * 2020-06-30 2020-10-30 海尔优家智能科技（北京）有限公司 Process scheduling method and device, storage medium and electronic device
CN111857995B (en) * 2020-06-30 2024-05-24 海尔优家智能科技（北京）有限公司 Process scheduling method and device, storage medium and electronic device

Also Published As

Publication number Publication date
CN107783843B (en) 2021-08-06

Similar Documents

Publication Publication Date Title
CN103699437B (en) A kind of resource regulating method and equipment
CN107783843A (en) A kind of processing method and processing device of cloud data
CN103294728B (en) A kind of data processing method and system
Artalejo et al. Analysis of a stochasticclearing system with repeated attempts
CN104615486B (en) For searching for the multi-task scheduling of Extension Software Platform and executing methods, devices and systems
CN107015850A (en) Thread scheduling method and device
CN108230089A (en) A kind of online order processing method, system and device
CN110287206A (en) Method of the building for the distributed lock of Redis database
CN104636211B (en) Information interacting method and middleware system between a kind of software systems
CN110138662A (en) Session entry processing method, device and multiple nucleus system in multiple nucleus system
CN114257550B (en) Automatic control method and device for interface access flow, storage medium and server
CN101771670A (en) Method for concurrently identifying and processing sub-joins in multi-core systems
CN106789395B (en) A kind of Distributed PDM system monitoring data transmission method based on Web
Altiok et al. Single-stage, multi-product production/inventory systems with backorders
CN100531218C (en) Method, device and system for assuring correct execution
US11023146B1 (en) System and method for managing buffers in a computer system
CN107526690A (en) Remove the method and device of caching
CN111190913A (en) Distributed lock implementation method and system
CN110119301A (en) A kind of virtual machine migration method and system
CN107566677A (en) A kind of method and apparatus of resource transfers
CN105245471A (en) Message sending method and message sending device
US20240169281A1 (en) Shop management server, recording medium, andshop management method
CA2381739A1 (en) Adaptive spin latches
KR102319158B1 (en) Device and method of supporting transaction processing in asynchronous transaction processing architecture
CN116523170A (en) Tool management method, system, computer device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant