CN101561767A - Method and device for executing tasks based on operating system - Google Patents
Method and device for executing tasks based on operating system Download PDFInfo
- Publication number
- CN101561767A CN101561767A CNA2008100361201A CN200810036120A CN101561767A CN 101561767 A CN101561767 A CN 101561767A CN A2008100361201 A CNA2008100361201 A CN A2008100361201A CN 200810036120 A CN200810036120 A CN 200810036120A CN 101561767 A CN101561767 A CN 101561767A
- Authority
- CN
- China
- Prior art keywords
- task
- queue
- waiting
- reference position
- tasks
- 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
Links
Images
Landscapes
- Telephonic Communication Services (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention aims at providing a method and a device for executing tasks based on an operating system. Positions of tasks in a task queue are adjusted according to the execution states of the tasks in the task queue so as to lead uncompleted tasks to be preferentially detected and processed in a next detection period. If the current detection has no task needing to be processed, the next detection period begins after waiting preset time. The method for executing tasks provides a chance to be executed again for the known tasks needing to be executed within a shorter time on the one hand and provides different system detection frequencies on the other hand so as to save the occupation to system resources.
Description
Technical field
The present invention relates to a kind of task executing method and device thereof, relate in particular to task executing method and device thereof in real time operating system based on operating system.
Background technology
Along with computer system progressively is penetrated into the every nook and cranny of social life, also increasing gradually as operating system shared proportion in industry on computer system kernel and basis.Operating system not only is used for whole resources that managing computer system comprises hardware and software, also needs control program to carry out, and provides support for various tasks, so that all resources of computer system play a role to greatest extent.As a huge management control system, one of major function of operating system manages the execution of a plurality of different tasks exactly.
In the prior art, a plurality of tasks of operating system are detected by periodic scan successively according to the order of setting in the program code.If detect the current task of needing processing, then this task handled.For a particular task, if when program does not detect this task in order, then this task executions just must be waited for.Specifically, a plurality of tasks are put into task queue according to a definite sequence, the periodic scan of waiting system.System comprises some tasks in to formation in a sense cycle task groups scans task of whether having the current time to handle in this task groups to determine.If the task of needing processing is arranged in this task groups, then it is handled.If do not need the task of handling, perhaps handled all need handling of tasks in the current task group, then wait for next sense cycle, again other task in the formation is scanned.
Based on the task executing method of prior art, the invariant position of task in task queue, and can only latent period scanning detect.In other words, probability of happening height, the task that temporal correlation is strong are scanned detected probability and accidental task is basic identical, fail the task of finishing in a sense cycle, must wait for that also at least one sense cycle just may be performed once more.This task executing method may cause hot job in time not handle, and has also wasted the resource of system simultaneously.Another kind of solution commonly used is to utilize the strong task of interrupt mechanism processing time correlativity, but the interrupt resources of system is relatively limited, often can't satisfy the needs of system design.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, a kind of more Efficient and Flexible task execution mechanism is provided.
Based on this purpose, the invention provides a kind of task executing method and device based on operating system, concrete, the present invention dynamically adjusts its position in task queue according to the task executions state, to shorten the task dispatching time to be detected that needs execution.Preferably, when need handling when the current time in the detected task groups in the sense cycle of task fails all to finish in this sense cycle, the present invention places task queue foremost uncompleted task, and in next sense cycle detecting foremost from task queue, and carry out detected waiting task, thereby guarantee the chance that uncompleted task can be carried out once more in last sense cycle in the shortest time.Preferably, when all finishing in this sense cycle, need handling when the current time in the detected task groups in the sense cycle of task (needs to understand, this moment, this sense cycle did not finish probably as yet), the present invention does not wait for that this sense cycle finishes, but enter next sense cycle immediately, the waiting task in this task queue is detected.In addition,, then wait for the time of a predetermined length, enter next sense cycle again, be used for the resource overhead that task queue detects with reasonable control if do not need the task of handling in the current sense cycle.
According to an aspect of the present invention, a kind of task executing method based on operating system is provided, wherein, may further comprise the steps: a) whether detection comprises one or more waiting tasks since first quantity task of the detection reference position of a task queue; B) if comprise one or more waiting tasks in described first quantity task, then second quantity waiting task in described one or more waiting tasks is handled, obtaining second quantity treated task, and adjust its position in described task queue according to described each treated task executions state; C) repeating described step b) all handles until described one or more waiting tasks.
According to another aspect of the present invention, a kind of task execution device based on operating system also is provided, it is characterized in that, comprising: pick-up unit is used for detecting since first quantity task of the detection reference position of a task queue whether comprise one or more waiting tasks; Adjusting gear, if comprise one or more waiting tasks in described first quantity task, then described adjusting gear is used for second quantity waiting task of described one or more waiting tasks is handled, obtaining second quantity treated task, and adjust its position in described task queue according to described each treated task executions state; Control device is used to judge whether that described one or more waiting task all handles, if untreated waiting task is arranged, then controls described adjusting gear described untreated waiting task is operated.
Compared with prior art, the present invention can the task for known needs execution provide the chance that is performed once more in the shorter time.On the other hand, if have waiting task to handle in the cycle in current detection, then after the processing of finishing current detection detected waiting task in the cycle, need not to wait for continue to detect whether also have other waiting task in the task queue at once.This manner of execution provides the higher detection frequency more for a long time in system task, provides lower detection frequency in system task more after a little while, has realized " frequency conversion operation " in a sense, has improved the efficient of operating system.
Description of drawings
By reading the detailed description of doing with reference to the following drawings that non-limiting example is done, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 is the synoptic diagram of the task queue of an embodiment according to the present invention;
Fig. 2 is the process flow diagram based on the task executing method of operating system according to a specific embodiment of the present invention;
Fig. 3 is the process flow diagram based on the task executing method of operating system according to a preferred embodiment of the present invention; Fig. 4 is the synoptic diagram according to the task queue of one embodiment of the present invention;
Fig. 5 is the task execution device block diagram based on operating system according to a specific embodiment of the present invention;
Wherein, same or analogous Reference numeral is represented same or analogous steps characteristic or device/module.
Embodiment
Understand the part notion that at first occurs in the herein interpreted for convenient:
-task queue: herein, the formation that will be used for the task that storage system may need to carry out (comprising that task that the current detection cycle need handle and current detection cycle do not need the task of handling) is called task queue.
-the first quantity task: be scanned the task of detection in a sense cycle, according to the difference of embodiment, this first quantity task may be the part or all of task in the task queue.
-waiting task: the task that need are handled in the cycle in current detection, specifically detect from first quantity task by scanning and determine.
-treated task: a waiting task promptly becomes a treated task after treatment in current detection in the cycle, note that treated task may finish dealing with, and also may be untreated and finish;
-the second quantity task: operating system is the waiting task of interior executed in parallel at one time, for instance, for only supporting single-threaded uniprocessor dispatching system, owing to can only have a thread to be in executing state on each processor, therefore can only support the execution one by one to waiting task, then the number of this second quantity task of this kind situation correspondence is 1.For the system of scheduling of multiprocessor thread, can be integer smaller or equal to above-mentioned first quantity according to the number of this second quantity task of difference of system thread scheduling mechanism.
-executing state: carry out corresponding various task status with task.Execution herein is meant process the taking processor of task, be that the thread of task is scheduled and enters running status, and the adjustment itself of task relevant information operations such as () status information that ordering detection for example etc. causes are read, priority is adjusted is not regarded as task executions.Executing state comprises following three kinds: do not carry out, promptly in waiting for implementation; Do not finish, promptly through execution but task be not done, wait for next time and carrying out; Finish, promptly through carrying out and being done.Thus, it will be understood by those skilled in the art that the state of a treated task may be for not finishing or finishing.
In addition, it should be noted that and to adjust the position of task in formation herein, task is carried out the various operations that task is carried out such as (no matter whether finishing execution) be referred to as
" processing ".Consider such a case, suppose to work as previous task for sending message information, still owing to reasons such as network timeout or this user device system overload temporarily can't be finished transmission to another subscriber equipment.Should be appreciated that this task passes through " processing " this moment, but do not finish " execution ", and its executing state is " not finishing ".
-detect reference position: refer to that current detection needs the reference position of first quantity task that detects in the cycle.The synoptic diagram that shown in Figure 1 is according to the task queue of a specific embodiment of the present invention.Suppose that system can detect 64 (promptly first quantity is 64) tasks in a sense cycle, detecting reference position is primary importance shown in Fig. 1.Then system detects 64 tasks that begin from the task of being positioned at primary importance in the formation shown in Figure 1 in current sense cycle, determines wherein whether to comprise waiting task.Suppose that the detection reference position of next sense cycle is current detection reference position (primary importance) position of first quantity task afterwards, then the detection reference position of next sense cycle is the second place shown in Fig. 1.
, as do not have other special instruction herein, above-mentioned notion all can be explained by foregoing.
Below in conjunction with process flow diagram method of the present invention partly is described.
Fig. 2 is the process flow diagram based on the task executing method of operating system according to a specific embodiment of the present invention.Followingly task executing method provided by the invention is described, suppose that primary importance shown in Fig. 1 is the current detection reference position, and first quantity is 8 with reference to Fig. 1.It will be appreciated by those skilled in the art that said method can be applied to include but not limited to personal computer (PC), the webserver, intelligent mobile terminal, set-top box or the like among any equipment based on operating system, below is example with PC.
In step S10, the operating system among the PC (hereinafter to be referred as system) detects 8 tasks that begin from the reference position of detection shown in Fig. 1, thereby knows whether the waiting task that needs processing is wherein arranged.Each task in the real time operating system can corresponding specific zone bit, and according to the difference setting of operating system, whether system can walk abreast or sequentially come these 8 tasks by the special sign position is that waiting task detects.
Concrete, system reserves certain storage space and stores whether be used to identify each task be the identification information of waiting task, described identification information comprises a plurality of zone bits, and whether corresponding specific detected task of each zone bit wherein, its value of 1 or 0 be then to should task being waiting task.System comes at any time according to definition status, priority or other relevant informations of each task or the value of the corresponding zone bit of regular update.For instance, task for one-period, it promptly needs to carry out once every a predetermined period, for this reason, system will judge that whether over and done with this task carried out from last time an above-mentioned predetermined period by the surveillance clock, if judged result is a "Yes", then revise this task corresponding marker bit.
According to above-mentioned explanation, the system of one of ordinary skill in the art will readily recognize that can know whether the task of needing processing is arranged in these 8 tasks by the corresponding zone bit that reads in the above-mentioned identification information.The ground that is without loss of generality supposes that zone bit is that 1 its corresponding task of expression is a waiting task, and zone bit is 0 expression corresponding task for being not (the current detection cycle) waiting task.Be appreciated that if the identification information that system reads is 0,000 1010, can know that then the 5th and the 7th task (being task A shown in Fig. 1 and task B) that begins from initial detection position is detected waiting task.Optionally, read one by one by zone bit and can determine that equally whether its corresponding task is waiting task, no longer gives an example herein above-mentioned 8 tasks.For detected waiting task, system can set corresponding position and identify in its relevant information, " waiting task in this sense cycle " this identity with clear and definite this task, certainly, be this purpose, system also can put into newly-built formation or array with the pointer of each above-mentioned waiting task, repeats no more herein.Suppose, as shown in Figure 1,, determine that task A and task B are waiting task from detecting in 8 tasks that reference position begins in the present embodiment by detecting.Detect in first quantity task and have waiting task, so this method enters step S11.
In step S11, waiting task A and the waiting task B that determines among the step S10 handled.For ease of understanding, at first cite a plain example, wherein, second quantity is 1, also, at one time, operating system is only handled individual task.In view of waiting task has two, system handles task A earlier.Suppose that task A is: receive a data block from port one, the possibility of result after then it being handled is to have finished the reception (finishing) to this data block or temporarily can't carry out this data block is received (not finishing).It should be noted that after this no matter whether task A can be done through above-mentioned processing, it all will be regarded as treated task at the state of current detection in the cycle.
After task A carried out above-mentioned processing, system also upgrades the executing state of task A, concrete, its executing state is not updated to " not finishing " or " finishing " from " carrying out ", this renewal process can realize by being provided with the corresponding relevant information of task A, this belongs to the mature technology that those skilled in the art understand, and repeats no more herein.
If system has finished the operation that receives this data block from port one in this sense cycle, the executing state after then task A upgrades is " finishing ".Operating system comes the position of task A in this task queue adjusted according to executing state and the pre-defined rule after upgrading, and for example, is placed on A ' position as shown in Figure 1 in the task queue.Preferably, when the position of treated task is adjusted, uncompleted each treated task can be placed in task queue before completed each treated task.Preferred, can completed treated task be placed the tail of the queue of task queue according to the task executions state with the head of the queue of uncompleted each treated task as for task queue.Be appreciated that, by above preferred implementation, the task of needing processing in next sense cycle badly can be concentrated on the forward position of formation, processing mode of the prior art relatively, can save the detection time that needs Processing tasks, also reduce the probability that waits as long for that incident causes.
Task A after treatment, system judges whether detected before this all waiting tasks all processed (executing state after no matter it is handled is why) in step S12.If untreated waiting task is still arranged then return step S11 and proceed to handle.In this example, system knows that by judgement waiting task B handles as yet, then returns step S11.Described judgement can be based on multiple existing techniques in realizing.For example, in step S10, the numerical value (being 2 in this example) of the quantity of waiting task being put into a memory block, whenever finish a task handling is just subtracted one with the numerical value in this memory block, is 0 to finish the processing to all waiting tasks until this numerical value.Prior art has multiple ripe support scheme to described judgement, herein specifically citation no longer one by one.
Suppose that task B is: send one group of data to port 2, but, in step S11, fail to finish execution to task B owing to reasons such as network congestion or port 2 temporary closes.According to the elaboration of front, system can be considered as task B treated task, and its executing state is not updated to " not finishing " from " carrying out ", and task B is put into the position of task queue B ' according to the executing state after this renewal.Wherein the particular location of B ' is determined by the preset rules of system, and for example, being detected handling of task once more places the forward position of formation as early as possible with hope, need not frequent detected task place task queue by the position, back.
Task B after treatment, system judges whether that by step S12 all waiting tasks (A, B) are all processed once more, if all finish, then finishes the operation in current detection cycle.Preferably, this method enters step S10 immediately, begins a new sense cycle.A lot of existing operating systems can be supported the parallel processing to a plurality of tasks.Therefore, it will be understood by those skilled in the art that the task number of mentioning that can be executed in parallel at one time (i.e. " second quantity ") can be a plurality of herein.For example, can handle waiting task A in this example and waiting task B simultaneously based on the windows XP task scheduling system of double-core or multinuclear, in the case, the value of second quantity is 2.Consideration takies and the requirement of multiple factor such as speed system resource, can the adjusting of this second quantity by the software program setting, and not only by the hardware configuration decision of system.
The value of carrying out second quantity task and second quantity simultaneously is greater than under 1 the situation, and after each that can be in handling second quantity task, the executing state after upgrading according to this task is immediately adjusted the position of this task in task queue.Also can handle and upgrade its executing state separately earlier to second quantity task, by the time all tasks in second quantity task are all finished handle and the operation of executing state renewal after, come its position to adjust respectively according to the executing state after each the renewal in this second quantity task again.Be to be understood that; owing to have a plurality of tasks in the task queue; and core concept of the present invention is to adjust its position in task queue according to each task executions state; therefore; ordinal relation in the present embodiment between related each step does not constitute the qualification to protection domain of the present invention only at specific embodiment.
More than, be that the supposition of waiting task is introduced this method based on task A among Fig. 1 and task B.And if comprise all non-waiting task of above-mentioned 8 tasks of task A, B, then optimum system choosing ground was waited for after the time of certain-length, repeated execution of steps S10.
Below, with reference to Fig. 3 and in conjunction with Fig. 1 an optimization modification of above-mentioned embodiment is done concise and to the point description.Fig. 3 is the process flow diagram based on the task executing method of operating system according to a preferred embodiment of the present invention.
Total task number of supposing to exist in the task queue also, can't intactly scan whole task queue in each sense cycle greater than first quantity.So, in step S12 through judgement learn first quantity task handling has been all over after, this method also may need all the other tasks in the task queue are detected and handle.Wherein a kind of optional mode, be exactly after each sense cycle, detect reference position according to certain Policy Updates, and first quantity task that in next sense cycle the detection reference position after renewal is begun detects corresponding every follow-up each step operation of execution then.After the detection and processing of the first quantity task of finishing the current detection cycle, redefine the detection reference position, begin the next cycle of treatment that detects again.Cooperate relevant detection reference position update rule, use the task executing method mechanism of this this kind circular treatment can adjust the probability that each task in the formation is detected and carry out, help the reasonable distribution of system process time quota.
For instance, still suppose task queue as shown in Figure 1, the detection reference position in current detection cycle is a primary importance.This method is carried out to step S12, after the processing of each waiting task in first quantity task that primary importance is begun has been finished in judgement, upgrades detecting reference position in step S13.The renewal of described detection position can be passed through accomplished in many ways, for instance, task queue can be stored with the form of array, wherein an element in the corresponding array of each task.The pointer of corresponding each task is stored in array successively, and specifies the corresponding element that detects reference position in the independent pointed array.In each sense cycle, first quantity positions of elements behind current this pointer corresponding element of the pointed that this is independent (the detection reference position in current detection cycle), and with this position as the detection reference position (being the second place among Fig. 1) after upgrading.It will be understood by those skilled in the art that the not unique correspondence of detection reference position after the renewal, can adjust flexibly upgrading the rule that detects reference position according to the designing requirement of system.Preferably and since according to the present invention current detection in the cycle uncompleted treated task will concentrate on task queue than front end, therefore can detect the head of the queue that reference position is updated to task queue.
Finish the renewal that detects reference position by step S13 after, execution in step S10 detects processing to the task that the second place begins.
Preferably, if do not have waiting task in first quantity task that current detection detected in the cycle, then system to detect reference position and upgrade and wait for time cycle of a predetermined length after again the task in the task queue is detected.Still be described below with reference to Fig. 3.
Suppose that the primary importance among Fig. 1 is the detection reference position.In step S10, detect first quantity task that begins from the task of being arranged in primary importance and whether comprise waiting task.
If wherein comprise waiting task, then enter step S11 and come waiting task is done further processing, repeat no more herein.
If wherein do not comprise waiting task, then enter step S14, wait for the time cycle of a predetermined length.Concrete, the length transition of the time cycle that needs can be waited for is the number of system clock cycle (system clock) correspondingly, and this numerical value is kept in the memory block, each new clock period subtracts 1 with this numerical value later, when this numerical value is reduced to 0, begin next sense cycle.Certainly, also can utilize the timer of system to keep or jump out above-mentioned waiting status.In view of the technological means well-known to those skilled in the art that is set to, no longer add herein to describe to time cycle of predetermined length.
Preferably, in step S14, can also upgrade detecting reference position, with the detection reference position of the residing position of first quantity task after the task on the current detection reference position as next sense cycle, that is, the position of task is used as detection reference position in the next sense cycle on the second place among Fig. 1.
It should be noted that foregoing as the position of first quantity task " this current detection reference position after " describe be adjust each task according to executing state the position before as described in the relative position relation of each task in the formation.Imagine a kind of extreme case, the task number that comprises in the task queue is considerably beyond first quantity, each all is a waiting task detect to determine first quantity task that begins from the task of being arranged in primary importance through step S10, and wherein each task executions state all becomes " executed " after processed.Therefore, the position of above-mentioned first quantity task all is adjusted according to pre-defined rule, is without loss of generality, and supposes that they are placed in the tail end of task queue.In other words, in the task queue shown in Figure 1 from primary importance to the second place before the task of a position all be adjusted to the tail end of task queue, and moved forward successively from some positions that the task of being positioned at the second place originally begins in the back task.
After the completing place is adjusted, the task of being arranged in the second place originally will be positioned at the primary importance of Fig. 1, so the detection reference position after the renewal is arranged in the primary importance of formation, promptly originally be positioned at the new position of the task of the second place.Preferably, if the current detection reference position has been positioned at the position, back of task queue, system can't find the position (because of it does not exist) of first quantity task after the current detection reference position, then can be with the head of the queue of task queue as new detection reference position.
It will be understood by those skilled in the art that in the above described manner and upgrade, can avoid having handled in the cycle in current detection in a large number of task is carried out duplicate detection detecting reference position.After the time cycle of waiting for a predetermined length, repeating step S10.
Be appreciated that by among the step S11 to position adjust of treated task in task queue can with current detection in the cycle executing state in the high task-set of the task of " not finishing " and occurrence frequency at the specific part of task queue.By concentrating this specific part is detected afterwards, and detected waiting task is handled, can improve the execution efficient of system.If do not detect waiting task, exist the probability of need handling in the cycle in current detection of task then less relatively in the then whole task queue at this specific part.At this moment, the time cycle of waiting for a predetermined length can be reduced detection taking system resource.
Below in conjunction with Fig. 3 and in conjunction with Fig. 4 another preferred embodiment of the present invention is further described, wherein, Fig. 4 is the synoptic diagram according to the task queue of one embodiment of the present invention, to the description of described another preferred embodiment hypothesis based on following indefiniteness:
1) operating system of subscriber equipment is embedded system;
2) corresponding and continuous other subscriber equipment (promptly with between the current user equipment being connected the subscriber equipment that can exchange information, IP address that each subscriber equipment is corresponding unique) of this subscriber equipment of each task by effective communication link;
3) can comprise one or more incidents in a task, and this incident can send data for other subscriber equipment to its correspondence or receive data by other subscriber equipment of its correspondence;
4) under original state, all tasks in the task queue all are waiting task;
5) comprise 26 tasks of a-z as shown in Figure 4 in the task queue successively;
6) in a sense cycle, 4 tasks are detected (value of first quantity is 4);
7) operating system is single-threaded (promptly the value of second quantity is 1);
8) only comprise an incident in each task.
In step S10, at first task a, b, c and the d that is in the 1-4 of queue position detected, and a that sets the tasks, b, c are the waiting task in current detection cycle, then in step S11, earlier task a is handled and finishes execution, therefore the executing state with task a is updated to " finishing ", and task a placed the end of task queue, promptly after the task z.Then task b is repeated aforesaid operations, the executing state with task b is updated to " finishing " equally, is placed on the end of task queue, promptly after the task a.
Then, task c is repeated aforesaid operations, suppose task c for to its relative user equipment I transmission 500k data, and owing to the reason of network congestion, temporarily can't continue to carry out after the data transmission of finishing 50k, then the executing state with task c is updated to " not finishing ".And be placed on task queue foremost, promptly be positioned at before the d.Therefore, when the current detection end cycle, task queue four tasks foremost are followed successively by cdef, and four tasks of least significant end are as yzab (task b is last task of task queue).Because there is executing state in current detection in the cycle is the waiting task of " not finishing ", therefore upgrading the detection reference position is the head of the queue of task queue.After finishing aforesaid operations, enter next sense cycle (, next sense cycle being defined as second sense cycle) at once for ease of describing.
In second sense cycle, to detecting 4 tasks of reference position (residing position task c this moment) beginning: task cdef detects, wherein all tasks are the waiting task in current detection cycle, and then repeating step S11 and S12 handle one by one to it.Suppose task c because network congestion still can't be carried out subsequent treatment, then this task is continued to place task queue foremost, task d and e are processed and finish execution, then are placed on tail of the queue, promptly after the task b.Suppose that task f sends Query Information to port II, and, then task f is placed head of the queue because network congestion can't be finished execution.Therefore, when second sense cycle finished, task queue four tasks foremost were followed successively by fcgh, and six tasks of least significant end are followed successively by yzabde (task b is last task of task queue).
In the 3rd sense cycle, fcgh detects to task.By that analogy, after a plurality of sense cycle, suppose that network recovery is unimpeded, then task f gets and finishes, and is placed in the afterbody of task queue after finishing; If port I still closes, then task c still is in the forward position of task queue, can obtain more machines meeting.
It will be appreciated by those skilled in the art that in the process of Network Transmission the breakpoint transmission of data block sends and receives and can not once finish is multiple incident in the transmission course, also need to be divided into a plurality of file fragmentations for the big data block of needs transmission and operate.Therefore, one sends or corresponding other subscriber equipment that carries out data interaction with this subscriber equipment of reception incident, and one with this subscriber equipment carry out other subscriber equipment of data interaction often corresponding a plurality of transmissions or reception incident.Since between individual event and other user device IP address for concerning one to one, those skilled in the art understand, through after a plurality of sense cycle, be adjusted to forward position in the task queue at this subscriber equipment place with the mutual more pairing task of other subscriber equipment of this subscriber equipment.If detecting reference position is the head of the queue of task queue, then corresponding with mutual more other subscriber equipment of this subscriber equipment task can obtain the chance of how detected processing, need not wait for that the scanning to all user device IP address can preferentially realize its data transmission.And and the pairing task of other subscriber equipment of temporary transient no datat piece transmission between this subscriber equipment be placed in the end of task queue, promptly reduced priority.
Understand easily, transmission network as network based on the point-to-point transmission agreement in, this subscriber equipment may link to each other with a plurality of other subscriber equipmenies, but may be only carries out data interaction (for example, send or receiver, video voice data) with several other subscriber equipmenies wherein in a certain particular moment.By the position of adjustment task in formation, can improve the detected probability of handling of the pairing task of other subscriber equipment that has data interaction with this subscriber equipment, thereby the minimizing temporary file takies buffer memory, saves the time of detecting and waiting for, improves transmission of data blocks efficient simultaneously.
Below in conjunction with block diagram device part of the present invention is described.
Fig. 5 is the task execution device block diagram based on operating system according to a specific embodiment of the present invention.Followingly task execution device 1 provided by the invention is described with reference to Fig. 1.This task execution device 1 comprises pick-up unit 10, adjusting gear 11, control device 12 and waits for updating device 13.Suppose that primary importance shown in Fig. 1 is the current detection reference position, and first quantity is 8.It will be appreciated by those skilled in the art that said apparatus can be applied to include but not limited to personal computer (PC), the webserver, intelligent mobile terminal, set-top box or the like among any equipment based on operating system, below is example with PC.
Pick-up unit 10 among the PC detects 8 tasks that begin from the reference position of detection shown in Fig. 1, thereby knows whether the waiting task that needs processing is wherein arranged.Each task in the real time operating system can corresponding specific zone bit, and according to the difference setting of operating system, whether system can walk abreast or sequentially come these 8 tasks by the special sign position is that waiting task detects.
Concrete, system reserves certain storage space and stores whether be used to identify each task be the identification information of waiting task, described identification information comprises a plurality of zone bits, and whether corresponding specific detected task of each zone bit wherein, its value of 1 or 0 be then to should task being waiting task.System comes at any time according to definition status, priority or other relevant informations of each task or the value of the corresponding zone bit of regular update.For instance, task for one-period, it promptly needs to carry out once every a predetermined period, for this reason, system will judge that whether over and done with this task carried out from last time an above-mentioned predetermined period by the surveillance clock, if judged result is a "Yes", then revise this task corresponding marker bit.
According to above-mentioned explanation, one of ordinary skill in the art will readily recognize that pick-up unit 10 can know whether the task of needing processing is arranged in these 8 tasks by the corresponding zone bit that reads in the above-mentioned identification information.The ground that is without loss of generality supposes that zone bit is that 1 its corresponding task of expression is a waiting task, and zone bit is that 0 expression corresponding task is not to be-preceding sense cycle) waiting task.Be appreciated that if the identification information that pick-up unit 10 reads is 0,000 1010, can know that then the 5th and the 7th task (being task A shown in Fig. 1 and task B) that begins from initial detection position is detected waiting task.Optionally, read one by one by zone bit and can determine that equally whether its corresponding task is waiting task, no longer gives an example herein above-mentioned 8 tasks.For detected waiting task, pick-up unit 10 can be set corresponding position and identify in its relevant information, " waiting task in this sense cycle " this identity with clear and definite this task, certainly, be this purpose, pick-up unit 10 also can be put into newly-built formation or array with the pointer of each above-mentioned waiting task, repeats no more herein.Suppose, as shown in Figure 1,, determine that task A and task B are waiting task from detecting in 8 tasks that reference position begins in the present embodiment by detecting.Pick-up unit 10 detects in first quantity task and has waiting task, and will to detect task A and task B be that the information of waiting task sends adjusting gear 11 to.
11 pairs of pick-up units of adjusting gear, 10 detected waiting task A and waiting task B handle.For ease of understanding, at first cite a plain example, wherein, second quantity is 1, also, at one time, operating system is only handled individual task.In view of waiting task has two, adjusting gear 11 is handled task A earlier.Suppose that task A is: receive a data block from port one, the possibility of result after then it being handled is to have finished the reception (finishing) to this data block or temporarily can't carry out this data block is received (not finishing).It should be noted that after this no matter whether task A can be done through above-mentioned processing, it all will be regarded as treated task at the state of current detection in the cycle.
After task A carried out above-mentioned processing, adjusting gear 11 also upgrades the executing state of task A, concrete, its executing state is not updated to " not finishing " or " finishing " from " carrying out ", this renewal process can realize by being provided with the corresponding relevant information of task A, this belongs to the mature technology that those skilled in the art understand, and repeats no more herein.
If system has finished the operation that receives this data block from port one in this sense cycle, the executing state after then task A upgrades is " finishing ".Adjusting gear 11 comes the position of task A in this task queue adjusted according to executing state and the pre-defined rule after upgrading, and for example, is placed on A ' position as shown in Figure 1 in the task queue.Preferably, when the position of treated task is adjusted, uncompleted each treated task can be placed in task queue before completed each treated task.Preferred, can completed treated task be placed the tail of the queue of task queue according to the task executions state with the head of the queue of uncompleted each treated task as for task queue.Be appreciated that, by above preferred implementation, the task of needing processing in next sense cycle badly can be concentrated on the forward position of formation, processing mode of the prior art relatively, can save the detection time that needs Processing tasks, also reduce the probability that waits as long for that incident causes.So far, adjusting gear 11 is finished the processing to task A.
After adjusting gear 11 was finished processing to task A, control device 12 judged whether detected before this all waiting tasks all processed (executing state after no matter it is handled why).If untreated waiting task still arranged then proceed to handle by adjusting gear 11.In this example, control device 12 knows that by judgement waiting task B handles as yet, is proceeded to handle by adjusting gear 11.Described judgement can be based on multiple existing techniques in realizing.For example, pick-up unit 10 is put into a memory block with the numerical value (being 2 in this example) of the quantity of waiting task, whenever finish a task handling is just subtracted one with the numerical value in this memory block, and be 0 to finish processing until this numerical value to all waiting tasks.Prior art has multiple ripe support scheme to described judgement, herein specifically citation no longer one by one.
Suppose that task B is: send one group of data to port 2, but owing to reasons such as network congestion or port 2 temporary closes, fail to finish execution to task B at current detection cycle middle regulator 11.According to the elaboration of front, adjusting gear 11 can be considered as treated task with task B, and its executing state is not updated to " not finishing " from " carrying out ", and task B is put into the position of task queue B ' according to the executing state after this renewal.Wherein the particular location of B ' is determined by the preset rules of system, and for example, being detected handling of task once more places the forward position of formation as early as possible with hope, need not frequent detected task place task queue by the position, back.
Task B after treatment, control device 12 judges whether that by step S12 all waiting tasks (A, B) are all processed once more, if all finish, then finishes the operation in current detection cycle.Preferably, monitoring device 10 carries out detecting operation next time immediately, begins a new sense cycle.A lot of existing operating systems can be supported the parallel processing to a plurality of tasks.Therefore, it will be understood by those skilled in the art that the task number of mentioning that can be adjusted device 11 executed in parallel at one time (i.e. " second quantity ") can be a plurality of herein.For example, can handle waiting task A in this example and waiting task B simultaneously based on the windows XP task scheduling system of double-core or multinuclear, in the case, the value of second quantity is 2.Consideration takies and the requirement of multiple factor such as speed system resource, can the adjusting of this second quantity by the software program setting, and not only by the hardware configuration decision of system.
Adjusting gear 11 is carried out the value of second quantity task and second quantity simultaneously greater than under 1 the situation, adjusting gear 11 can be in handling second quantity task each after, the executing state after upgrading according to this task is immediately adjusted the position of this task in task queue.Also can handle and upgrade its executing state separately earlier to second quantity task, by the time all tasks in second quantity task are all finished handle and the operation of executing state renewal after, come its position to adjust respectively according to the executing state after each the renewal in this second quantity task again.Be to be understood that; owing to have a plurality of tasks in the task queue; and core concept of the present invention is to adjust its position in task queue according to each task executions state; therefore; ordinal relation in the present embodiment between related each step does not constitute the qualification to protection domain of the present invention only at specific embodiment.
More than, be that the supposition of waiting task is introduced this method based on task A among Fig. 1 and task B.And if comprise all non-waiting task of above-mentioned 8 tasks of task A, B, then pick-up unit 10 detects after preferably waiting for time of certain-length next time.
Below, refer again to Fig. 5 and a modification of above-mentioned embodiment is done concise and to the point description in conjunction with Fig. 1.Wherein control device 12 also comprises control updating device 120.
Total task number of supposing to exist in the task queue also, can't intactly scan whole task queue in each sense cycle greater than first quantity.So, in control device 12 through judgement learn first quantity task handling has been all over after, task execution device 1 also may need all the other tasks in the task queue are detected and handle.Wherein a kind of optional mode, be exactly after each sense cycle, pick-up unit 10 detects reference position according to certain Policy Updates, and first quantity task that in next sense cycle the detection reference position after renewal is begun detects corresponding every follow-up each step operation of execution then.After the detection and processing of the first quantity task of finishing the current detection cycle, control updating device 120 redefines the detection reference position, begins the next cycle of treatment that detects again.Cooperate relevant detection reference position update rule, use the task executing method mechanism of this this kind circular treatment can adjust the probability that each task in the formation is detected and carry out, help the reasonable distribution of system process time quota.
For instance, still suppose task queue as shown in Figure 1, the detection reference position in current detection cycle is a primary importance.After the processing of each waiting task in first quantity task that primary importance is begun had been finished in control device 12 judgements, 120 pairs of control updating devices detected reference position and upgrade.The renewal of described detection position can be passed through accomplished in many ways, for instance, task queue can be stored with the form of array, wherein an element in the corresponding array of each task.The pointer of corresponding each task is stored in array successively, and specifies the corresponding element that detects reference position in the independent pointed array.In each sense cycle, first quantity positions of elements behind current this pointer corresponding element of the pointed that this is independent (the detection reference position in current detection cycle), and with this position as the detection reference position (being the second place among Fig. 1) after upgrading.It will be understood by those skilled in the art that the not unique correspondence of detection reference position after the renewal, can adjust flexibly upgrading the rule that detects reference position according to the designing requirement of system.Preferably and since according to the present invention current detection in the cycle uncompleted treated task will concentrate on task queue than front end, therefore can detect the head of the queue that reference position is updated to task queue.So far, control updating device 120 is finished detecting the renewal of reference position.
After control updating device 120 was finished the renewal that detects reference position, the task that 10 pairs of second places of pick-up unit begin detected processing.
Preferably, if do not have waiting task in current detection first quantity task that pick-up unit 10 detects in the cycle, then control 120 pairs of detections of updating device reference position and upgrade, pick-up unit 10 detects the task in the task queue again after the time cycle of waiting for a predetermined length.Below in conjunction with Fig. 5 and with reference to Fig. 1 the preferred embodiment is described further.
Suppose that the primary importance among Fig. 1 is the detection reference position.Pick-up unit 10 detects first quantity task that begins from the task of being arranged in primary importance and whether comprises waiting task.
If pick-up unit 10 detects waiting task, then 11 pairs of waiting tasks of adjusting gear are done further processing, repeat no more herein.
If wherein do not comprise waiting task, then wait for the time cycle of a predetermined length of updating device 13 control waits.Concrete, the length transition of the time cycle that needs can be waited for is the number of system clock cycle (system clock) correspondingly, and this numerical value is kept in the memory block, each new clock period subtracts 1 with this numerical value later, when this numerical value is reduced to 0, begin next sense cycle.Certainly, also can utilize the timer of system to keep or jump out above-mentioned waiting status.In view of the technological means well-known to those skilled in the art that is set to, no longer add herein to describe to time cycle of predetermined length.
Preferably, wait for that updating device 13 can also upgrade detecting reference position, with the detection reference position of the residing position of first quantity task after the task on the current detection reference position as next sense cycle, that is, the position of task is used as detection reference position in the next sense cycle on the second place among Fig. 1.
It should be noted that foregoing as the position of first quantity task " this current detection reference position after " describe be adjust each task according to executing state the position before as described in the relative position relation of each task in the formation.Imagine a kind of extreme case, the task number that comprises in the task queue is considerably beyond first quantity, install 10 after testing and detect to determine first quantity task that begins from the task of being arranged in primary importance each all is a waiting task, and wherein each task executions state all becomes " executed " after being adjusted device 11 processing.Therefore, the position of above-mentioned first quantity task all is adjusted device 11 according to pre-defined rule and adjusts, and is without loss of generality, and supposes that they are placed in the tail end of task queue.In other words, in the task queue shown in Figure 1 from primary importance to the second place before the task of a position all be adjusted to the tail end of task queue, and moved forward successively from some positions that the task of being positioned at the second place originally begins in the back task.
After the completing place is adjusted, the task of being arranged in the second place originally will be positioned at the primary importance of Fig. 1, so the detection reference position after the renewal is arranged in the primary importance of formation, promptly originally be positioned at the new position of the task of the second place.Preferably, if the current detection reference position has been positioned at the position, back of task queue, wait for that updating device 13 can't find the position (because of it does not exist) of first quantity task after the current detection reference position, then can be with the head of the queue of task queue as new detection reference position.
It will be understood by those skilled in the art that in the above described manner and upgrade, can avoid having handled in the cycle in current detection in a large number of task is carried out duplicate detection detecting reference position.After the time cycle of waiting for a predetermined length, repeating step S10.
Be appreciated that by position adjust of 11 pairs of treated tasks of adjusting gear in task queue can with current detection in the cycle executing state in the high task-set of the task of " not finishing " and occurrence frequency at the specific part of task queue.Concentrate by pick-up unit 10 afterwards this specific part is detected, and detected waiting task is handled, can improve the execution efficient of system.If do not detect waiting task, exist the probability of need handling in the cycle in current detection of task then less relatively in the then whole task queue at this specific part.At this moment, the time cycle of waiting for a predetermined length can be reduced detection taking system resource.
Below in conjunction with Fig. 5 and with reference to Fig. 4 another preferred embodiment of the present invention is further described, wherein, Fig. 5 is the task execution device block diagram according to one embodiment of the present invention, Fig. 4 is the synoptic diagram of the task queue of an embodiment according to the present invention, and the following description of above-mentioned another preferred embodiment is based on the hypothesis of following indefiniteness:
1) operating system of subscriber equipment is embedded system;
2) corresponding and continuous other subscriber equipment (promptly with between the current user equipment being connected the subscriber equipment that can exchange information, IP address that each subscriber equipment is corresponding unique) of this subscriber equipment of each task by effective communication link;
3) can comprise one or more incidents in a task, and this incident can send data for other subscriber equipment to its correspondence or receive data by other subscriber equipment of its correspondence;
4) under original state, all tasks in the task queue all are waiting task;
5) comprise 26 tasks of a-z as shown in Figure 4 in the task queue successively;
6) in a sense cycle, 4 tasks are detected (value of first quantity is 4);
7) operating system is single-threaded (promptly the value of second quantity is 1);
8) only comprise an incident in each task.
At first, task a, b, c and d that 10 pairs of pick-up units are in the 1-4 of queue position detect, and a that sets the tasks, b, c are the waiting task in current detection cycle.Then, execution is handled and finished to adjusting gear 11 to task a earlier, and therefore the executing state with task a is updated to " finishing ", and task a is placed the end of task queue, promptly after the task z.Know that by control device to be still waiting Processing tasks not processed then, then 11 couples of task b of adjusting gear repeat aforesaid operations, and the executing state with task b is updated to " finishing " equally, is placed on the end of task queue, promptly after the task a.
Then, the 1 couple of task c of task execution device repeats aforesaid operations, suppose task c for to its relative user equipment I transmission 500k data, and owing to the reason of network congestion, temporarily can't continue to carry out after the data transmission of finishing 50k, then the executing state with task c is updated to " not finishing ".And be placed on task queue foremost, promptly be positioned at before the d.Therefore, when the current detection end cycle, task queue four tasks foremost are followed successively by cdef, and four tasks of least significant end are as yzab (task b is last task of task queue).Because there is executing state in current detection in the cycle is the waiting task of " not finishing ", therefore upgrading the detection reference position is the head of the queue of task queue.After finishing aforesaid operations, enter next sense cycle (, next sense cycle being defined as second sense cycle) at once for ease of describing.
In second sense cycle, 10 pairs of pick-up units detect 4 tasks of reference position (residing position task c this moment) beginning: task cdef detects, wherein all tasks are the waiting task in current detection cycle, and then adjusting gear 11 and control device 12 are handled one by one to it.Suppose task c because network congestion still can't be carried out subsequent treatment, then this task is continued to place task queue foremost, task d and e are processed and finish execution, then are placed on tail of the queue, promptly after the task b.Suppose that task f sends Query Information to port II, and, then task f is placed head of the queue because network congestion can't be finished execution.Therefore, when second sense cycle finished, task queue four tasks foremost were followed successively by fcgh, and six tasks of least significant end are followed successively by yzabde (task b is last task of task queue).
In the 3rd sense cycle, 10 couples of task fcgh of pick-up unit detect.By that analogy, after a plurality of sense cycle, suppose that network recovery is unimpeded, then task f gets and finishes, and is placed in the afterbody of task queue after finishing; If port I still closes, then task c still is in the forward position of task queue, can obtain more machines meeting.
It will be appreciated by those skilled in the art that in the process of Network Transmission the breakpoint transmission of data block sends and receives and can not once finish is multiple incident in the transmission course, also need to be divided into a plurality of file fragmentations for the big data block of needs transmission and operate.Therefore, one sends or corresponding other subscriber equipment that carries out data interaction with this subscriber equipment of reception incident, and one with this subscriber equipment carry out other subscriber equipment of data interaction often corresponding a plurality of transmissions or reception incident.Since between individual event and other user device IP address for concerning one to one, those skilled in the art understand, through after a plurality of sense cycle, be adjusted to forward position in the task queue at this subscriber equipment place with the mutual more pairing task of other subscriber equipment of this subscriber equipment.If detecting reference position is the head of the queue of task queue, then corresponding with mutual more other subscriber equipment of this subscriber equipment task can obtain the chance of how detected processing, need not wait for that the scanning to all user device IP address can preferentially realize its data transmission.And and the pairing task of other subscriber equipment of temporary transient no datat piece transmission between this subscriber equipment be placed in the end of task queue, promptly reduced priority.
Understand easily, transmission network as network based on the point-to-point transmission agreement in, this subscriber equipment may link to each other with a plurality of other subscriber equipmenies, but may be only carries out data interaction (for example, send or receiver, video voice data) with several other subscriber equipmenies wherein in a certain particular moment.By the position of adjustment task in formation, can improve the detected probability of handling of the pairing task of other subscriber equipment that has data interaction with this subscriber equipment, thereby the minimizing temporary file takies buffer memory, saves the time of detecting and waiting for, improves transmission of data blocks efficient simultaneously.
More than specific embodiments of the invention are described.It will be appreciated that the present invention is not limited to above-mentioned specific implementations, those skilled in the art can make various distortion or modification within the scope of the appended claims.
Claims (15)
1. task executing method based on operating system wherein, may further comprise the steps:
A. detect since first quantity task of the detection reference position of a task queue and whether comprise one or more waiting tasks;
If comprise one or more waiting tasks in b. described first quantity task, then second quantity waiting task in described one or more waiting tasks is handled, obtaining second quantity treated task, and adjust its position in described task queue according to described each treated task executions state;
C. repeating described step b all handles until described one or more waiting tasks.
2. method according to claim 1 is characterized in that, described step c also comprises:
-all handle when described one or more waiting tasks, upgrade described detection reference position and repeat described step a.
3. method according to claim 2 is characterized in that, also comprises after the described step a:
If do not comprise one or more waiting tasks in first quantity task that the detection reference position of-described task queue begins, after then waiting for the time cycle of a predetermined length, with the position of first quantity task after the described detection reference position as new detection reference position, repeating said steps a.
4. according to claim 2 or 3 described methods, it is characterized in that described step b also comprises:
-adjust its position in described task queue according to described each treated task executions state, so that uncompleted each treated task is in before completed each treated task in described task queue.
5. method according to claim 4 is characterized in that, described step b also comprises:
-according to described each treated task executions state uncompleted each treated task is placed the head of the queue of described task queue, completed each treated task is placed the end of described task queue.
6. according to each described method among the claim 1-5, it is characterized in that described detection reference position is fixed and is set to the head of the queue of described task queue.
7. according to each described method among the claim 1-6, it is characterized in that described operating system is embedded system.
8. the task execution device based on operating system is characterized in that, comprising:
Pick-up unit is used for detecting since first quantity task of the detection reference position of a task queue whether comprise one or more waiting tasks;
Adjusting gear, if comprise one or more waiting tasks in described first quantity task, then described adjusting gear is used for second quantity waiting task of described one or more waiting tasks is handled, obtaining second quantity treated task, and adjust its position in described task queue according to described each treated task executions state;
Control device is used to judge whether that described one or more waiting task all handles, if untreated waiting task is arranged, then controls described adjusting gear described untreated waiting task is operated.
9. task execution device according to claim 8 is characterized in that, described control device also comprises the control updating device:
Described control updating device is used for upgrade described detection reference position, and the control detection device being operated when described one or more waiting tasks have all been handled.
10. task execution device according to claim 9 is characterized in that, also comprises the wait updating device:
When described wait updating device is used for first quantity task that the detection reference position in described task queue begins and does not comprise one or more waiting task, after waiting for the time cycle of a predetermined length, as new detection reference position, and the control detection device is operated with the position of first quantity task after the described detection reference position.
11. according to claim 9 or 10 described task execution devices, it is characterized in that, described adjusting gear also is used for adjusting its position in described task queue according to described each treated task executions state, so that uncompleted each treated task is in described task queue before completed each treated task.
12. task execution device according to claim 11, it is characterized in that, described adjusting gear also is used for according to described each treated task executions state completed each treated task being placed the end of described task queue, and uncompleted each treated task is placed the head of the queue of described task queue.
13. each described task execution device is characterized in that according to Claim 8-12, described detection reference position is fixed and is set to the head of the queue of described task queue.
14. each described task execution device is characterized in that according to Claim 8-13, described operating system is embedded system.
15. the equipment based on operating system is characterized in that, comprises each described task execution device based on operating system among the claim 8-14.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810036120 CN101561767B (en) | 2008-04-16 | 2008-04-16 | Method and device for executing tasks based on operating system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810036120 CN101561767B (en) | 2008-04-16 | 2008-04-16 | Method and device for executing tasks based on operating system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101561767A true CN101561767A (en) | 2009-10-21 |
| CN101561767B CN101561767B (en) | 2012-06-06 |
Family
ID=41220579
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200810036120 Expired - Fee Related CN101561767B (en) | 2008-04-16 | 2008-04-16 | Method and device for executing tasks based on operating system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101561767B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102279730A (en) * | 2010-06-10 | 2011-12-14 | 阿里巴巴集团控股有限公司 | Parallel data processing method, device and system |
| CN103034475A (en) * | 2011-10-08 | 2013-04-10 | ***通信集团四川有限公司 | Distributed parallel computing method, device and system |
| CN107104822A (en) * | 2017-03-29 | 2017-08-29 | 杭州云纪网络科技有限公司 | Server preparedness processing method, device, storage medium and electronic equipment |
| CN107370639A (en) * | 2017-08-14 | 2017-11-21 | 苏州众天力信息科技有限公司 | A kind of more scenery control methods of gateway based on trapezoidal temporal algorithm |
| CN108009022A (en) * | 2017-11-06 | 2018-05-08 | 联动优势科技有限公司 | A kind of message treatment method and server |
| CN108021439A (en) * | 2016-10-31 | 2018-05-11 | 阿里巴巴集团控股有限公司 | A kind of task polling method, task arrangement method and relevant apparatus |
| CN112395067A (en) * | 2020-12-04 | 2021-02-23 | 深圳前海微众银行股份有限公司 | Task scheduling method, system, device and medium |
| CN113806042A (en) * | 2021-08-25 | 2021-12-17 | 北京市遥感信息研究所 | Task scheduling method of multi-core real-time embedded system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100349149C (en) * | 2003-01-24 | 2007-11-14 | 中兴通讯股份有限公司 | Process scheduling method in embedded type real time operating system |
| CN1801101A (en) * | 2006-01-17 | 2006-07-12 | 浙江大学 | Thread implementation and thread state switching method in Java operation system |
-
2008
- 2008-04-16 CN CN 200810036120 patent/CN101561767B/en not_active Expired - Fee Related
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102279730B (en) * | 2010-06-10 | 2014-02-05 | 阿里巴巴集团控股有限公司 | Parallel data processing method, device and system |
| CN102279730A (en) * | 2010-06-10 | 2011-12-14 | 阿里巴巴集团控股有限公司 | Parallel data processing method, device and system |
| CN103034475A (en) * | 2011-10-08 | 2013-04-10 | ***通信集团四川有限公司 | Distributed parallel computing method, device and system |
| CN103034475B (en) * | 2011-10-08 | 2015-11-25 | ***通信集团四川有限公司 | Distributed Parallel Computing method, Apparatus and system |
| CN108021439B (en) * | 2016-10-31 | 2022-07-15 | 阿里巴巴集团控股有限公司 | Task polling method, task arrangement method and related device |
| CN108021439A (en) * | 2016-10-31 | 2018-05-11 | 阿里巴巴集团控股有限公司 | A kind of task polling method, task arrangement method and relevant apparatus |
| CN107104822A (en) * | 2017-03-29 | 2017-08-29 | 杭州云纪网络科技有限公司 | Server preparedness processing method, device, storage medium and electronic equipment |
| CN107104822B (en) * | 2017-03-29 | 2020-09-08 | 恒生电子股份有限公司 | Server disaster recovery processing method and device, storage medium and electronic equipment |
| CN107370639A (en) * | 2017-08-14 | 2017-11-21 | 苏州众天力信息科技有限公司 | A kind of more scenery control methods of gateway based on trapezoidal temporal algorithm |
| CN108009022A (en) * | 2017-11-06 | 2018-05-08 | 联动优势科技有限公司 | A kind of message treatment method and server |
| CN112395067A (en) * | 2020-12-04 | 2021-02-23 | 深圳前海微众银行股份有限公司 | Task scheduling method, system, device and medium |
| CN113806042A (en) * | 2021-08-25 | 2021-12-17 | 北京市遥感信息研究所 | Task scheduling method of multi-core real-time embedded system |
| CN113806042B (en) * | 2021-08-25 | 2023-06-16 | 北京市遥感信息研究所 | Task scheduling method of multi-core real-time embedded system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101561767B (en) | 2012-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101561767B (en) | Method and device for executing tasks based on operating system | |
| CN106557369B (en) | Multithreading management method and system | |
| US11558253B2 (en) | Data processing method and apparatus, and computing node for updating container images | |
| CN103346974B (en) | Controlling method of service process and network device | |
| KR950000506B1 (en) | Producing management method and producing apparatus and manufacturing apparatus in semiconductor producing system | |
| CN101211277B (en) | Method of selecting one of execution schedules of guest OS and virtual machine monitor employing the method | |
| CN103532823A (en) | Message processing method, device and system for Internet of Things | |
| CN1818933A (en) | Rf-tag reading system and rf-tag reader/writer control system, and interference avoidance method therefor | |
| CN101135981A (en) | Method and device for realizing batch report generation | |
| CN102455940B (en) | Processing method and system of timers and asynchronous events | |
| CN103019805A (en) | Timer setting method under multithreading environment | |
| CN105786600A (en) | Task scheduling method and device | |
| CN103279331A (en) | Multi-task concurrent processing method and device for Android system | |
| US7893836B2 (en) | Method and apparatus for avoiding collision between each of radio frequency identification readers | |
| JP2008152618A (en) | Job assignment program, method and device | |
| CN1859184A (en) | Method for increasing communication link layer protocol efficiency | |
| CN111722944B (en) | NIO-based AIRT-ROS communication method and system | |
| CN111158875A (en) | Multi-module-based multi-task processing method, device and system | |
| CN108509257B (en) | Message processing method and device based on multithreading | |
| CN112711522A (en) | Docker-based cloud testing method and system and electronic equipment | |
| CN101502149B (en) | A network managing system for completing customized operation automatically and a method thereof | |
| CN102508720A (en) | Method for improving efficiency of preprocessing module and efficiency of post-processing module and system | |
| US20190340034A1 (en) | Device and method for processing jobs | |
| US20170142227A1 (en) | Data processing apparatus and data processing method | |
| CN101320443A (en) | Electronic work order processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120606 Termination date: 20200416 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |