CN106020988A - Off-line task scheduling method and device for intelligent video monitoring system - Google Patents
Off-line task scheduling method and device for intelligent video monitoring system Download PDFInfo
- Publication number
- CN106020988A CN106020988A CN201610391312.9A CN201610391312A CN106020988A CN 106020988 A CN106020988 A CN 106020988A CN 201610391312 A CN201610391312 A CN 201610391312A CN 106020988 A CN106020988 A CN 106020988A
- Authority
- CN
- China
- Prior art keywords
- task
- time
- processor
- processed
- scheduling
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000012544 monitoring process Methods 0.000 title claims abstract description 53
- 238000012545 processing Methods 0.000 claims abstract description 83
- 230000008569 process Effects 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000009471 action Effects 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000003637 basic solution Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
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
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5044—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering hardware capabilities
-
- 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
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The embodiment of the invention discloses an off-line task scheduling method and device for an intelligent video monitoring system. The off-line task scheduling method for the intelligent video monitoring system is applied to a server in the video monitoring system. The video monitoring system further comprises a plurality of processors used for processing off-line tasks of the intelligent video monitoring system and a processing capability model pre-established for the processors. The method comprises the steps that an off-line task request of the intelligent video monitoring system is received, and the task request comprises at least one task to be processed; the weighting time length of each task to be processed is calculated; according to the processing capability model, the time needed for processing all the tasks to be processed at present of each processor is calculated; according to the weighting time length and the time, the tasks to be processed in the task request are scheduled to the processors. By means of the off-line scheduling method and device, the loads of the processors can be balanced..
Description
Technical Field
The invention relates to the technical field of offline tasks of intelligent video monitoring systems, in particular to an offline task scheduling method and device of an intelligent video monitoring system.
Background
In the face of a large number of monitoring videos, although an Intelligent Video Analysis (IVA) technology is used in a conventional offline video processing system, due to the limitation of a single-machine computing capability, a task of processing the monitoring videos within a time required by a client cannot be completed normally, and therefore a distributed processing technology is introduced. In distributed computing, task management, task scheduling and resource management are three major basic functions of a system, wherein the task scheduling is an important aspect influencing the performance of the system. Task scheduling is an important component of operating systems, and for real-time operating systems, task scheduling directly affects their real-time performance. Task scheduling algorithms can be divided into events: a driving scheduling algorithm, which is used for scheduling the execution of the tasks according to the sequence of the events and the priority of the tasks; clock driven scheduling algorithms, typically used for periodic tasks.
The traditional distributed task scheduling methods comprise Min-Min, Max-Min and the like. The idea of the Min-Min scheduling algorithm is to allocate each task as soon as possible to the earliest available and fastest performing processor. The Min-Min algorithm is based on the minimum completion time, and it is considered in each mapping that all unallocated tasks allocate the tasks to the fastest executing processor each time, which easily causes a severe imbalance in resource load and makes the resources not fully utilized.
Most of the traditional distributed task scheduling methods are knowledge deterministic algorithms, and require a globally deterministic knowledge of system resources and user tasks. Because the running time of the intelligent video processing task cannot be known in advance, and the processing capacity of the intelligent video analysis server (IVU) changes along with the change of time, it can be seen that when the traditional distributed task scheduling method is applied to an intelligent video monitoring system, the cluster load of a processor is unbalanced, the resource utilization is insufficient, and the current intelligent monitoring video processing task requirement cannot be met.
Disclosure of Invention
The embodiment of the invention aims to provide an offline task scheduling method and device for an intelligent video monitoring system, so that the load of a processor is balanced.
In order to achieve the above object, an embodiment of the present invention discloses an offline task scheduling method for an intelligent video monitoring system, which is applied to a server in the intelligent video monitoring system, wherein the intelligent video monitoring system further includes a plurality of processors for processing offline tasks of the intelligent video monitoring system, and a processing capability model for the processors is pre-established, and the method includes:
receiving an offline task request of an intelligent video monitoring system, wherein the task request comprises at least one task to be processed;
calculating the weighted time length of each task to be processed;
calculating the time required by each processor to process all tasks to be processed currently according to the processing capacity model;
and scheduling the tasks to be processed in the task requests to a processor according to the weighted time length and the time.
Preferably, the processing capability model is:
wherein, theFor a weighted length of time of a task currently to be processed by processor i, said vlastFor a pre-recorded latest calculation speed of processor i, saidThe t is the time required for the processor i to process the task.
Preferably, the scheduling the to-be-processed task in the task request to the processor according to the weighted time length and the time includes:
judging whether the time required for processing all the tasks to be processed currently by each processor is not more than a preset first threshold or not;
and if so, scheduling the task to be processed corresponding to the maximum weighted time length to the processor with the minimum required time.
Preferably, the scheduling the to-be-processed task in the task request to the processor according to the weighted time length and the time further includes:
under the condition that the time required by each processor to process all the tasks to be processed is greater than the preset first threshold value, calculating the difference value between the maximum required time and the minimum required time;
judging whether the difference value is larger than a preset second threshold value or not;
and if so, dispatching the task corresponding to the weighted time length closest to the difference value to the processor corresponding to the minimum required time.
Preferably, the weighted time length is:
wherein, the J isLFor the weighted time length of each task to be processed, n is the number of data blocks contained in each task, and L isiFor the time length of the ith data block, Li=qm*t,i∈[1,n]Said q ismAnd dividing the weight value for the quality corresponding to the ith data block, wherein t is the difference between the termination time and the start time of the data block.
In order to achieve the above object, the present invention further discloses an offline task scheduling device for an intelligent video surveillance system, which is applied to a server in the intelligent video surveillance system, wherein the intelligent video surveillance system further comprises a plurality of processors for processing offline tasks of the intelligent video surveillance system, a processing capability model for the processors is pre-established, and the device comprises:
the system comprises a receiving module, a processing module and a processing module, wherein the receiving module is used for receiving an offline task request of an intelligent video monitoring system, and the task request comprises at least one task to be processed;
the first calculation module is used for calculating the weighted time length of each task to be processed;
the second calculation module is used for calculating the time required by each processor to process all tasks to be processed currently according to the processing capacity model;
and the scheduling module is used for scheduling the tasks to be processed in the task requests to the processor according to the weighted time length and the time.
Preferably, the processing capability model is:
wherein, theFor a weighted length of time of a task currently to be processed by processor i, said vlastFor a pre-recorded latest calculation speed of processor i, saidThe t is the time required for the processor i to process the task.
Preferably, the scheduling module includes:
the first judging unit is used for judging whether the time required for processing all the tasks to be processed currently of the first judging unit is not more than a preset first threshold value or not for each processor;
and the first scheduling unit is used for scheduling the task to be processed corresponding to the maximum weighted time length to the processor with the minimum required time under the condition that the judgment result of the first judging unit is yes.
Preferably, the scheduling module further includes:
the calculation unit is used for calculating the difference value between the maximum required time and the minimum required time under the condition that the time required by each processor to process all the tasks to be processed is greater than the preset first threshold value;
the second judgment unit is used for judging whether the difference value is larger than a preset second threshold value or not;
and a second scheduling unit, configured to schedule, when the determination result of the second determining unit is yes, the task corresponding to the weighted time length closest to the difference value to the processor corresponding to the minimum required time.
Preferably, the weighted time length is:
wherein, the J isLFor the weighted time length of each task to be processed, n is the number of data blocks contained in each task, and L isiFor the time length of the ith data block, Li=qm*t,i∈[1,n]Said q ismAnd dividing the weight value for the quality corresponding to the ith data block, wherein t is the difference between the termination time and the start time of the data block.
According to the technical scheme, the offline task scheduling method and device for the intelligent video monitoring system, provided by the embodiment of the invention, are used for receiving an offline task request of the intelligent video monitoring system, wherein the task request comprises at least one task to be processed; calculating the weighted time length of each task to be processed; calculating the time required by each processor to process all tasks to be processed currently according to the processing capacity model; and scheduling the tasks to be processed in the task requests to a processor according to the weighted time length and the time.
Therefore, the processing capacity model for the processor is established in advance, and task scheduling is carried out according to the model, so that the load of the processor is balanced, the resources of the processor are fully utilized, the task is processed efficiently, and the advantages of distributed computing are fully exerted.
Of course, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of an offline task scheduling method for an intelligent video monitoring system according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an offline task scheduling device of an intelligent video monitoring system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses an offline task scheduling method and device of an intelligent video monitoring system, which are respectively explained in detail below.
It should be noted that the embodiment of the present invention is preferably applicable to a server in an intelligent video monitoring system, and in practical applications, the intelligent video monitoring system further includes a plurality of processors for processing offline tasks of the intelligent video monitoring system, and a processing capability model for the processors is pre-established.
Referring to fig. 1, fig. 1 is a schematic flowchart of an offline task scheduling method of an intelligent video monitoring system according to an embodiment of the present invention, including the following steps:
s101, receiving an offline task request of an intelligent video monitoring system, wherein the task request comprises at least one task to be processed;
the to-be-processed task is an offline task of the intelligent video monitoring system to be processed.
Wherein, the server can be: the intelligent video processing algorithm scheduling server, the processor can be: intelligent video analytics server (IVU), the task request may be: and the intelligent video monitoring system carries out offline tasks and processing requests of users.
In one particular implementation, a user submits an intelligent surveillance video and a processing request. According to a processing request of a user, dividing an intelligent video into a plurality of data blocks, for example, to avoid communication between IVUs, a plurality of related data blocks are combined into an intelligent video monitoring system offline task, for example, a video is divided into 39 data blocks, and 3 data blocks which are adjacent in sequence from the beginning to the end of the video are combined into one task, and 13 tasks are total. The server receives task requests, wherein the task requests comprise 13 offline tasks of the intelligent video monitoring system to be processed.
S102, calculating the weighted time length of each task to be processed;
wherein, the weighted time length of each task to be processed is as follows:
wherein, JLFor the weighted time length of each task to be processed, n is the number of data blocks contained in each task to be processed, LiIs the weighted time length, L, of the ith data blocki=qm*t,i∈[1,n],qmAnd dividing an initial weight for the quality corresponding to the ith data block, wherein t is the difference between the termination time and the start time of the data block. In practical application, the intelligent monitoring video can be subjected to quality division according to a processing request of a user, and further the quality division initial weight of each data block is obtained.
S103, calculating the time required by each processor to process all tasks to be processed currently according to the processing capacity model;
wherein the processing capability model for the processor is:
wherein,for a weighted length of time, v, of a task currently to be processed by processor ilastFor a pre-recorded latest calculation speed of processor i,the trend of the processing power of the processor i is shown, and t is the time required for the processor i to process the task.
In practical application, according to a pre-established processing capability model, the time required by the processor i to process each task of all tasks to be processed currently can be calculated, and the time required by the processor i to process all tasks to be processed currently can be further obtained.
The process of pre-establishing the processing capability model for the processor may be:
after receiving an offline task processing request of an intelligent video monitoring system submitted by a user for the first time, calculating the weighted time length of each task to be processed, and adding each task into a task queue according to the sequence of size, wherein in practical application, the task queue may or may not exist, and particularly, the purpose of establishing a processing capability model is achieved;
the server schedules the first eta x tau tasks in the task queue to tau processors for processing, and the processors feed back actual processing time to the server after processing an off-line task of the intelligent video monitoring system, wherein eta is a positive-even integer threshold value, controls the number of the first scheduling tasks, and can be automatically set according to the total number of the tasks to be processed, for example: the first 9 tasks are dispatched to the 3 processors A, B and C for processing, each processor containing 3 tasks;
according to the weighted time length of the task and the actual processing time required by processing, an n-element linear equation set is established, wherein the equation set is as follows:
the coefficient matrix is:
wherein, TnActual processing time, t, required for the processor to process the nth tasknnIs the start-stop time difference of the nth data block in the nth task.
The solution of the n-ary system of equations is:
qm=ξ1+ξ2+…+ξn-r
ξ1、ξ2、ξn-ris a basic solution system of an n-element linear equation set, r ═ r (d);
calculate each video quality partition qmAnd for each qmSetting a set for storing the result of each calculation, and taking the average value of data in the set as qmThe accurate weight of (2);
calculating the processing speed V of each processor, and putting the processing speed of the processor which is the last three times into a set M, wherein V is JLT, said M ═ Vt1,Vt2,Vt3},t1、t2、t3Representing the moment at which the processing speed is calculated, T being the actual processing time required by the processor to process a task;
calculating the processing acceleration a of each processor according to the set M as the variation trend of the processing capacity of the processor;
a processing capability model for each processor is established based on the weighted time length of the task, the processing speed and the processing acceleration of the processor.
And S104, scheduling the task to be processed in the task request to a processor according to the weighted time length and the time.
Specifically, the to-be-processed tasks in the task request are dispatched to the processors according to the weighted time length and the weighted time, and whether the time required for processing all the current to-be-processed tasks of each processor is not more than a preset first threshold value or not can be judged for each processor; if so, if the server still has the tasks to be processed which are not scheduled to the processor, the tasks to be processed corresponding to the maximum weighted time length are scheduled to the processor with the minimum required time.
Specifically, in practical application, under the condition that the time required by each processor to process all the tasks currently to be processed is greater than a preset first threshold, the difference between the maximum required time and the minimum required time can be calculated; judging whether the difference value is larger than a preset second threshold value or not; if so, if the server still has the tasks to be processed which are not scheduled to the processor, the task corresponding to the weighted time length closest to the difference value is scheduled to the processor corresponding to the minimum required time.
Illustratively, in a specific implementation, the video surveillance system includes 3 processors A, B, C for processing video tasks, the corresponding time required for processing all the tasks is 10s, 15s, and 18s, respectively, the preset first threshold is 13s, and the preset second threshold is 6 s. There are also unscheduled pending tasks 1, 2, 3 and 4 in the server, with corresponding weighted time lengths of 12s, 7.5s, 9s and 10s, respectively. When the time required by the processor A is not more than a preset first threshold value, the server dispatches the task 1 with the largest weighted time length to the processor A for processing; aiming at the condition that the time required by the processor B is greater than a preset first threshold value, calculating the difference between the maximum required time 18s and the minimum required time 10s to be 8s, wherein the difference is greater than a preset second threshold value, the server also has unscheduled tasks to be processed 2, 3 and 4, and knowing that the weighted time length 7.5s of the task to be processed 2 is closest to the difference 8s, the server schedules the unscheduled task to be processed 2 to the processor B for processing; and aiming at the condition that the time required by the processor C is greater than a preset first threshold, calculating the difference between the maximum task time 18s and the minimum task time 10s to be 8s, wherein the difference is greater than a preset second threshold, the server also comprises unscheduled tasks 3 and 4 to be processed, the corresponding weighted time lengths are respectively 9s and 10s, and the server schedules the unscheduled task 3 to be processed to the processor C for processing if the weighted time length of the task 3 to be processed is closest to the difference. And continuing to execute the next round of scheduling until no unscheduled tasks to be processed exist in the server.
Therefore, the processing capacity model for the processor is established in advance, and the task scheduling is carried out according to the processing capacity model, so that the load of the processor can be balanced, the resources of the processor can be fully utilized, the task can be efficiently processed, and the advantages of distributed computing can be fully exerted.
Referring to fig. 2, fig. 2 is a schematic structural diagram of an offline task scheduling device of an intelligent video monitoring system according to an embodiment of the present invention, where the task scheduling device is applied to a server in the intelligent video monitoring system, the intelligent video monitoring system further includes a plurality of processors for processing offline tasks of the intelligent video monitoring system, a processing capability model for the processors is pre-established, and the task scheduling device corresponds to the flow shown in fig. 1, and may include a receiving module 201, a first calculating module 202, a second calculating module 203, and a scheduling module 204.
The receiving module 201 is configured to receive an offline task request of the intelligent video monitoring system, where the task request includes at least one task to be processed;
a first calculating module 202, configured to calculate a weighted time length of each to-be-processed task;
the second calculating module 203 is configured to calculate, according to the processing capability model, time required by each processor to process all tasks currently to be processed;
and the scheduling module 204 is configured to schedule the to-be-processed task in the task request to the processor according to the weighted time length and the time.
Specifically, the processing capability model is as follows:
wherein,for a weighted length of time, v, of a task currently to be processed by processor ilastFor a pre-recorded latest calculation speed of processor i,the trend of the processing power of the processor i is shown, and t is the time required for the processor i to process the task.
Specifically, the scheduling module may include: a first judging unit and a first scheduling unit (not shown in the figure);
the first judging unit is used for judging whether the time required for processing all the tasks to be processed currently of the first judging unit is not more than a preset first threshold value or not for each processor;
and the first scheduling unit is used for scheduling the task to be processed corresponding to the maximum weighted time length to the processor with the minimum required time under the condition that the judgment result of the first judging unit is yes.
Specifically, the scheduling module may further include: a calculating unit, a second judging unit and a second scheduling unit (not shown in the figure);
the calculation unit is used for calculating the difference value between the maximum required time and the minimum required time under the condition that the time required by each processor to process all the tasks to be processed is greater than the preset first threshold value;
the second judgment unit is used for judging whether the difference value is larger than a preset second threshold value or not;
and a second scheduling unit, configured to schedule, when the determination result of the second determining unit is yes, the task corresponding to the weighted time length closest to the difference value to the processor corresponding to the minimum required time.
Specifically, the weighting time length is as follows:
wherein, the J isLFor said n is the number of data blocks contained by each of said tasks, said LiFor the time length of the ith data block, Li=qm*t,i∈[1,n]Said q ismAnd dividing the weight value for the quality corresponding to the ith data block, wherein t is the difference between the termination time and the start time of the data block.
Therefore, the processing capacity model for the processor is established in advance, and task scheduling is carried out according to the model, so that the load of the processor is balanced, the resources of the processor are fully utilized, the task is processed efficiently, and the advantages of distributed computing are fully exerted.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
Those skilled in the art will appreciate that all or part of the steps in the above method embodiments may be implemented by a program to instruct relevant hardware to perform the steps, and the program may be stored in a computer-readable storage medium, which is referred to herein as a storage medium, such as: ROM/RAM, magnetic disk, optical disk, etc.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. An off-line task scheduling method of an intelligent video monitoring system is applied to a server in the intelligent video monitoring system, and is characterized in that the intelligent video monitoring system also comprises a plurality of processors for processing off-line tasks of the intelligent video monitoring system, and a processing capacity model aiming at the processors is pre-established; the method comprises the following steps:
receiving an offline task request of an intelligent video monitoring system, wherein the task request comprises at least one task to be processed;
calculating the weighted time length of each task to be processed;
calculating the time required by each processor to process all tasks to be processed currently according to the processing capacity model;
and scheduling the tasks to be processed in the task requests to a processor according to the weighted time length and the time.
2. The method of claim 1, wherein the processing power model is:
wherein,for a weighted length of time, v, of a task currently to be processed by processor ilastFor a pre-recorded latest calculation speed of processor i,the trend of the processing power of the processor i is shown, and t is the time required for the processor i to process the task.
3. The method of claim 1, wherein said scheduling the pending task in the task request to the processor according to the weighted time duration and the time comprises:
judging whether the time required for processing all the tasks to be processed currently by each processor is not more than a preset first threshold or not;
and if so, scheduling the task to be processed corresponding to the maximum weighted time length to the processor with the minimum required time.
4. The method of claim 3, wherein the scheduling the pending task in the task request to the processor according to the weighted time duration and the time further comprises:
under the condition that the time required by each processor to process all the tasks to be processed is greater than the preset first threshold value, calculating the difference value between the maximum required time and the minimum required time;
judging whether the difference value is larger than a preset second threshold value or not;
and if so, dispatching the task corresponding to the weighted time length closest to the difference value to the processor corresponding to the minimum required time.
5. The method of claim 1, wherein the weighted duration is:
wherein, the J isLFor the weighted time length of each task to be processed, n is the number of data blocks contained in each task, and L isiFor the time length of the ith data block, Li=qm*t,i∈[1,n]Said q ismAnd dividing the weight value for the quality corresponding to the ith data block, wherein t is the difference between the termination time and the start time of the data block.
6. An intelligent video monitoring system offline task scheduling device is applied to a server in an intelligent video monitoring system and is characterized in that the intelligent video monitoring system also comprises a plurality of processors for processing the offline tasks of the intelligent video monitoring system, and a processing capacity model aiming at the processors is pre-established; the device comprises:
the system comprises a receiving module, a processing module and a processing module, wherein the receiving module is used for receiving an offline task request of an intelligent video monitoring system, and the task request comprises at least one task to be processed;
the first calculation module is used for calculating the weighted time length of each task to be processed;
the second calculation module is used for calculating the time required by each processor to process all tasks to be processed currently according to the processing capacity model;
and the scheduling module is used for scheduling the tasks to be processed in the task requests to the processor according to the weighted time length and the time.
7. The apparatus of claim 6, wherein the processing power model is:
wherein,for a weighted length of time, v, of a task currently to be processed by processor ilastFor a pre-recorded latest calculation speed of processor i,is a processori the trend of the processing power, t is the time required for the processor i to process the task.
8. The apparatus of claim 6, wherein the scheduling module comprises:
the first judging unit is used for judging whether the time required for processing all the tasks to be processed currently of the first judging unit is not more than a preset first threshold value or not for each processor;
and the first scheduling unit is used for scheduling the task to be processed corresponding to the maximum weighted time length to the processor with the minimum required time under the condition that the judgment result of the first judging unit is yes.
9. The apparatus of claim 8, wherein the scheduling module further comprises:
the calculation unit is used for calculating the difference value between the maximum required time and the minimum required time under the condition that the time required by each processor to process all the tasks to be processed is greater than the preset first threshold value;
the second judgment unit is used for judging whether the difference value is larger than a preset second threshold value or not;
and a second scheduling unit, configured to schedule, when the determination result of the second determining unit is yes, the task corresponding to the weighted time length closest to the difference value to the processor corresponding to the minimum required time.
10. The apparatus of claim 6, wherein the weighted duration is:
wherein, the J isLFor each weighted time length of the task to be processedN is the number of data blocks included in each task, and L isiFor the time length of the ith data block, Li=qm*t,i∈[1,n]Said q ismAnd dividing the weight value for the quality corresponding to the ith data block, wherein t is the difference between the termination time and the start time of the data block.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610391312.9A CN106020988B (en) | 2016-06-03 | 2016-06-03 | A kind of offline method for scheduling task of intelligent video monitoring system and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610391312.9A CN106020988B (en) | 2016-06-03 | 2016-06-03 | A kind of offline method for scheduling task of intelligent video monitoring system and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106020988A true CN106020988A (en) | 2016-10-12 |
CN106020988B CN106020988B (en) | 2019-03-15 |
Family
ID=57089481
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610391312.9A Active CN106020988B (en) | 2016-06-03 | 2016-06-03 | A kind of offline method for scheduling task of intelligent video monitoring system and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106020988B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110545242A (en) * | 2018-05-29 | 2019-12-06 | 杭州海康威视数字技术股份有限公司 | target analysis method and intelligent analysis equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102831012A (en) * | 2011-06-16 | 2012-12-19 | 日立(中国)研究开发有限公司 | Task scheduling device and task scheduling method in multimode distributive system |
CN102902587A (en) * | 2011-07-28 | 2013-01-30 | 中国移动通信集团四川有限公司 | Distribution type task scheduling method, distribution type task scheduling system and distribution type task scheduling device |
CN103226467A (en) * | 2013-05-23 | 2013-07-31 | 中国人民解放军国防科学技术大学 | Data parallel processing method and system as well as load balancing scheduler |
-
2016
- 2016-06-03 CN CN201610391312.9A patent/CN106020988B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102831012A (en) * | 2011-06-16 | 2012-12-19 | 日立(中国)研究开发有限公司 | Task scheduling device and task scheduling method in multimode distributive system |
CN102902587A (en) * | 2011-07-28 | 2013-01-30 | 中国移动通信集团四川有限公司 | Distribution type task scheduling method, distribution type task scheduling system and distribution type task scheduling device |
CN103226467A (en) * | 2013-05-23 | 2013-07-31 | 中国人民解放军国防科学技术大学 | Data parallel processing method and system as well as load balancing scheduler |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110545242A (en) * | 2018-05-29 | 2019-12-06 | 杭州海康威视数字技术股份有限公司 | target analysis method and intelligent analysis equipment |
Also Published As
Publication number | Publication date |
---|---|
CN106020988B (en) | 2019-03-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9916183B2 (en) | Scheduling mapreduce jobs in a cluster of dynamically available servers | |
US8997107B2 (en) | Elastic scaling for cloud-hosted batch applications | |
CN114072766A (en) | System and method for digital labor intelligent organization | |
CN110889510B (en) | Online scheduling method and device for distributed machine learning task | |
CN116702907B (en) | Server-unaware large language model reasoning system, method and equipment | |
US20210200585A1 (en) | System for real-time scheduling in an ansynchronous transfer mode communication network | |
CN110099083A (en) | A kind of load equilibration scheduling method and device for server cluster | |
CN115033357A (en) | Micro-service workflow scheduling method and device based on dynamic resource selection strategy | |
CN116048721A (en) | Task allocation method and device for GPU cluster, electronic equipment and medium | |
Mejia-Alvarez et al. | An incremental approach to scheduling during overloads in real-time systems | |
Zikos et al. | A clairvoyant site allocation policy based on service demands of jobs in a computational grid | |
CN109189581B (en) | Job scheduling method and device | |
CN106020988B (en) | A kind of offline method for scheduling task of intelligent video monitoring system and device | |
Teng et al. | Scheduling real-time workflow on MapReduce-based cloud | |
CN105550025A (en) | Distributed IaaS (Infrastructure as a Service) scheduling method and system | |
Naik | A deadline-based elastic approach for balanced task scheduling in computing cloud environment | |
Lin et al. | Two-tier project and job scheduling for SaaS cloud service providers | |
CN116700925A (en) | Digital employee group cooperation method based on RPA | |
CN111767125A (en) | Task execution method and device, electronic equipment and storage medium | |
Rahni et al. | Feasibility analysis of non-concrete real-time transactions with edf assignment priority | |
CN115129481B (en) | Computing resource allocation method and device and electronic equipment | |
CN112988363B (en) | Resource scheduling method, device, server and storage medium | |
CN115712501A (en) | Cloud simulation method and system suitable for engineering machinery | |
CN108228334B (en) | Container cluster expansion method and device | |
CN114513423B (en) | Bandwidth adjustment method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |