CN114327877A - Multitask scheduling method, device and system - Google Patents
Multitask scheduling method, device and system Download PDFInfo
- Publication number
- CN114327877A CN114327877A CN202111582379.8A CN202111582379A CN114327877A CN 114327877 A CN114327877 A CN 114327877A CN 202111582379 A CN202111582379 A CN 202111582379A CN 114327877 A CN114327877 A CN 114327877A
- Authority
- CN
- China
- Prior art keywords
- task
- executed
- queue
- node
- 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.)
- Pending
Links
Images
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention provides a multitask scheduling method, a device and a system, wherein a task queue to be executed and a task processing table are set, a task number is used as a main key, when a plurality of threads scan the same task to be executed, one thread is required to successfully insert the task number of the task to be executed into the task processing table, the other threads fail, the successfully inserted thread executes the task to be executed, the unsuccessfully inserted thread re-scans the task queue to be executed, the same task to be executed is prevented from being simultaneously called and executed by a plurality of threads, no matter how large the concurrency of the system is, the interference on the normal scheduling of the task is avoided, and the multitask scheduling efficiency is improved.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, and a system for multi-task scheduling.
Background
With the continuous development of internet big data application, business application scenes gradually move to mobile internet, more clients and click volumes are attracted through mobile equipment, but a larger number of concurrent processing task requests are brought along, and great challenges are brought to the original business system.
In the prior art, a table locking mode is often adopted for preventing the multi-task scheduling problem, the executed task information is inserted into an individual task scheduling table, system resources are consumed in the locking and unlocking processes, and under the scene of over-large concurrency of tens of thousands of times per second, most task threads are caused to queue for a long time, so that the system performance is obviously reduced.
Disclosure of Invention
In view of this, the present invention provides a method, an apparatus, and a system for multi-task scheduling, which prevent a same task to be executed from being simultaneously invoked and executed by multiple threads, and improve efficiency of multi-task scheduling.
In order to achieve the above purpose, the invention provides the following specific technical scheme:
in a first aspect, an embodiment of the present invention provides a multitask scheduling method, which is applied to a thread in a node of a distributed system, where the method includes:
scanning a task queue to be executed, wherein each task to be executed in the task queue to be executed corresponds to a unique task number respectively;
acquiring a task to be executed, and inserting a task number of the task to be executed into a task processing table;
executing the task to be executed under the condition of successful insertion;
and in the case of insertion failure, rescanning the task queue to be executed.
In some embodiments, the method further comprises:
after the task to be executed is executed, rescanning the task queue to be executed;
and releasing the thread resources under the condition that no task to be executed exists in the task queue to be executed.
In a second aspect, an embodiment of the present invention provides a multitask scheduling method, which is applied to a master node in a distributed system, where the method includes:
creating a task queue to be executed and a task processing table, wherein each task to be executed in the task queue to be executed corresponds to a unique task number;
and under the condition that a thread scans the queue of the tasks to be executed to obtain the tasks to be executed, and successfully inserts the task number of the tasks to be executed into the task processing table, other threads scan the queue of the tasks to be executed again.
In some embodiments, the method further comprises:
periodically acquiring resource use data of each node in the distributed system;
and sending a risk early warning signal under the condition that the resource use data of the node reaches an early warning interval.
In some embodiments, the method further comprises:
and under the condition that the resource use data of the node reaches the early warning interval, dynamically allocating resources for the node by adopting a preset dynamic allocation strategy.
In some embodiments, the method further comprises:
the method comprises the steps of periodically collecting task processing data of each node in a distributed system;
respectively judging whether the running tasks of each node are excessive according to historical task processing data;
and under the condition that the running tasks of the nodes are excessive, adjusting the task distribution strategy.
In a third aspect, an embodiment of the present invention provides a multitask scheduling device, which is applied to a thread in a node of a distributed system, where the device includes:
the task scanning unit is used for scanning the task queues to be executed, and each task to be executed in the task queues to be executed corresponds to a unique task number respectively;
the task processing device comprises a task obtaining unit, a task processing unit and a scanning unit, wherein the task obtaining unit is used for obtaining a task to be executed, inserting a task number of the task to be executed into a task processing table, triggering the task executing unit under the condition of successful insertion, and triggering the scanning unit under the condition of failed insertion;
and the task execution unit is used for executing the task to be executed.
In some embodiments, the apparatus further comprises:
the resource releasing unit is used for rescanning the task queue to be executed after the task to be executed is executed; and releasing the thread resources under the condition that no task to be executed exists in the task queue to be executed.
In a fourth aspect, an embodiment of the present invention provides a multitask scheduling apparatus, which is applied to a master node in a distributed system, where the apparatus includes:
the system comprises a creating unit, a task processing unit and a processing unit, wherein the creating unit is used for creating a task queue to be executed and a task processing table, and each task to be executed in the task queue to be executed corresponds to a unique task number; and under the condition that a thread scans the queue of the tasks to be executed to obtain the tasks to be executed, and successfully inserts the task number of the tasks to be executed into the task processing table, other threads scan the queue of the tasks to be executed again.
In some embodiments, the apparatus further comprises:
the first acquisition unit is used for periodically acquiring resource use data of each node in the distributed system;
and the risk early warning unit is used for sending a risk early warning signal under the condition that the resource use data of the nodes reach the early warning interval.
In some embodiments, the apparatus further comprises:
and the resource dynamic allocation unit is used for dynamically allocating resources for the node by adopting a preset dynamic allocation strategy under the condition that the resource use data of the node reaches the early warning interval.
In some embodiments, the apparatus further comprises:
the second acquisition unit is used for periodically acquiring task processing data of each node in the distributed system;
the judging unit is used for respectively judging whether the running tasks of each node are excessive according to the historical task processing data;
and the distribution strategy adjusting unit is used for adjusting the task distribution strategy under the condition that the running tasks of the nodes are excessive.
In a fifth aspect, an embodiment of the present invention provides a distributed system, including: a master node and at least one slave node;
the master node is configured to perform the method as described in any implementation manner of the second aspect;
the master node and the slave node comprise at least one thread for performing the method as described in any of the implementations of the first aspect.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses a multitask scheduling method, which is characterized in that a task queue to be executed and a task processing table are arranged, a task number is used as a main key, when a plurality of threads scan the same task to be executed, one thread is necessary to successfully insert the task number of the task to be executed into the task processing table, the other threads fail, the successfully inserted thread executes the task to be executed, the unsuccessfully inserted thread re-scans the task queue to be executed, the same task to be executed is prevented from being simultaneously called and executed by the plurality of threads, no matter how large the concurrency of a system is, the normal scheduling of the task cannot be interfered, and the multitask scheduling efficiency is improved.
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 embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flowchart of a multitask scheduling method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating another multitask scheduling method according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating another multitask scheduling method according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a multi-task scheduling device applied to threads 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 invention provides a multitask scheduling method, a multitask scheduling device and a multitask repeated scheduling exception capture processing mechanism, which are used for realizing accurate control on multithreading concurrent task processing, preventing thread execution interference and avoiding repeated scheduling of tasks.
Specifically, referring to fig. 1, the method for scheduling multiple tasks disclosed in this embodiment is applied to a thread in a node of a distributed system, and specifically includes the following steps:
s101: scanning a task queue to be executed, wherein each task to be executed in the task queue to be executed corresponds to a unique task number;
the task queue to be executed is created by the main node in the distributed system, is a set of various tasks waiting for processing in a queue and is managed in a queue mode.
The thread is created by the node where the thread is located, and the task queue to be executed is scanned after the thread is started.
S102: acquiring a task to be executed, and inserting a task number of the task to be executed into a task processing table;
the following table is a task processing table, which represents a collection of various tasks that have been called by threads and are being executed, and in which the task number is the Primary Key (PK).
| Task number (PK) | Calling time |
| 00001 | 20190508121212985 |
| 00002 | 20190508121332243 |
| 00003 | 20190508121225732 |
| ...... | ...... |
It can be understood that when a plurality of threads scan the same task to be executed, one thread must successfully insert the task number of the task to be executed into the task processing table, and the other threads all fail.
S103: executing the task to be executed under the condition of successful insertion;
and under the condition that the task number of the task to be executed is successfully inserted into the task processing table, returning a success mark, setting the task state of the task to be executed in the task queue to be under execution, and starting the execution work of the subsequent task.
S104: and in the case of insertion failure, rescanning the task queue to be executed.
Under the condition that the task number of the task to be executed is failed to be inserted into the task processing table, the thread reports a main key conflict exception error, captures a multi-task repeated scheduling exception, determines that the task to be executed is called by other threads, then re-scans the task queue to be executed, and repeatedly executes S101-S104.
By utilizing a multitask repeated scheduling exception capturing and processing mechanism, task processing logic is simpler and clearer, and no matter the number of tasks is more and the concurrent processing frequency is higher, each task to be executed can be scheduled and executed by only one thread and cannot be interfered by other threads, so that a correct result is ensured, accurate control over processing of the multithread concurrent tasks is realized, and repeated scheduling of the tasks is avoided.
Further, in each node in the distributed system, multi-task parallel processing is started. After one task is processed, the thread does not immediately release resources, but continues to search for the next task to be executed from the task queue to be executed, if yes, the process is started immediately, and the thread resources are released until no task to be executed exists in the task queue to be executed, so that the extra resource consumption of the system caused by frequent starting and releasing of the thread is avoided, and the processing performance is greatly improved.
The embodiment also discloses a multitask scheduling method, which is applied to the main node in the distributed system, and the method comprises the following steps:
and creating a task queue to be executed and a task processing table, wherein each task to be executed in the task queue to be executed corresponds to a unique task number.
And under the condition that the thread scans the queue of the task to be executed to obtain the task to be executed and successfully inserts the task number of the task to be executed into the task processing table, other threads scan the queue of the task to be executed again.
Furthermore, in order to realize intelligent monitoring of system operation risk and task processing, the main node monitors the system operation risk, monitors node resources and handles the forward movement with the risk, and the system operation efficiency and stability are improved.
Referring to fig. 2, the present embodiment provides a method for multitask scheduling, including the following steps:
s201: periodically acquiring resource use data of each node in the distributed system;
specifically, the resource usage data of each node in the distributed system may be collected according to a preset period, such as every second, every minute, and the like, where the resource usage data includes: CPU, memory, disk storage, network I/O.
S202: sending a risk early warning signal under the condition that the resource use data of the node reaches an early warning interval;
it should be noted that the early warning interval is determined according to an operation risk monitoring data mining model.
The risk monitoring data mining model is constructed in advance according to historical data and is used for analyzing the comparison relation between the task processing quantity and the processing efficiency and the use condition of system resources; analyzing the interval of CPU, memory, storage and network I/O resource consumption, wherein the system has the most stable processing performance and the most processing tasks; the system operation efficiency is obviously reduced in which interval, and the task execution time is obviously prolonged; and the running risk of the system is caused by the problems of downtime, network blockage and memory overflow of cluster nodes in various intervals.
And under the condition that the monitored resource use data of the nodes reach the early warning interval, sending a risk early warning signal to prompt manual intervention processing, improving the timeliness of risk early warning and being beneficial to manual processing as soon as possible.
S203: and dynamically allocating resources for the node by adopting a preset dynamic allocation strategy.
In order to process the resource usage risk as soon as possible, in this embodiment, a preset dynamic allocation policy may be further adopted to automatically allocate available resources in the system to alleviate the operation risk, such as dynamically allocating storage resources, memory resources, and the like, when it is monitored that the resource usage data of the node reaches the early warning interval.
Referring to fig. 3, the present embodiment provides a method for multitask scheduling, including the following steps:
s301: the method comprises the steps of periodically collecting task processing data of each node in a distributed system;
specifically, task processing data, such as task processing data, processing data amount, weighting processing time, and the like, of each node in the distributed system may be acquired according to a preset period, such as every second, every minute, and the like.
The weighted processing time is the time after weighted processing of the task processing times with different priorities, and it can be understood that the task with higher priority has larger weight, and the weighted processing time is about longer, that is, the task with higher priority allows longer task processing time.
S302: respectively judging whether the running tasks of each node are excessive according to historical task processing data;
and acquiring comparison data of historical system CPU, memory, storage and other configuration resources and historical execution quantity and average execution time of the tasks of the same type, calculating the pressure load condition of the current node in real time through big data simulation, and respectively judging whether the running task of each node is excessive.
S303: and under the condition that the running tasks of the nodes are excessive, adjusting the task distribution strategy.
The health and stability of the system cluster are ensured by adjusting the task distribution strategy under the condition that the running tasks of the nodes are excessive.
The task distribution strategy can be any one of the existing task distribution strategies, and the task quantity processed by each node in the distributed system is balanced by adjusting the task distribution strategy, so that the overall task processing efficiency of the distributed system is improved.
The embodiment also provides a multitask scheduling device, which is applied to a thread in a node of a distributed system, please refer to fig. 4, where the multitask scheduling device includes:
the task scanning unit 401 is configured to scan a to-be-executed task queue, where each to-be-executed task in the to-be-executed task queue corresponds to a unique task number;
a task obtaining unit 402, configured to obtain a task to be executed, insert a task number of the task to be executed into a task processing table, trigger a task executing unit 403 when the insertion is successful, and trigger the scanning unit 401 when the insertion is failed;
and the task execution unit is used for executing the task to be executed.
In some embodiments, the apparatus further comprises:
the resource releasing unit is used for rescanning the task queue to be executed after the task to be executed is executed; and releasing the thread resources under the condition that no task to be executed exists in the task queue to be executed.
The multi-task scheduling device disclosed by the embodiment has the advantages that the task number is used as a main key by setting the task queue to be executed and the task processing table, when a plurality of threads scan the same task to be executed, one thread is necessary to successfully insert the task number of the task to be executed into the task processing table, other threads fail, the thread which is successfully inserted executes the task to be executed, the thread which is unsuccessfully inserted re-scans the task queue to be executed, the same task to be executed is prevented from being simultaneously called and executed by the plurality of threads, no matter how large the system concurrency is, the interference on the normal scheduling of the task can be avoided, and the multi-task scheduling efficiency is improved. And simultaneously, in each node in the distributed system, multi-task parallel processing is started. After one task is processed, the thread does not immediately release resources, but continues to search for the next task to be executed from the task queue to be executed, if yes, the process is started immediately, and the thread resources are released until no task to be executed exists in the task queue to be executed, so that the extra resource consumption of the system caused by frequent starting and releasing of the thread is avoided, and the processing performance is greatly improved.
The present embodiment further provides a multitask scheduling device, which is applied to a master node in a distributed system, where the multitask scheduling device includes:
the system comprises a creating unit, a task processing unit and a processing unit, wherein the creating unit is used for creating a task queue to be executed and a task processing table, and each task to be executed in the task queue to be executed corresponds to a unique task number; and under the condition that a thread scans the queue of the tasks to be executed to obtain the tasks to be executed, and successfully inserts the task number of the tasks to be executed into the task processing table, other threads scan the queue of the tasks to be executed again.
In some embodiments, the apparatus further comprises:
the first acquisition unit is used for periodically acquiring resource use data of each node in the distributed system;
and the risk early warning unit is used for sending a risk early warning signal under the condition that the resource use data of the nodes reach the early warning interval.
In some embodiments, the apparatus further comprises:
and the resource dynamic allocation unit is used for dynamically allocating resources for the node by adopting a preset dynamic allocation strategy under the condition that the resource use data of the node reaches the early warning interval.
In some embodiments, the apparatus further comprises:
the second acquisition unit is used for periodically acquiring task processing data of each node in the distributed system;
the judging unit is used for respectively judging whether the running tasks of each node are excessive according to the historical task processing data;
and the distribution strategy adjusting unit is used for adjusting the task distribution strategy under the condition that the running tasks of the nodes are excessive.
The multitask scheduling device disclosed by the embodiment realizes intelligent monitoring of system operation risk and task processing while realizing a task processing mode based on an abnormal capturing anti-interference mechanism, and the main node forwards the system operation risk monitoring, node resource monitoring and risk processing, so that the system operation efficiency and stability are improved.
The embodiment also provides a distributed system, which can be a Hadoop distributed system, and is responsible for multitask management and scheduling execution through a Spark frame, wherein the Hadoop is a distributed data storage frame, mass data can be rapidly stored through a distributed file system HDFS, and various means for rapid retrieval and processing are provided. Spark is a fast parallel computing framework based on memory, improves the fast response of data processing in a mass data environment, and simultaneously ensures high fault tolerance and low cost.
The distributed system includes: a master node and at least one slave node;
the main node is used for executing the following multitask scheduling method:
creating a task queue to be executed and a task processing table, wherein each task to be executed in the task queue to be executed corresponds to a unique task number;
and under the condition that a thread scans the queue of the tasks to be executed to obtain the tasks to be executed, and successfully inserts the task number of the tasks to be executed into the task processing table, other threads scan the queue of the tasks to be executed again.
Further, the method further comprises:
periodically acquiring resource use data of each node in the distributed system;
and sending a risk early warning signal under the condition that the resource use data of the node reaches an early warning interval.
Further, the method further comprises:
and under the condition that the resource use data of the node reaches the early warning interval, dynamically allocating resources for the node by adopting a preset dynamic allocation strategy.
Further, the method further comprises:
the method comprises the steps of periodically collecting task processing data of each node in a distributed system;
respectively judging whether the running tasks of each node are excessive according to historical task processing data;
and under the condition that the running tasks of the nodes are excessive, adjusting the task distribution strategy.
The master node and the slave node comprise at least one thread, and the thread is used for executing the following multitask scheduling method:
scanning a task queue to be executed, wherein each task to be executed in the task queue to be executed corresponds to a unique task number respectively;
acquiring a task to be executed, and inserting a task number of the task to be executed into a task processing table;
executing the task to be executed under the condition of successful insertion;
and in the case of insertion failure, rescanning the task queue to be executed.
Further, the method further comprises:
after the task to be executed is executed, rescanning the task queue to be executed;
and releasing the thread resources under the condition that no task to be executed exists in the task queue to be executed.
In the distributed system disclosed in this embodiment, the master node sets the to-be-executed task queue and the task processing table, and uses the task number as the master key, when a plurality of threads scan the same to-be-executed task, one thread must successfully insert the task number of the to-be-executed task into the task processing table, and all the other threads fail, the thread that is successfully inserted executes the to-be-executed task, and the thread that is unsuccessfully inserted rescans the to-be-executed task queue, so that the same to-be-executed task is prevented from being simultaneously invoked and executed by the plurality of threads, no matter how large the system concurrency is, no interference is generated to the normal scheduling of the task, and the efficiency of multi-task scheduling is improved.
The thread adopts a task continuous execution mechanism, when one task is processed, the thread is not released immediately, the next task to be executed is continuously searched from the task list, if yes, the processing is started immediately, and the thread resource is released until no task to be executed exists, so that the additional resource consumption of the system caused by frequent starting and releasing of the thread is avoided, and the processing performance is greatly improved.
The distributed system operation risk and task processing intelligent monitoring management enables the system operation risk monitoring, node resource monitoring and risk processing to be advanced, and system operation efficiency and stability are greatly improved.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It is further 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.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above embodiments can be combined arbitrarily, and the features described in the embodiments in the present specification can be replaced or combined with each other in the above description of the disclosed embodiments, so that those skilled in the art can implement or use the present application.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A multitask scheduling method is applied to threads in nodes of a distributed system, and the method comprises the following steps:
scanning a task queue to be executed, wherein each task to be executed in the task queue to be executed corresponds to a unique task number respectively;
acquiring a task to be executed, and inserting a task number of the task to be executed into a task processing table;
executing the task to be executed under the condition of successful insertion;
and in the case of insertion failure, rescanning the task queue to be executed.
2. The method of claim 1, further comprising:
after the task to be executed is executed, rescanning the task queue to be executed;
and releasing the thread resources under the condition that no task to be executed exists in the task queue to be executed.
3. A multitask scheduling method is applied to a main node in a distributed system, and the method comprises the following steps:
creating a task queue to be executed and a task processing table, wherein each task to be executed in the task queue to be executed corresponds to a unique task number;
and under the condition that a thread scans the queue of the tasks to be executed to obtain the tasks to be executed, and successfully inserts the task number of the tasks to be executed into the task processing table, other threads scan the queue of the tasks to be executed again.
4. The method of claim 3, further comprising:
periodically acquiring resource use data of each node in the distributed system;
and sending a risk early warning signal under the condition that the resource use data of the node reaches an early warning interval.
5. The method of claim 4, further comprising:
and under the condition that the resource use data of the node reaches the early warning interval, dynamically allocating resources for the node by adopting a preset dynamic allocation strategy.
6. The method of claim 3, further comprising:
the method comprises the steps of periodically collecting task processing data of each node in a distributed system;
respectively judging whether the running tasks of each node are excessive according to historical task processing data;
and under the condition that the running tasks of the nodes are excessive, adjusting the task distribution strategy.
7. A multitask scheduling device, for application to threads in nodes of a distributed system, the device comprising:
the task scanning unit is used for scanning the task queues to be executed, and each task to be executed in the task queues to be executed corresponds to a unique task number respectively;
the task processing device comprises a task obtaining unit, a task processing unit and a scanning unit, wherein the task obtaining unit is used for obtaining a task to be executed, inserting a task number of the task to be executed into a task processing table, triggering the task executing unit under the condition of successful insertion, and triggering the scanning unit under the condition of failed insertion;
and the task execution unit is used for executing the task to be executed.
8. The apparatus of claim 7, further comprising:
the resource releasing unit is used for rescanning the task queue to be executed after the task to be executed is executed; and releasing the thread resources under the condition that no task to be executed exists in the task queue to be executed.
9. A multitask scheduling device, for use in a master node in a distributed system, the device comprising:
the system comprises a creating unit, a task processing unit and a processing unit, wherein the creating unit is used for creating a task queue to be executed and a task processing table, and each task to be executed in the task queue to be executed corresponds to a unique task number; and under the condition that a thread scans the queue of the tasks to be executed to obtain the tasks to be executed, and successfully inserts the task number of the tasks to be executed into the task processing table, other threads scan the queue of the tasks to be executed again.
10. A distributed system, comprising: a master node and at least one slave node;
the main node is used for executing the multitask scheduling method of any one of claims 3-6;
the master node and the slave node comprise at least one thread for performing the multitask scheduling method according to claim 1 or 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111582379.8A CN114327877A (en) | 2021-12-22 | 2021-12-22 | Multitask scheduling method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111582379.8A CN114327877A (en) | 2021-12-22 | 2021-12-22 | Multitask scheduling method, device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114327877A true CN114327877A (en) | 2022-04-12 |
Family
ID=81053715
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111582379.8A Pending CN114327877A (en) | 2021-12-22 | 2021-12-22 | Multitask scheduling method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114327877A (en) |
-
2021
- 2021-12-22 CN CN202111582379.8A patent/CN114327877A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10838777B2 (en) | Distributed resource allocation method, allocation node, and access node | |
| Lu et al. | Log-based abnormal task detection and root cause analysis for spark | |
| US8261266B2 (en) | Deploying a virtual machine having a virtual hardware configuration matching an improved hardware profile with respect to execution of an application | |
| CN111818159B (en) | Management method, device, equipment and storage medium of data processing node | |
| US20080162272A1 (en) | Methods and apparatus to collect runtime trace data associated with application performance | |
| WO2021159638A1 (en) | Method, apparatus and device for scheduling cluster queue resources, and storage medium | |
| CN112162865A (en) | Server scheduling method and device and server | |
| US9027025B2 (en) | Real-time database exception monitoring tool using instance eviction data | |
| WO2017107456A1 (en) | Method and apparatus for determining resources consumed by task | |
| US20220070099A1 (en) | Method, electronic device and computer program product of load balancing | |
| CN110727508A (en) | Task scheduling system and scheduling method | |
| CN112685160A (en) | Scheduling method and device of timing task, terminal equipment and computer storage medium | |
| CN110109741B (en) | Method and device for managing circular tasks, electronic equipment and storage medium | |
| CN113391911B (en) | Dynamic scheduling method, device and equipment for big data resources | |
| CN114461385A (en) | Thread pool scheduling method, device and equipment and readable storage medium | |
| CN111580951A (en) | Task allocation method and resource management platform | |
| da Silva et al. | Self-healing of operational workflow incidents on distributed computing infrastructures | |
| CN114327877A (en) | Multitask scheduling method, device and system | |
| CN115858499A (en) | Database partition processing method and device, computer equipment and storage medium | |
| Zhu et al. | Green scheduling: A scheduling policy for improving the energy efficiency of fair scheduler | |
| CN110750350A (en) | Large resource scheduling method, system, device and readable storage medium | |
| Gottumukkala et al. | Reliability-aware resource allocation in HPC systems | |
| CN115422010A (en) | Node management method and device in data cluster and storage medium | |
| CN105760215A (en) | Map-reduce model based job running method for distributed file system | |
| CN109302723A (en) | A kind of multinode real-time radio pyroelectric monitor control system Internet-based and control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |