Unlock instant, AI-driven research and patent intelligence for your innovation.

A two-level fault-tolerant multimedia distributed task processing method

A distributed task and processing method technology, applied in the field of multimedia distributed task processing, can solve problems such as difficult to take over other node tasks, full computing resources, etc.

Active Publication Date: 2018-09-18
ZHEJIANG UNIVIEW TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If a small-scale computing cluster is composed of 10 i7-3770CPU servers to process the video concentration tasks of 60-channel 1080P HD cameras, the computing resources of each server are basically occupied, and it is difficult to take over the tasks of other nodes

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A two-level fault-tolerant multimedia distributed task processing method
  • A two-level fault-tolerant multimedia distributed task processing method
  • A two-level fault-tolerant multimedia distributed task processing method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] Please refer to figure 2 , figure 2 The processing of continuing the task after the computing node restarts within a predetermined time after power failure is given, and the task processing includes task recovery and task synchronization. This process ensures the consistency of task information between computing nodes and scheduling nodes when realizing automatic recovery of unfinished tasks.

[0042] S21. The video processing algorithm module of the computing node loads unfinished tasks of the node from the database, and resumes execution of the tasks; and sends a notification to the cluster middleware module after all tasks resume execution.

[0043] S22. After receiving the notification, the cluster middleware module acquires currently executed task information from the video processing algorithm module, and the task information includes a task ID.

[0044] S23. The video processing algorithm module returns task information of all currently executed tasks.

[00...

Embodiment 2

[0054] Please refer to image 3 , image 3 The process of task recovery and task synchronization in which the video processing algorithm module process of the computing node crashes and resumes within a predetermined time is given.

[0055] S31. The video processing algorithm module of the computing node loads unfinished tasks of the node from the database, and resumes task execution; and sends a notification to the cluster middleware module after all tasks resume execution.

[0056] S32. After receiving the notification, the cluster middleware module acquires currently executed task information (including task ID, task handle, etc.) from the video processing algorithm module.

[0057] S33. The video processing algorithm module returns task information of all currently executed tasks.

[0058] S34. The cluster middleware module judges whether each task ID in all returned tasks can match the task ID recorded in its own task control block, if yes, then go to step S35, otherwis...

Embodiment 3

[0063] Embodiment 3 provides the situation that after the network between the computing node and the scheduling node is disconnected, it recovers within a predetermined time. After the network is restored, for the processing flow of computing nodes, please refer to Figure 4 .

[0064] S41. The cluster middleware module of the computing node sets an out-of-sync flag on all task control blocks of itself.

[0065] S42. The cluster middleware module initiates registration to the scheduling node.

[0066] S43. After the registration is successful, the scheduling node sends a task synchronization message to it, and the message carries task information of all unfinished tasks of the computing node recorded by the scheduling node.

[0067] S44. The computing node judges whether the task ID recorded in each task control block of the computing node can match the task ID in the synchronization message. If yes, execute step S45, otherwise execute step S46.

[0068] S45. Clear the out-...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a two-level fault-tolerant multimedia distributed task processing method applied to a distributed cluster system. The distributed cluster system comprises a scheduling node and a plurality of computing nodes. The method includes the steps that the restoring task of the computing node is processed in a certain time window after any computing node breaks down; the scheduling node executes redistribution of tasks of the computing node beyond the certain time window. According to the scheme, the scheduling node supports task fault takeover, and meanwhile the computing nodes have the fault restoration ability of the node tasks. As for most of fault situations, the method can guarantee that the tasks continue to be executed on the initially distributed computing nodes as much as possible, and the situation that computing resources occupied by the nodes are unbalanced after the tasks are redistributed to other nodes is effectively avoided.

Description

technical field [0001] The invention relates to the technical field of cloud computing, in particular to a two-level fault-tolerant multimedia distributed task processing method. Background technique [0002] It has become an increasingly urgent need to analyze and process massive video and picture data generated by high-definition cameras, bayonet cameras, and electric police in large-scale video surveillance systems such as public security and traffic. [0003] On the one hand, intelligent analysis of videos and pictures, and the extraction of structured information such as people, vehicles, and objects in them can facilitate video retrieval of key people and vehicles, and at the same time, in-depth data mining can be carried out based on these information. In addition, for massive surveillance video data, there are processing requirements such as concentration, summary, diagnosis, and transcoding. [0004] For a large-scale video surveillance system, a single device node...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F11/00
Inventor 成浩
Owner ZHEJIANG UNIVIEW TECH CO LTD