A method of managing machine and related apparatus

A technology of machine and main machine, applied in the field of development platform, can solve problems such as unsatisfactory

Active Publication Date: 2020-04-07
WUHAN DOUYU NETWORK TECH CO LTD
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In practical applications, for the pornography service framework, due to the large number of live broadcast rooms, one identification service machine cannot meet the online demand, so multiple identification machines are required

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 method of managing machine and related apparatus
  • A method of managing machine and related apparatus
  • A method of managing machine and related apparatus

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0079] Embodiment 2. A method for identifying illegal pictures;

[0080] In practical applications, with the popularization of artificial intelligence deep learning, deep learning is now used to intelligently identify whether a picture contains pornographic content. In general, deep learning consumes computer resources. For live broadcast platforms, there are tens of thousands of live broadcast rooms at the same time, and screenshots of the live broadcast room content must be taken every preset 30 seconds to check whether it contains pornographic content. Therefore, a dedicated deep learning machine with a large number of graphics processing unit (GPU) computing resources will be used. For the pornography service framework, it is necessary to make good use of the deep learning machine, and make the program more robust, while being able to cover all the live broadcast rooms of the live broadcast platform. Therefore, this article provides a method for identifying illegal pictures....

Embodiment 3

[0122] Embodiment 3. A task scheduling method;

[0123] In addition, for the pornography service framework, when the number of simultaneous broadcasts in the live room is relatively large, it will cause a lot of requests, and it is very likely that the number of requests at the same time is very large, which will cause many requests to time out and not be processed. So based on the machines with multiple recognition services, how to design reasonably so that all requested tasks will not cause timeout or discarding, and at the same time, multiple recognition machines can perform full-load recognition tasks without causing one recognition There are many tasks, and other machines are idle. Based on this, this article designs all the pornographic identification requests on the master machine as a cache queue. At the same time, a load-balancing scheduling control is also designed on the master machine, so that tasks are distributed as evenly as possible to different machines. When a ...

Embodiment 4

[0181] Embodiment 4. A picture recognition method;

[0182] For the pornography recognition service in this article, each image will use GPU resources, and GPU is the core resource in the entire recognition, and it is also the resource that is most likely to cause bottlenecks. Through the operation of the entire system, this article finds that For the live broadcast room, the strategy of using a preset time duration was adopted to intercept the pictures in the video for identification. It was found that in many cases, the content of the live broadcast room did not change much in a continuous period of time. It is understandable that if the pictures are similar, the corresponding recognition results are also similar. Therefore, the similarity of the pictures can be recognized by the GPU algorithm. If the pictures are similar, the previous recognition results will be returned directly, so that the GPU identification resources to reduce the possibility of bottlenecks. Then you need...

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 embodiment of the invention discloses a machine management method and a related device, which are used for scheduling different recognition machines according to requests, improving the auditing efficiency and achieving the purpose of disaster recovery backup. The method comprises the steps that a main machine designs a memory data structure through a String function, the memory data structureis stored through a list chain table of a standard template library STL, and the memory data structure is used for managing registration and logout of an auxiliary machine; when the auxiliary machineis started, the main machine receives a registration request message sent by the auxiliary machine, and the registration request message comprises the name of the auxiliary machine and the number ofthe auxiliary machine; in response to the registration request message, the main machine stores the name of the auxiliary machine and the number of the auxiliary machine by calling a function push _ back in the list chain table so as to realize registration of the auxiliary machine; the main machine receives a logout request message sent by the auxiliary machine, wherein the logout request messageis used for requesting logout from an recognition machine group of the main machine; in response to the logout request message, the primary machine logs out the secondary machine according to the logout request message.

Description

Technical field [0001] The invention relates to the field of development platforms, in particular to a method and related devices for managing machines. Background technique [0002] For the live broadcast platform, the content of the live broadcast on the live broadcast platform needs to be reviewed to confirm that the content of the live broadcast is legal and legal, and cannot be a pornographic broadcast. In the prior art, to confirm whether the content of the live broadcast is regular and legal, it can be identified by a machine in the pornography service framework. In actual applications, for the pornography service framework, due to the large number of live broadcast rooms, one recognition service-oriented machine cannot meet the online demand, so multiple recognition machines are required. [0003] Therefore, how to schedule the multiple identification machines more efficiently to improve the audit efficiency is a problem that needs to be solved urgently. Summary of the in...

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 Applications(China)
IPC IPC(8): G06F9/54G06F11/14H04N21/2187H04N21/266
CPCG06F9/547G06F11/1464H04N21/266H04N21/2187
Inventor 周志刚
Owner WUHAN DOUYU NETWORK TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap