Distributed transcoding system and method and device thereof

Abstract

The invention discloses a distributed transcoding system and method and a device thereof. The system comprises a transcoding module, a database and a shared file system, wherein the shared file system is used for receiving and storing a source file transmitted from an external requester side; the database is used for receiving and storing a task data packet transmitted from the external requester side; and the transcoding module is used for obtaining the task data packet from the database, obtaining a corresponding source file from the shared file system according to the task data packet, transcoding the source file, and string the transcoded source file into the shared file system. The distributed transcoding system and method and the device thereof have the advantages that a central task allocation system is removed and self-scheduling is achieved by various transcoding machines, so that the risk of a failure is reduced, the input cost of the machines is reduced and the difficulty of later maintenance is reduced.

Description

Technical field [0001] The invention relates to a distributed transcoding system, method and device. Background technique [0002] At present, there are some businesses that provide video hosting services for enterprises. One of the services is to transcode high-definition videos provided by enterprises into video streams suitable for Internet dissemination. This process involves the transfer of videos uploaded by users on the platform. The operation of encoding, because there will be a lot of users and user videos involved, it is impossible for us to use a single machine to transcode, and we must use a distributed method to transcode the videos uploaded by users. [0003] In general framework design, there is usually a task central task allocation system with multiple task processing machines to perform transcoding operations. This design completely relies on the central task allocation system. This system may be transitionally designed, except for task allocation scheduling , It...

AI Technical Summary

Problems solved by technology

[0003] In the general framework design, there is usually a central task allocation system for tasks, with multiple task processing machines for transcoding operations. This design completely relies on the central task allocation system. This system may be transitionally designed, except for the scheduling of task allocation. , it is also necessary to monitor all working machines in real time to determine whether to assign tasks to the working machine; and this scheduling system is very important, and it must be highly available to ensure that the scheduling system has no single point of failure risk; and when the central scheduling The system as a cluster itself requires multiple machines to be deployed, thereby increasing the cost of the machine

Images