Abstract
Video transcoding has become a key technology for video content distribution network service. In this paper, we propose a novel MapReduce-based parallel video transcoding method. In our method, video files are stored on a shared file system to reduce the overhead of disks I/O and networks in the Hadoop MapReduce. FFmpeg is used to compute the splitting point of the video and the actual video transcoding. Experimental results show that our method can significantly reduce the time of transcoding.
Highlights
With the rapid development of network video applications, how to support the access to video content of various terminal equipment (PCs, laptops, smartphones, etc.) has been an important issue concerned by the CDN(content distribution network) service providers
Currently,a popular distributed computing model-Hadoop MapReduce[3] has been widely used in the field of distributed video transcoding[4,5,6]
In this paper,we propose a novel parallel video transcoding method, which uses original MapReduce and FFmpeg [7] to achieve efficient distributed transcoding
Summary
With the rapid development of network video applications, how to support the access to video content of various terminal equipment (PCs, laptops, smartphones, etc.) has been an important issue concerned by the CDN(content distribution network) service providers. The video from one format to another format, including compression algorithm, resolution, bit rate, frame rate [2]. Currently,a popular distributed computing model-Hadoop MapReduce[3] has been widely used in the field of distributed video transcoding[4,5,6]. Storing video files on Hadoop HDFS can incur the overhead of disks I/O and networks. You have to modify the FileInputFormat class in Hadoop MapReduce or modify the transcoding software to support reading and writing files in HDFS. In this paper,we propose a novel parallel video transcoding method, which uses original MapReduce and FFmpeg [7] to achieve efficient distributed transcoding. Video files are stored on a shared file system to reduce the overhead of disks I/O and networks in the MapReduce. Video splitting is no need to split a video, just computing time code of video’s splitting point by FFmpeg and recording the time code
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have