【摘要】：視頻轉(zhuǎn)碼是視頻錄制到播放不可或缺的一個(gè)重要環(huán)節(jié)之一,然而單機(jī)版視頻轉(zhuǎn)碼耗時(shí)長,特別是大視頻轉(zhuǎn)碼。該文采用一種基于Hadoop的分布式視頻轉(zhuǎn)碼系統(tǒng),使用Hadoop的Map Reduce分布式框架,通過JNI對(duì)FFmpeg庫的調(diào)用,實(shí)現(xiàn)分布式的視頻轉(zhuǎn)碼。實(shí)驗(yàn)結(jié)果,相比單機(jī)版視頻轉(zhuǎn)碼,分布式視頻轉(zhuǎn)碼有明顯的優(yōu)勢。
[Abstract]:Video transcoding is one of the most important links between video recording and playback. However, single video transcoding takes a long time, especially for large video transcoding. In this paper, a distributed video transcoding system based on Hadoop is adopted. The distributed video transcoding is realized by using the Map Reduce distributed framework of Hadoop and the call of FFmpeg library by JNI. Experimental results show that distributed video transcoding has obvious advantages compared with single video transcoding.
【作者單位】： 蘇州科技大學(xué)網(wǎng)絡(luò)與教育技術(shù)中心;蘇州科技大學(xué)電子與信息工程學(xué)院;中國電信股份有限公司吳江分公司;
【分類號(hào)】：TP393

【相似文獻(xiàn)】

相關(guān)會(huì)議論文 前3條

1 楊君蔚;楊雪;丁君;劉欣s，

本文編號(hào)：2408718