LwTE: Light-Weight Transcoding at the Edge

Due to the growing demand for video streaming services, providers have deal with increasing resource requirements increasingly heterogeneous environments. To mitigate this problem, many works been proposed which aim ( i) improve cloud/edge caching efficiency, (ii) use computation power available in on-the-fly transcoding, and (iii) optimize trade-off among various cost parameters, e.g., storage, computation, bandwidth. In paper, we propose LwTE, a novel Light- weight Transcoding approach at Edge, context of HTTP Adaptive Streaming (HAS). During encoding process segment origin side, computationally intense search processes are going on. The main idea LwTE is store optimal results these as metadata each bitrate reuse them edge servers reduce required time computational resources transcoding. enables us only highest plus corresponding (of very small size) unpopular segments/bitrates. way, addition significant reduction bandwidth storage consumption, transcoding requested remarkably decreased by utilizing its metadata; unnecessary avoided. Popular segments/bitrates being stored. We investigate our Video-on-Demand (VoD) services optimizing (transcoding) costs then compare it conventional methods (store all bitrates, partial transcoding). indicate that reduces least 80% decreases aforementioned 12% 70% compared state-of-the-art approaches.