Storage efficiency opportunities and analysis for video repositories

Conventional wisdom states that deduplication techniques do not yield good storage efficiency savings for video data. Indeed, even if the video content is the same, the video files differ from each other due to differences in closed-captioning, text overlay, language, and video resolution. Therefore deduplication techniques have not yielded good storage efficiency savings. In this paper, we look at the effectiveness of four different deduplication algorithms on different scenarios. We evaluate fixedsized, variable-sized, and two content-aware deduplication techniques. Our study shows that content-aware and variable-sized deduplication techniques do provide significant storage efficiency savings.