Achieving high performance content streaming with adaptive chunklets and active queue management

Recent years have seen a rapid increase in the volume of video/content streaming traffic on the Internet. The demand for a better user experience has also concurrently increased, with viewers preferring quicker video startup, higher video quality, and fewer rebuffers. Consumer video streams are likely to be bottlenecked and impacted by emerging Active Queue Management (AQM) schemes. This thesis spans the entire breadth from an in-depth review of content streaming and AQM technologies to a proposal/validation of a novel and backward-compatible technique for improving streaming performance. This thesis is firmly rooted in the reality of today's Internet technology and represents a thoroughly practical study with a potential for immediate impact.