By Yasser F. Syed Ph.D., Comcast Distinguished Engineer and Weidong Mao Ph.D., Comcast Senior Fellow & IEEE Fellow
This paper will examine what will be needed for Cloud TV video services to utilize a virtualized software platform for orchestrating distributed transcoding workflows. There are many encoding constructs that are now becoming realizable in this environment. Until recently, customized IC and hardware-based designs had been a practical approach to getting the job done using general-purpose servers. But these general processors are getting faster, more efficient, and more cooperative in multi-processor configuration groups. Encoding/transcoding technologies, and especially those designed for Multiple Bit Rate (MBR) technologies using H.264/ Advanced Video Coding (AVC), as well as High Efficiency Video Coding (HEVC), are increasingly designed to become more flexible, adaptable, and scalable. Furthermore, encoding/transcoding quality continues to increase, resulting in higher density growth at a much lower cost using general-purpose servers. This paper discusses how transcoding workflows can be adjusted, so as to to take advantage of new server architectures, better networking, and distributed storage -- through a concept called a Transcoding Resource Manager.