Video transcoding is the process of producing from an original (already encoded) input video sequence, multiple output sequences, each at potentially different bitrate, resolution and/or format. Transcoding is essential to support video delivery towards clients that use different players and have different network access capabilities. In the most basic scheme the input sequence is decoded and then re-encoded at the desired levels. Although significant research on fast transcoding schemes exists, the transcoding process is still computationally intensive. For this reason efficientscheduling methods that allocate resources to transcoding jobs are necessary in order to achieve good overall performance. Such policies usually aim at allocating transcoding jobs over co-located servers, thus, they typically overlook parameters such as network traffic. Motivated by the case of transcoding in the Cloud, in this paper we investigate the problem of scheduling transcoding jobs over a distributed system comprising of processing nodes that are geographically dispersed and might be whole clusters or even separate data centers. We propose algorithms to minimize both the inter-node network traffic and the intra-node energy consumption, while meeting the deadlines and quality requirements. Through simulation experiments we conclude on thebest alternatives.