AWS News Blog

Amazon Elastic Transcoder Update

Transcoding is the process of adapting a media file (video or audio) to change the size, format, or other parameters in order to reduce the file size or to make it compatible with a particular type of device. Amazon Elastic Transcoder is a scalable, fully-managed service that works on a cost-effective pay per use model. You don’t have to license or install any software and you can take advantage of transcoding presets for a variety of popular output devices and formats. Amazon Elastic Transcoder outputs H.264 and VP8 video and AAC, MP3, and Vorbis audio in a number of package formats including MP4, WebM, MPEG2-TS, MP3, and OGG. Additionally, you may output segmented video files and manifest files to support HLS video streaming.

Enhanced Parallelism
Over the last couple of months we have quietly raised the level of parallelism within the service. If your job includes more than one type of output, work proceeds in parallel, up to the limit of 30 outputs per job. You can have up to four pipelines per AWS account and each pipeline can process between 12 and 20 jobs simultaneously by default (the limit is specific to each Region; check here for the particulars).

Elastic Transcoder has been tuned to minimize queue times, even as queues grow to 100 or more jobs. We have measured a median queue time of 0.6 seconds, with the P90 point at just 3.2 seconds. In other words, processing for 90% of the jobs submitted to Elastic Transcoder begins within 3.2 seconds. As an example, a 5 minute HD transcoding job submitted in the US East (Northern Virginia) Region with an input bitrate of 20 Mbps and multiple output bitrates of 800 Kbps, 1 Mbps, 2 Mbps all completed in about 3 minutes.

Getting Started
All AWS customers have access to 20 minutes of audio transcoding, 20 minutes of SD transcoding, and 10 minutes of HD transcoding at no charge as part of the AWS Free Usage Tier. Read the Elastic Transcoder Developer Guide and you will be up to speed in no time!

Jeff;