This tutorial has was created as part of a series of workshops for learning about automating around AWS MediaConvert. As of now, there are several other resources and code samples in this space that serve as a better starting point for implementing a watchfolder workflow:
VOD on AWS Foundation Solution - this solution implements a similar watchfolder architecture as the tutorial. It adds a tighter security posture and CloudFront origin for video play out.
VOD on AWS Solution - this solution builds on the Foundation architecture by adding DynamoDB state management to track transcoding jobs triggered by the workflow and Step Functions for workflow orchestration.
This project contains examples for automating Video On Demand (VOD) workflows on AWS. These are code samples to get you started on common tasks rather than an end to end architecture suitable for production. We will prefer simple, easy to understand examples that are appropriate to answer the "How do I ...?" kind of questions someone might find on the AWS Forums.
The building blocks here are organized into sample workflows such as workload monitoring, content analytics, and video quality checks.
Some, but not all, of the AWS resources used in this sample are supported under the AWS free tier.
The following services will incur charges in your account regardless of your free tier status:
MediaConvert, Kinesis Streams, Kinesis Firehose, Elasticsearch.
The total cost for running this sample depends on the size and number of media and metadata files For full details, see the pricing webpage for each AWS service you will be using in this sample.
Workflows are independent of one another unless noted in the Prerequisite section of the workflow README file.
-
Video ingest and conversion basic workflows
- MediaConvert watchfolder with SNS notifications (tutorial and CloudFormation): Automatically trigger video transcoding when a video is added to an S3 bucket. This workflow also has a companion tutorial to step through creating resources in the console.
-
Video conversion job metrics and monitoring
- Collecting MediaConvert job progress metrics Analyze and track the progress of MediaConvert jobs. Calculate new metrics about in-progress jobs from MediaConvert CloudWatch events.
- Monitoring MediaConvert job workloads Track and visualize MediaConvert workloads over time using CloudWatch events, Kinesis Streams, Kinesis Firehose and Elasticsearch
-
Video analysis
- Analyze MediaConvert inputs using MediaInfo (tutorial and CloudFormation) Collect and store mediainfo output whenever MediaConvert INPUT_INFORMATION CloudWatch events occur using Lambda and Dynamodb