This post forms part of a three-part series on how we migrate content to AWS for Media and Entertainment and Broadcast customers. This post addresses the second stage of the migration, a media audit having already been done.

Part 2 of 3 – Metadata Extraction and Database Creation

Tape Ark has developed a unique technology stack to help customers migrate large legacy tape collections to AWS. We deploy our customized Mass Ingest Workflow to move the customer’s visual assets from physical media to AWS, depositing it in the customer’s bucket.

Tape Ark extracts several kinds of metadata from the customer’s tape collection to help assist them in managing their assets. To create and establish a database in the cloud, Tape Ark uses Amazon Simple Queue Service (SQS) and Dynamo DB, among other tools. The tape catalogue files are extracted using AWS Lambda and interrogated to populate an SQS database with extraction instructions based on the file size and memory requirements. We develop custom AWS Lambda/Serverless routines to facilitate the resolution of missing, duplicate or multi-segment file issues – the reconciliation of files spanning across multiple tapes. Checksums are automatically generated and cross-referenced against pre-existing checksums, enabling the customer to confidently verify that the assets received are the same as the originals. DynamoDB is used to keep track of movement so that all the data ends up where it should.

In addition to converting the customer’s media content and transferring it to the AWS, Tape Ark can also create low-resolution proxy clips that allow the customer to view their videos from highly available storage at a reduced cost. The higher resolution broadcast videos are then put in latent storage like Amazon S3 Glacier, and only retrieved after watching the proxy and once they are sure that they indeed want the clip. This allows for both cost savings and high availability at the same time.

Read Part 3 here: