The repository shall have adequate hardware and software support for backup functionality sufficient for preserving the repository content and tracking repository functions.

This is necessary in order to ensure continued access to and tracking of preservation functions applied to the digital objects in their custody.

Documentation of what is being backed up and how often; audit log/inventory of backups; validation of completed backups; disaster recovery plan, policy and documentation; fire drills; testing of backups; support contracts for hardware and software for backup mechanisms; demonstrated preservation of system metadata such as access controls, location of replicas, audit trails, checksum values.

The repository should be able to demonstrate the adequacy of the processes, hardware, and software for its backup systems and the full range of ingest, preservation, and dissemination functions required of a repository entrusted with long-term preservation. Simple backup mechanisms must preserve not only the repository main content, but also the system metadata generated by the preservation functions. Repositories need to develop backup plans that ensure their continuity of operations across all failure modes.

APTrust provides for backups and DR of content and repository functionality in multiple ways and at multiple layers for very high resiliency. 

All deposited content is stored/hosted on S3, Wasabi, and Glacier, each of which provides either erasure coding or object replication across many drives and/or multiple datacenters. Depositors have the option of depositing to multiple or a single location.

Traditional backups are done for 7 days of the Postgres database, with backups replicated in an automated fashion to the US-West-2 Region (Oregon) for DR This database has records of all the transactions, so the history of the system is maintained as well. 

Any need for persistent data in the containers is managed by mounted storage with replication across multiple data centers within the US East-1 region (Virginia). (Replication.) 

Data resiliency for other services is provided by multi-zone (multi-datacenter) lifecycles on AWS. 

APTrust has a designated technical team to be sure that backup processes are working and reviewed for appropriateness on a regular basis. Additional support resources are available with AWS due to the Business support Plan utilized by APTrust.