The concept of storage policies is central to the usage and protection of data within the Vault v2 platform.
The policies determine or impact:
- how data is stored within the system upon ingress
- how data is made available in the event of a site-wide disaster
- the overall amount of data capacity consumed by an object and the associated costs passed onto the consuming organisation.
Vault v2's service resilience capabilities consist of Replication and Erasure Coding (EC). In general EC is more storage efficient than replication and has a lower storage utilisation overhead. For more information about the service resilience capabilities available from the platform, download the Vault v2 White Paper here.
Storage policies are assigned at the bucket level depending on your organisation's internal application, business, commercial and risk requirements. CCL allows you to store data within Vault v2 using the following methodologies:
|Three-site Erasure Coded||
In addition to selecting a replication or erasure coding scheme, a consistency level can be selected for write operations. Consistency levels impose requirements on the portion of the data and metadata associated with each S3 request, that must be completed successfully, before the system can return a successful response to the S3 clients.
Vault v2 supports object versioning. It can be enabled on buckets so that each version of an object in the bucket is retained. This includes both the current and prior versions.
For example, if a user uploads a newly created document and then uploads further revised versions of the document, the system will store each individual version of the document.
Versioning protects organisations against losing an object due to accidentally overwriting it - since the previous version is retained as well as the new version. Versioning also allows access to all the past versions of an object throughout its history.
For more information see: