Data replication to the cloud

Learn about the advantages and challenges of replicating your data to cloud storage for processing.

The first common hybrid model is based on replicating data. Here, enterprise data and metadata are copied from the on-premise environment to cloud storage, typically object stores like Amazon S3, Azure Blob Storage, or Google Cloud Storage. Processing is performed entirely in the cloud, using the replicated data, as shown in the diagram.

This architecture is a good fit when an organisation can clearly separate workloads between the cloud and on-premise, and is willing to perform a lift-and-shift migration of a subset of applications to the cloud.

However, for temporary burst to cloud, this architecture leads to complex overheads in maintaining data copies, increased data storage costs, data inconsistencies between data copies, and duplication of security and governance contexts.