Cloud Storage
Cloud Storage in Flashback is the object-storage layer of the Cloud and AI Gateway. It lets you connect buckets/containers from multiple providers behind one operational model, so storage resources can be consumed consistently by repositories, automation, and AI/LLM workflows.
At a high level:
A Bucket in Flashback is the representation of a remote vendor’s object storage resource.
Depending on the provider, this resource can be called a bucket (S3, GCS) or a container (Azure Blob).
Flashback standardizes access to those resources while preserving provider-specific connectivity and identity options.
What Flashback supports for buckets and object storage
Flashback supports object storage resources exposed through:
AWS S3 endpoints.
S3-compatible APIs from external or on-prem providers through custom endpoints.
Google Cloud Storage (GCS) endpoints.
GCS-compatible APIs from external or on-prem providers through custom endpoints.
Microsoft Azure Blob storage accounts and containers.
This allows you to build multi-cloud, hybrid, and on-prem storage topologies where repositories can target heterogeneous storage backends from a single gateway layer.
Buckets, containers, and object storage fundamentals
Object storage is designed around:
Objects (data + metadata),
grouped in buckets/containers,
addressed through HTTP APIs,
and controlled through API credentials or delegated identities.
In Flashback, this is especially useful when your storage resources are used alongside AI/LLM patterns (for example: grounding data, archived corpora, binary artifacts, logs, and generated outputs), because you can connect storage providers with consistent governance and repository-level operations.
Configure bucket resources
For exact creation fields, provider-specific parameters, access mechanisms, storage-type constraints, and warnings, use the detailed guide:
You can also configure buckets via API:
Security and secret encryption
Secret encryption is a platform-wide security principle in Flashback, not only a Cloud Storage behavior.
For the full model (how secrets are encrypted, when decryption happens, and why this applies to buckets, AI providers, and other connected resources), see:
Last updated
Was this helpful?