Skip to main content
Migration Notice
We're migrating documentation from the old portal into this one. Some things may look a little different or out of place in the meantime — we know, and we're working to get it right. If something's unclear or doesn't look right, let us know.
Version: 2.15.0

Concept for Data Orchestration for Dell

Introduction​

Superna Data Orchestration — built on the Golden Copy engine — is a PowerScale-integrated solution that provides high-speed, metadata-aware file-to-object copy and sync of files to native object storage over the S3 protocol. It is designed to move data intelligently between file and object storage in both directions: file-to-object for backup, archive, and cost-tiering, and object-to-file for recall, restore, and cloud-to-premises data mobility.

This page introduces what Data Orchestration does, the problems it solves, its core capabilities, and the terminology used throughout the rest of this documentation.

The Problem Data Orchestration Solves​

Enterprises increasingly need their file data to move fluidly between on-premises PowerScale storage and cloud or on-premises object storage, for several overlapping reasons:

  • 3-2-1 backup compliance: keeping a synced backup copy of file data in a separate storage tier (object storage) satisfies the requirement for an additional copy of data on different media.
  • Hybrid cloud data mobility: business processes and end users increasingly need to move data between on-premises SMB shares and cloud object stores — for cloud-based analytics, distributed teams, or cost-driven data placement — without needing separate tools for each direction.
  • Cost-driven data lifecycle management: not all data needs to stay on primary, high-performance storage. Aging or cold data can be archived to lower-cost object storage tiers, freeing up primary storage capacity.
  • Media and creative workflows: large media files often need to move between on-premises storage and the cloud to support burst compute, rendering, or collaboration workflows.

Data Orchestration addresses these needs with a single engine that understands both the file system and object storage sides of the equation — including file metadata (ownership, permissions, and timestamps) that a generic copy tool would otherwise lose.

Core Capabilities​

  • PowerScale-integrated sync: uses the PowerScale snapshot change list API to support fast incremental syncs, rather than re-scanning the entire file system on every job.
  • Direct Restore: restores data from object storage back to PowerScale and re-applies file metadata (owner, group, ACLs, and timestamps) directly to the file system, without an out-of-band copy step. Restore supports redirecting a recall to a different cluster, a different folder path, or filtering by a modified-date range.
  • Metadata protection: file metadata is automatically protected using S3 object metadata tags (ACLs, owner, group, and timestamps).
  • Data integrity verification: optional MD5 checksum support for copy jobs.
  • File system to object store auditing: audit jobs can compare file system and object store content.
  • Bandwidth control: rate limiting can be applied per copy job.
  • Flexible copy behavior: sync mode, copy mode, or both, with configurable handling of deletes — including a recycle mode that copies deleted files to a separate bucket with its own retention/TTL, rather than propagating deletes immediately to the object target.
  • Bi-directional hybrid cloud workflows: the Pipeline feature set enables data movement in both directions — on-premises to object storage and object storage back to on-premises file shares — on a schedule or based on detection of new or changed data.
  • End-user cloud data mobility: a web portal (Cloud Browser for Data Mobility) gives end users a single interface to view and move data between cloud object storage and on-premises SMB shares, while respecting SMB-level security so users only see data they are authorized to access.
  • Self-service archiving: the Archive Engine (a separately licensed capability) provides an end-user web portal for self-service archive and recall to cloud or on-premises object storage, along with administrator-defined age-based archive policies.

Key Terminology​

  • Virtual Accelerator Node (VAN): an additional Golden Copy VM deployed solely to increase copy throughput. The primary Golden Copy VM acts as the control node; VAN nodes handle copy work only. All nodes require an NFS mount to the source cluster.
  • Folder definition: a configured source path on PowerScale, together with its target object storage bucket/endpoint and copy settings. Folder definitions are the unit of configuration for what gets copied and where.
  • Copy job: an individual full or incremental copy (or recall) operation executed against one or more folder definitions.
  • Pipeline: the Data Orchestration feature set that enables bi-directional data movement (file-to-object and object-to-file) on a schedule or in response to detected changes, as opposed to one-way backup/archive.
  • Backup Bundle: the licensed capability set that enables synced backup use cases, including version-aware restore/recall.
  • Cloud Browser for Data Mobility (CBDM): the end-user-facing web portal for browsing and moving data between cloud object storage and on-premises file shares.
  • Archive Engine: the separately licensed self-service archiving capability, including the end-user archive/recall portal and administrator age-based archive policies.

How Data Orchestration Fits with Other Cyberstorage for Dell Solutions​

Data Orchestration is focused on data mobility, backup, and archive between file and object storage — a different problem from the threat detection, immutability, and site-failover problems addressed elsewhere in Cyberstorage for Dell:

  • AirGap isolates a last-resort, immutable copy of production data on a dedicated vault cluster. Data Orchestration's object storage copies are a complementary, separate backup tier — not a replacement for an immutable vault copy.
  • Disaster Recovery protects against site failure by replicating data to a DR cluster for rapid failover. Data Orchestration's object storage targets are not intended as a DR failover target; they serve backup, archive, and hybrid cloud mobility use cases instead.

See Use Cases for how these use cases translate into concrete deployment scenarios.

See Also​

  • Pre-Requisites – Platform, licensing, and infrastructure requirements before deploying Data Orchestration.
  • Design Guide – Architecture and performance planning for Golden Copy VMs and Virtual Accelerator Nodes.
  • Use Cases – Common scenarios for Data Orchestration.