Backup snapshot storage growth cost
Parent: Atlas Cluster Topology and Tier Economics · Published reference · snapshot 2026-10-22 · skill ranked-87a9ba58
↓ Facts as markdownall context files
Source-anchored research for Backup snapshot storage growth cost. Run status: COMPLETED.
These notes link each claim to its source. Some citations identify archived research reports that are no longer hosted on this site. A published reference means the content is available; it does not certify independent review or accuracy.Read the editorial policy and follow the sources before relying on a claim.
Structure and components
- Atlas Core can retain multiple full snapshots after a node or region failover, tier/storage/IOPS configuration change, backup-region priority change, provider maintenance, or topology change such as adding a shard. This reduces the share of incremental data and raises storage cost. Depending on provider reporting, Atlas can bill up to the whole volume capacity; its example bills 4 TB when the provider reports a 4 TB volume even though the backups contain only 500 GB. (high confidence) (verified as of 2026-10-20) [source]
How it works
- Atlas Core Cloud Backup snapshots are usually incremental: after the first snapshot, a new snapshot generally stores only changed data, and the cloud provider determines the exact incrementality. AWS EBS also uses changed-block snapshots. For Azure Blob Storage without smart tier, shared blocks are billed once across a base blob and its snapshots only when their tiers have not been explicitly set or changed; smart-tier accounts bill snapshots at full content length. Snapshot count alone is therefore not a reliable storage-growth estimate. (high confidence) [source]
- Write churn can grow retained snapshot storage even when logical database content changes little. Atlas says application access patterns affect snapshot incrementality; Azure bills rewritten blocks as new unique data even if their content and ID match an earlier block. In one October 2026 AWS EBS account, column-store compaction changed about 97.7 GiB per day on 1.28 TB volumes, producing daily deltas close to full copies; this is an EBS case, not an Atlas measurement. (medium confidence) (verified as of 2026-10-01) [source]
- Atlas Infinite uses a separate backup meter: Atlas measures uncompressed backup storage hourly in GB-hours, includes 24 hours of continuous cloud backup, and bills additional retained snapshots or incremental changes when retention exceeds that window. (low confidence) (verified as of 2026-10-14) [source]
- Atlas bills cross-region data transfer and additional storage when backups are copied to other regions. If all backups are copied to one additional region, the primary backup and its copy together incur approximately twice the backup storage: the copy adds approximately the same storage amount as the primary. Storage rates vary by region. (medium confidence) (verified as of 2026-10-14) [source]
Measurements and reference values
- Atlas may bill backup storage up to the total capacity of the volume if the cloud provider reports the snapshot footprint as the full volume capacity, even when the protected data is smaller. Atlas documents an example where a 4 TB volume with 500 GB of backup data is billed as 4 TB. (low confidence) (verified as of 2026-10-14) [source]
Facts and statements
- Atlas Core calculates Cloud Backup charges from the total snapshot storage reported by the cloud provider, per GB-month, with rates that vary by provider and region; invoices show usage in daily GB-days. A live-data-size × retained-snapshot-count estimate is not a reliable bill forecast. (low confidence) (verified as of 2026-10-20) [source]
- Atlas says deleting one snapshot does not remove unique data while another retained snapshot still references it. Its documented cost controls include deleting unneeded snapshots, lowering snapshot frequency or retention, reducing copies to additional regions, and improving application write patterns to improve snapshot incrementality. (high confidence) [source]
- Atlas Core Continuous Cloud Backup has a separate invoice line from standard Cloud Backup. Its billed footprint combines snapshot storage with oplog storage, and a longer point-in-time recovery window increases oplog size and backup cost; provider, region, and usage tier also affect the rate. (high confidence) (verified as of 2026-10-20) [source]
- For Atlas Core Cloud Backups, snapshots are usually incremental after the first: later snapshots generally store only changed data, and the cloud provider determines the exact incrementality. AWS EBS and Azure managed-disk documentation describe a first full snapshot followed by changed-data snapshots; Google Cloud likewise defaults to incremental snapshots. Live database size multiplied by retained snapshot count is therefore not a dependable estimate of stored backup bytes. (high confidence) [source]
- Retained change data drives snapshot storage growth. Higher write churn can add more changed blocks to the retained snapshot history, while longer retention keeps that history for more time. Atlas lists reducing snapshot frequency, retention, or copies to additional regions as cost controls; cloud-provider documentation confirms that incremental snapshot sizes reflect changed data. (high confidence) [source]
- Atlas can retain more than one full snapshot per replica set or shard after node or region failover, cluster configuration or topology changes, region-priority changes, or cloud-provider maintenance. This reduces the incremental share and can increase storage and cost. If a snapshot storage volume becomes invalid, Atlas creates a new volume and takes a full-copy snapshot before continuing incrementally. (medium confidence) (verified as of 2026-10-14) [source]
Children
- No children recorded.