Storage Autoscaling

Parent: MongoDB Atlas Cost Optimization · Published reference · snapshot 2026-10-15

Depth-first rabbithole dossier for Storage Autoscaling; source-anchored research pack.

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.

Definitions

[medium] On an Atlas Core cluster, cluster storage is the disk attached to each node, and `diskSizeGB` sets its size; Atlas delays an auto-scaling event until the oplog window covers the estimated scaling time. Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md, https://www.mongodb.com/docs/atlas/customize-storage.md [source]

How it works

[low] When you enable auto-scaling for compute and storage, Atlas automatically uses both reactive and predictive autoscaling to scale up your clusters. Sources: https://www.mongodb.com/docs/atlas/architecture/current/scalability.md [source]
[high] Atlas can reduce storage capacity by provisioning new volumes and syncing data from old to new volumes with an in-place data sync process, which it always uses for capacity decreases regardless of volume modification cooldown periods. Sources: https://www.mongodb.com/docs/atlas/customize-storage.md, https://www.mongodb.com/community/forums/t/scale-down-cluster-storage-with-no-downtime/312168 [source]
[low] Atlas generally will not scale a cluster if scaling would result in a tier outside the specified size range or if memory or CPU usage would exceed the capacity of the scaled tier, though cluster tier and storage might scale in parallel if disk needs require it. Sources: https://www.mongodb.com/docs/atlas/architecture/current/scalability.md [source]

Measurements and reference values

[medium] Atlas enables cluster storage auto-scaling by default, automatically increasing cluster storage when disk space used reaches 90% for any node in the cluster. Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md, https://www.mongodb.com/docs/atlas/customize-storage.md, https://www.mongodb.com/docs/atlas/sizing-tier-selection.md [source]
[low] In an auto-scaling event, Atlas might increase the cluster tier even if Cluster Tier Scaling is disabled if the current tier lacks the required storage capacity; for example, an M30 cluster at maximum 480 GB storage reaching 90% usage requires scaling to 600 GB, forcing an upgrade to M40. (verified as of 2026-10-15) Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md [source]
[low] Atlas NVMe clusters auto-scale to the next higher tier when 90% of the storage space is full. Sources: https://www.mongodb.com/docs/atlas/nvme-storage.md [source]

Problems, failure modes and limitations

[low] When Atlas overrides the maximum cluster tier, it disables the cluster from automatically scaling down; if the cluster is at the configured maximum tier, Atlas disables downscaling because smaller tiers cannot accommodate the storage, otherwise it raises the minimum cluster tier to the current tier. Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md [source]
[medium] Atlas Infinite clusters do not have Atlas Core storage auto-scaling because storage is decoupled from compute and grows independently; to control costs, you set a maximum storage size, and Atlas blocks writes when the cluster reaches that limit. Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md, https://www.mongodb.com/docs/atlas/sizing-tier-selection.md [source]
[low] The new volume size can't exceed the supported capacity of its file system and partitioning scheme, and there are limits to the maximum aggregated storage that can be requested across volume modifications. Sources: https://docs.aws.amazon.com/ebs/latest/userguide/ebs-modify-volume.html [source]
[low] During public preview, Atlas might throttle writes before a cluster reaches the configured storage limit to preserve system availability. (verified as of 2026-10-15) Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md [source]
[low] For Azure clusters that support Extended Storage/IOPS, you can make up to four disk IOPS/throughput changes within any 24-hour period; if this limit is reached, Atlas replaces the virtual machine to bypass the limitation. Sources: https://www.mongodb.com/docs/atlas/customize-storage.md [source]
One source states Atlas can reduce storage capacity in place by provisioning new volumes and syncing data, while another states you cannot scale down storage directly in Atlas on AWS but must migrate to a smaller cluster. Sources: https://www.mongodb.com/community/forums/t/scale-down-cluster-storage-with-no-downtime/312168 [source]

Comparisons and alternatives

[low] Unlike AWS, Google Cloud does not have volume modification cooldown periods, but it does not allow in-place storage capacity reduction. Sources: https://www.mongodb.com/docs/atlas/customize-storage.md [source]

Facts and statements

[medium] Atlas auto-scales cluster storage up only; you can manually reduce cluster storage from the Edit Cluster page, but AWS and Azure do not allow in-place storage capacity reduction. Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md, https://www.mongodb.com/docs/atlas/customize-storage.md [source]
[low] By default, Atlas sets the minimum oplog retention window to 24 hours, and for storage auto-scaling to complete successfully, Atlas requires the minimum oplog retention window to be up to (60 seconds) * (GB of disk space configured). Sources: https://www.mongodb.com/docs/atlas/customize-storage.md [source]
[low] If a cluster's storage capacity decreases, Atlas automatically scales the oplog size down to ensure it fits in the decreased storage size. Sources: https://www.mongodb.com/docs/atlas/customize-storage.md [source]
[low] With auto-scaling, your cluster automatically scales its tier, storage capacity, or both in response to cluster usage, reducing the need to manually optimize your cluster to adapt to your current workload. Sources: https://www.mongodb.com/docs/atlas/billing/cluster-configuration-costs.md [source]
[low] On Atlas Core clusters, scaling of clusters that use the local NVMe SSD storage option requires an initial sync. Sources: https://www.mongodb.com/docs/atlas/nvme-storage.md [source]
[low] MongoDB offers storage compression, enabled by default with WiredTiger, the default MongoDB storage engine. Sources: https://www.vantage.sh/blog/mongodb-cost-optimization-tips [source]
[low] Changes to a cluster are applied in a rolling fashion, starting with the secondaries, and often require migrating to new servers and storage volumes. Sources: https://www.mongodb.com/community/forums/t/scale-down-cluster-storage-with-no-downtime/312168 [source]
[low] For development and testing environments, do not enable auto-scaling compute and auto-scaling storage to save costs in non-production environments. Sources: https://www.mongodb.com/docs/atlas/architecture/current/scalability.md [source]
[low] If a cluster receives a burst of high-speed write activity, such as a bulk insert, automatic scaling might not occur due to a temporary spike in disk storage capacity. Sources: https://www.mongodb.com/docs/atlas/cluster-autoscaling-storage.md [source]

Related concepts

Storage — is a part of Storage Autoscaling
Autoscaling — is a part of Storage Autoscaling

Children

No children recorded.

← the whole tree · 3D view· how to read this page