Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.coreweave.com/llms.txt

Use this file to discover all available pages before exploring further.

Distributed File Storage (DFS) enforces a storage quota per CoreWeave Kubernetes Service (CKS) cluster, which sets an upper limit on the total provisioned size of DFS-backed PersistentVolumeClaims (PVCs) in that cluster.

​
Quotas

By default, each CKS cluster has a maximum DFS quota of 100 TiB. This is the total provisioned capacity across all DFS PVCs in that cluster.
QuotaDefault Value
Maximum capacity (per CKS cluster)100 TiB
You can monitor your DFS storage usage and quotas per cluster in the DFS Usage dashboard in Grafana.

​
How quotas affect PVCs

You can create and resize DFS-backed PVCs as long as the sum of provisioned DFS capacity in a cluster remains within that cluster’s quota. If a request to create or resize a PVC would exceed the quota, the request fails with a clear error instead of over-provisioning the cluster.

​
Request a quota increase

If your workloads require more storage than your current quota allows, contact CoreWeave support to request an increase before PVC creation is blocked. To request a quota increase, contact CoreWeave support and include:
  • Your organization name and Org ID, found in the Settings page.
  • The region and zone.
  • The affected CKS cluster names.
  • The requested DFS quota (TiB) for each cluster.
  • Optional: A brief description of the workloads and expected storage growth (for example, new clusters or changes over the next 30 to 90 days).

​
Storage capacity alerts

CoreWeave DFS sends alerts when your provisioned storage approaches or is projected to exceed your cluster quota, giving you time to act before PVC creation is blocked.
Alert nameConditionSeverity
DFSQuotaProjectedNearLimitProvisioned storage is projected to exceed 90% of quota within 7 daysWarning
DFSQuotaNearLimitProvisioned storage exceeds 90% of quotaWarning
DFSQuotaProjectedLimitProvisioned storage is projected to exceed 100% of quota within 7 daysCritical
When multiple alerts fire at the same time for the same cluster and zone, only the highest-severity alert fires and the others are suppressed. Each notification covers a single cluster in a single zone, so multiple clusters or zones approaching the limit produce separate notifications. To set up Slack or webhook delivery, see CoreWeave Alerts.

​
PromQL for custom monitoring

You can use the following PromQL expressions against your Grafana datasource to build your own dashboards or alerting rules.

​
Current provisioned storage as a percentage of quota

sum by (cluster_org, cluster, zone) (
  pvmo_watcher_bound_pvc_bytes{provisioner="csi.vastdata.com"}
  + pvmo_watcher_inflight_pvc_bytes{provisioner="csi.vastdata.com"}
)
/ on(cluster_org, cluster, zone)
sum by (cluster_org, cluster, zone) (
  pvmo_quota_validation_quota_bytes{
    mode="enabled",
    provisioner="csi.vastdata.com"
  }
)
* 100

​
Projected provisioned storage in 7 days as a percentage of quota

The following example uses avg_over_time over a 7-day window sampled hourly, which matches the projection method used by the DFSQuotaProjectedNearLimit and DFSQuotaProjectedLimit alerts: 
avg_over_time(
  (
    sum by (cluster_org, cluster, zone) (
      pvmo_watcher_bound_pvc_bytes{provisioner="csi.vastdata.com"}
      + pvmo_watcher_inflight_pvc_bytes{provisioner="csi.vastdata.com"}
    )
  )[7d:1h]
)
/ on(cluster_org, cluster, zone)
sum by (cluster_org, cluster, zone) (
  pvmo_quota_validation_quota_bytes{
    mode="enabled",
    provisioner="csi.vastdata.com"
  }
)
* 100
Last modified on May 26, 2026