An administrator adds a single new node to an existing 3-node AHV cluster. The cluster uses a default storage container with a Replication Factor (RF) of 2. The new node contributes 10 TB of raw flash capacity. The administrator expects the total usable capacity of the cluster to increase by 10 TB. However, the capacity dashboard shows only a 5 TB increase in usable capacity. What is the primary reason for this discrepancy?

Nutanix storage policies apply at write time. With Replication Factor 2, each new data block is written to two nodes so a single node failure can be tolerated. That means raw capacity is not the same as usable capacity: approximately half of the raw space added to an RF=2 container is consumed by the second replica. A 10 TB raw contribution therefore adds only about 5 TB of usable capacity for new workload data. Curator garbage collection is unrelated to this immediate capacity calculation; it reclaims space from deleted, overwritten, or stale blocks, not from newly added physical disks. Hypervisor overhead does reserve a small fixed amount of storage for AHV system files and operations, but it is not the dominant reason for a 50% reduction in an RF=2 environment. Separating a node into another container would isolate capacity from the existing policy, not reduce usable capacity in the original container; normal cluster expansion makes disks available to containers according to placement and policy. Exam caveat: always distinguish raw capacity from usable capacity after applying RF or erasure coding. Operational check: compare the container’s raw capacity and usable capacity in Prism, then recalculate expected usable gain using the container’s RF before approving expansion.