vSAN ESA/OSA Capacity Calculator
Raw → usable capacity for vSAN ESA and OSA. Accounts for RAID overhead, slack space, and FTT.
Architecture notes
- OSA (Original Storage Architecture): cache + capacity tiers, vSAN 7 and earlier.
- ESA (Express Storage Architecture): flat NVMe pool, vSAN 8+. Better RAID-5/6 performance.
- FTT (Failures to Tolerate): how many node failures the cluster can survive without data loss.
- Slack space: vSAN needs free space for rebuild and rebalance operations.
How to calculate vSAN usable capacity
This calculator converts raw vSAN capacity into usable capacity by applying the RAID overhead of your storage policy, then reserving slack space. The formula: usable = (nodes × raw_per_node) × RAID_efficiency × (1 − slack).
RAID efficiency by policy: RAID-1 FTT=1 stores data twice (50% efficient). RAID-1 FTT=2 stores it three times (33%). RAID-5 FTT=1 uses single parity (75%, ESA recommended). RAID-6 FTT=2 uses double parity (66.7%).
ESA vs OSA: on vSAN 8 ESA, RAID-5/6 erasure coding performs close to RAID-1, so the capacity savings of RAID-5 come without the old OSA write penalty. On OSA most teams stay on RAID-1 despite the higher capacity cost.
Slack space: VMware recommends keeping 25–30% free for rebuilds and rebalancing. This calculator reserves 25% by default.
Frequently asked questions
How much usable capacity does RAID-5 give on vSAN ESA?
RAID-5 FTT=1 is about 75% efficient, so 100 TB raw yields roughly 75 TB after RAID and about 56 TB usable after 25% slack. ESA makes RAID-5 practical because erasure coding no longer carries the OSA write penalty.
What's the capacity difference between OSA and ESA?
The raw-to-usable math is the same per policy; the difference is that ESA makes RAID-5/6 performant enough to actually use, so you can choose capacity-efficient policies that were impractical on OSA.
How much slack space does vSAN need?
VMware recommends 25–30% free capacity for rebuild and rebalance operations. Running with less risks performance degradation and failed rebuilds during a node loss.
What does FTT mean for capacity?
FTT (Failures to Tolerate) sets how many node failures the cluster survives. Higher FTT means more redundancy and lower capacity efficiency — FTT=2 RAID-1 stores data three times (33% efficient) versus twice for FTT=1 (50%).