SAN Capacity
Calculate usable storage area network capacity after RAID overhead and reserved space.
About SAN Capacity
Raw disk capacity in a SAN is reduced in three stages before it becomes usable: RAID overhead (parity or mirror copies needed for redundancy), one or more hot spares set aside to automatically rebuild after a drive failure, and metadata/filesystem overhead reserved by the array controller and file system structures. Each stage takes a bite out of what's left from the previous stage, so the order of subtraction matters — this calculator applies RAID overhead first, then reserves the hot spare from the RAID-protected capacity, then applies metadata overhead to what remains.
Formula
Usable capacity
usable = (raw × (1 − raid_overhead%) − hot_spare) × (1 − metadata_overhead%)
- R
- Raw capacity
- o_{raid}
- RAID overhead fraction
- S_{hot}
- Hot spare reservation
- o_{meta}
- Metadata overhead fraction
Worked example
With the default values below, SAN Capacity returns Usable Capacity of 68.87. Change any input above and every figure here is recalculated in your browser.
Inputs used
- Raw Capacityminimum 0 TB
- 100 TB
- RAID Overheadaccepts 0–90
- 25
- Hot Spare Reservationminimum 0 TB
- 4 TB
- Metadata / Filesystem Overheadaccepts 0–20
- 3
Results
- Usable Capacity
- 68.87 TB
- Usable Efficiency
- 68.9%
- Total Overhead (RAID + spare + metadata)
- 31.13 TB
- Capacity After RAID Overhead
- 75.00 TB
Step by step
- 1Values used
- 2Usable capacity
- 3Usable Capacity= 68.87 TB
- 4Usable Efficiency= 68.9
- 5Total Overhead (RAID + spare + metadata)= 31.13 TB
- 6Capacity After RAID Overhead= 75.00 TB
Assumptions and limits
- The result is derived only from the 4 inputs listed above. Anything not among them is not part of the calculation.
- In particular, this calculator does not take inflation, fees or charges or compounding frequency into account.
- Figures are rounded for display to 2 decimal places, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.
How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.
Frequently Asked Questions
Why does RAID overhead vary so much between RAID levels?
RAID 10 mirrors every drive, so it always costs 50% of raw capacity regardless of drive count. RAID 5 costs one drive's worth of parity per group (overhead shrinks as group size grows), and RAID 6 costs two drives' worth of dual parity for better fault tolerance — so an 8-disk RAID 6 group has roughly 25% overhead while the same group in RAID 5 has about 12.5%.
Do I need more than one hot spare?
Larger disk pools and higher-capacity drives (which take longer to rebuild) increase the odds of a second failure during a rebuild window, so many SAN admins reserve one spare per 20-30 drives or per shelf rather than a single global spare.
Why is metadata overhead applied after the hot spare instead of on raw capacity?
Metadata and filesystem structures are allocated on the capacity actually presented to hosts after redundancy and spares are accounted for, not on the raw disk capacity — so applying it last reflects how storage arrays actually carve up space.