redzilla
All tools

IP Addressing

Layer 2

Cabling

Performance

Wireless

CCTV

Optical fiber

Power

Datacenter

AV

Voice

Diagnostics

Calculation

Security

Datacenter

RAID Capacity

Enter the number of disks, the size of each one and the RAID level to get usable capacity, raw capacity, overhead, fault tolerance and the minimum number of disks. Validates counts and parity per level. Everything is computed in your browser.

TB

Assumes all disks are the same size.

Examples
redzilla.cl — raid 
 
  
 
 
Usable capacity
 
—
 
—
 
 
 
Fault tolerance
 
—
 
—
 
 
 
Overhead
 
—
 
parity/mirror capacity
 
 
  
 
 Usable —  
 
 Raw —  
 
 Tolerance —  
 
  
 
The same setup at other levels
 
 
  
LevelUsableOverheadToleranceMin.
   
 
 

With 4 disks of 4 TB.
          Flagged rows do not meet the minimum or the parity requirement.

 
 
 How it is calculated · capacity and tolerance per level 
 
 N number of disks · s
size per disk · g disks per group
            (RAID 50/60) · groups = N / g.

 
 RAID 0 usable = N·s · tolerance 0 · min 2.

 
 RAID 1 usable = s (mirror) · tolerance N−1 · min 2.

 
 RAID 5 usable = (N−1)·s · tolerance 1 · min 3.

 
 RAID 6 usable = (N−2)·s · tolerance 2 · min 4.

 
 RAID 10 usable = (N/2)·s · tolerance ≥1 (1 per
            mirror) · min 4, N even.

 
 RAID 50 usable = (N−groups)·s · tolerance = groups
            (1 per RAID 5 group) · min 6.

 
 RAID 60 usable = (N−2·groups)·s · tolerance 2·groups
            (2 per RAID 6 group) · min 8.

 
 JBOD usable = N·s · tolerance 0 · min 1 (disks are
            concatenated, no redundancy).

 
 Overhead = (raw − usable) / raw. Tolerance is the
            guaranteed worst case; with RAID 10/50/60 more failures may survive depending
            on which disks fail.

 
 
 
 
 
Runs locally in your browser · no sign-up · nothing leaves your browser.
  
 
 
How it works
 
The calculator takes the number of disks (N), the size of each one and the RAID level, and returns the usable capacity, the raw capacity, the overhead (percentage dedicated to parity or mirroring), the guaranteed fault tolerance and the minimum disks for the level. It covers RAID 0, 1, 5, 6, 10, 50, 60 and JBOD, validates each level requirements (even N in RAID 10, N a multiple of disks per group in 50/60) and compares the same configuration across all levels.
The formulas assume equal-size disks s: RAID 0 and JBOD use everything (N×s, tolerance 0); RAID 1 leaves only s (mirror, survives N−1 failures); RAID 5 subtracts one parity disk ((N−1)×s, survives 1); RAID 6 subtracts two ((N−2)×s, survives 2); RAID 10 leaves half (N/2×s, survives at least 1); and RAID 50/60 subtract 1 or 2 disks per each RAID 5/6 group combined in a stripe. Overhead is (raw − usable) ÷ raw.
 
 
Example: 4 disks of 4 TB in RAID 5
 
     
  1. Raw capacity: 4 × 4 TB = 16 TB.
  2. RAID 5 dedicates the equivalent of one disk to parity: usable = (4 − 1) × 4 = 12 TB.
  3. Overhead: (16 − 12) ÷ 16 = 25%, tolerating the failure of 1 disk.
    4.  
 
 
 
 
Frequently asked questions
 
 
 How much usable capacity does RAID 5 give with 4 disks of 4 TB? 
It gives 12 TB usable out of 16 TB raw: RAID 5 always subtracts the equivalent of one disk for parity, so the formula is (N−1) × size. With the same disks, RAID 6 would give 8 TB (subtracts two) and RAID 10 would give 8 TB (half).
 
 What is the difference between RAID 5 and RAID 6, and when should I use each? 
RAID 5 survives 1 disk failure and sacrifices one disk of capacity; RAID 6 survives 2 simultaneous failures and sacrifices two. With large disks (8 TB or more) a rebuild takes many hours and the risk of a second failure during that window is real, so RAID 6 is the usual recommendation for large arrays.
 
 Does RAID replace backups? 
No. RAID protects against physical disk failure, but not against accidental deletion, ransomware, data corruption or disasters affecting the whole system. The tolerance shown by the calculator is the guaranteed worst case of disks that can fail without data loss; backups remain mandatory.
 
 What does the group requirement mean in RAID 50 and RAID 60? 
RAID 50/60 stripe several RAID 5/6 groups together: the total disk count must be a multiple of the disks per group (minimum 3 per group in RAID 50 and 4 in RAID 60, with at least 2 groups). Guaranteed tolerance is 1 failure per group in RAID 50 and 2 per group in RAID 60; two failures inside the same RAID 5 group destroy the array.
 
 
 
  
   
 Was this tool useful? 
      
   
 
 
 Disclaimer We take great care to keep every tool accurate and review it thoroughly; even so, we can't guarantee it is free of errors or take responsibility for how the results are used. We recommend double-checking anything critical. 
 Found an error? Let us know →