RAID 60
RAID 60 Quick Facts
What Is RAID 60?
RAID 60, also called RAID 6+0, is a nested RAID architecture. Multiple independent RAID 6 groups are created, and a RAID 0-style stripe is then placed across those groups.
Each underlying RAID 6 group maintains two independent sets of distributed parity. This means each group can tolerate the failure of up to two of its member drives. Striping across multiple RAID 6 groups allows the larger RAID 60 virtual disk to distribute I/O across several parity groups.
Multiple RAID 6 Groups Are Striped Together
The smallest conventional RAID 60 configuration uses two four-drive RAID 6 groups for a total of eight drives. Each group contains data plus dual parity, while the top-level stripe spans both groups.
Larger RAID 60 arrays can use more drives per group, additional RAID 6 groups, or both, subject to the capabilities and limits of the RAID controller or software implementation.
Simplified conceptual illustration of an eight-drive RAID 60 composed of two four-drive RAID 6 groups.
RAID 60 Usable Capacity
Each RAID 6 group consumes the equivalent capacity of two member drives for dual parity. The usable capacities of all RAID 6 groups are then combined by the top-level stripe.
For equal-sized groups, the same relationship can also be written as (total drives − 2 × number of RAID 6 groups) × smallest drive capacity.
Eight 4 TB Drives
Two 4-drive RAID 6 groups:
2 × (4 − 2) × 4 TB = 16 TB nominal RAID 60 capacity .
Twelve 8 TB Drives
Two 6-drive RAID 6 groups:
2 × (6 − 2) × 8 TB = 64 TB nominal RAID 60 capacity .
RAID 60 Can Survive Multiple Drive Failures
Every RAID 6 group can tolerate up to two member-drive failures. A RAID 60 array can therefore survive multiple simultaneous failures when no individual RAID 6 group loses more than two members.
Failure Placement Still Matters
RAID 60 does not provide unlimited multi-drive fault tolerance. If a third required member of the same RAID 6 group fails before sufficient redundancy is restored, that parity group can fail and the complete RAID 60 virtual disk can become unavailable.
“Two Drives Per Group” Is More Accurate Than a Single Array-Wide Number
Consider a RAID 60 array with two RAID 6 groups. The array could potentially survive four simultaneous member-drive failures if exactly two failed drives are located in each group.
However, three failures concentrated in one of those RAID 6 groups can exceed that group's dual-parity protection. This is why RAID 60 fault tolerance should be described in terms of failures per underlying RAID 6 group rather than simply stating a fixed number for the entire array.
RAID 60 Adds Parallelism Across Multiple RAID 6 Groups
The top-level stripe allows reads and writes to be distributed across multiple RAID 6 groups. This can provide greater aggregate throughput than a single RAID 6 group when the controller, interfaces, drives, and workload can take advantage of the additional parallelism.
RAID 60 still carries the dual-parity write overhead of RAID 6 within every group. Small random writes can therefore require substantially more work than non-parity layouts such as RAID 10. Controller cache, media type, stripe size, queue depth, workload pattern, and implementation can significantly affect real-world results.
Rebuilds Occur Within the Affected RAID 6 Group
When a member drive fails, reconstruction uses the surviving data and parity within the affected RAID 6 group. Other groups do not need to reconstruct the failed member's contents.
Dual Parity Preserves Additional Protection
After the first member failure within a RAID 6 group, that group still retains protection against one additional member failure. This is one of RAID 60's major resilience advantages over RAID 50.
Single Parity per Group vs Dual Parity per Group
RAID 50 stripes across RAID 5 groups, with each group able to tolerate one member failure. RAID 60 stripes across RAID 6 groups, with each group able to tolerate two member failures.
RAID 60 therefore provides greater group-level fault tolerance, but consumes the equivalent capacity of two drives per group for parity instead of one and performs additional parity work during writes.
RAID 60 Strengths
- Dual-parity protection within every RAID 6 group.
- Can survive multiple failures when no group loses more than two members.
- Retains additional fault tolerance after the first failure within a group.
- Provides parallelism across multiple RAID 6 groups.
- Well suited to large arrays where resiliency is a major priority.
- Smaller parity groups can limit the scope of individual rebuild operations.
RAID 60 Limitations
- Requires at least eight drives.
- Consumes two drives' worth of capacity per RAID 6 group for parity.
- Dual-parity calculations create write overhead.
- More complex to plan than basic RAID levels.
- Fault tolerance still depends on how failures are distributed among groups.
- RAID 60 does not replace independent backups.
When RAID 60 May Be Appropriate
RAID 60 is commonly considered for larger storage systems where dual-parity protection, capacity, and aggregate throughput are more important than minimizing drive count or parity write overhead.
RAID 60 Compared With RAID 6, RAID 10 & RAID 50
| RAID Level | Architecture | Minimum Drives | Redundancy | Primary Tradeoff |
|---|---|---|---|---|
| RAID 6 | Single dual-parity group | 4* | Any 2 drives | Strong redundancy with dual-parity write overhead |
| RAID 10 | Striped mirrors | 4 | Mirror-topology dependent | Strong performance with approximately 50% efficiency |
| RAID 50 | Striped RAID 5 groups | 6 | 1 drive per RAID 5 group | Higher efficiency with single-parity groups |
| RAID 60 | Striped RAID 6 groups | 8 | 2 drives per RAID 6 group | Greater resilience with additional parity capacity and write overhead |
* RAID-level support and minimum drive requirements can vary by controller or software implementation. Standard RAID 6 can be implemented with four drives, while some specific platforms may impose higher minimums.
Calculate Your RAID 60 Configuration
Use the Tech Supply Direct RAID Calculator to compare RAID 60 capacity, storage efficiency, parity overhead, fault tolerance, estimated performance, hot-spare impact, and hardware cost.
RAID 60 FAQ
How many drives are required for RAID 60?
RAID 60 requires at least two RAID 6 groups. With four drives per minimum RAID 6 group, the minimum RAID 60 configuration contains eight drives.
How many drives can RAID 60 lose?
Each underlying RAID 6 group can tolerate up to two failed members. Multiple groups can therefore experience failures simultaneously as long as no group exceeds two failed members.
Can RAID 60 survive four drive failures?
A two-group RAID 60 can potentially survive four failures if no more than two failed drives occur in either RAID 6 group. Four failures are not guaranteed to be survivable because failure placement matters.
What is the difference between RAID 50 and RAID 60?
RAID 50 stripes across RAID 5 groups with single parity. RAID 60 stripes across RAID 6 groups with dual parity. RAID 60 therefore provides greater fault tolerance within each underlying group.
Is RAID 60 faster than RAID 6?
RAID 60 can provide greater aggregate performance by distributing I/O across multiple RAID 6 groups. Actual performance depends on the controller, media, interfaces, group layout, cache, and workload.
Is RAID 60 a backup?
No. RAID 60 provides storage redundancy against supported drive-failure combinations but does not replace an independent and tested backup strategy.
RAID 60 Redundancy Is Not a Backup
RAID 60 provides substantial drive-failure tolerance, but RAID redundancy does not protect against every cause of data loss. Accidental deletion, corruption, ransomware, application errors, controller or system failures, catastrophic hardware loss, and other events can still affect the array. Important data should also be protected with independent, tested backups.