Understanding RAID groups

About RAID groups in Data ONTAP

A RAID group consists of one or more data disks, across which client data is striped and stored, plus one or two parity disks. The purpose of a RAID group is to provide parity protection from data loss across its included disks. RAID4 uses one parity disk to ensure data recoverability if one disk fails within the RAID group. RAID-DP uses two parity disks to ensure data recoverability even if two disks within the RAID group fail.

RAID group disk types

Data ONTAP assigns and makes use of four different disk types to support data storage, parity protection, and disk replacement.

Disk	Description
Data disk	Holds data stored on behalf of clients within RAID groups (and any data generated about the state of the storage system as a result of a malfunction).
Hot spare disk	Does not hold usable data, but is available to be added to a RAID group in an aggregate. Any functioning disk that is not assigned to an aggregate functions as a hot spare disk.
Parity disk	Stores data reconstruction information within RAID groups.
dParity disk	Stores double-parity information within RAID groups, if RAID-DP is enabled.

Levels of RAID protection

Data ONTAP supports two levels of RAID protection, RAID4 and RAID-DP, which you can assign on a per-aggregate basis.

If an aggregate is configured for RAID4 protection, Data ONTAP reconstructs the data from a single failed disk within a RAID group and transfers that reconstructed data to a spare disk.
If an aggregate is configured for RAID-DP protection, Data ONTAP reconstructs the data from one or two failed disks within a RAID group and transfers that reconstructed data to one or two spare disks as necessary.

RAID4 protection

RAID4 provides single-parity disk protection against single-disk failure within a RAID group. The minimum number of disks in a RAID4 group is two: at least one data disk and one parity disk. If there is a single data or parity disk failure in a RAID4 group, Data ONTAP replaces the failed disk in the RAID group with a spare disk and uses the parity data to reconstruct the failed disk's data on the replacement disk. If there are no spare disks available, Data ONTAP goes into a degraded mode and alerts you of this condition.

Attention
With RAID4, if there is a second disk failure before data can be reconstructed from the data on the first failed disk, there will be data loss. To avoid data loss when two disks fail, you can select RAID-DP. This provides two parity disks to protect you from data loss when two disk failures occur in the same RAID group before the first failed disk can be reconstructed.

The following figure diagrams a traditional volume configured for RAID4 protection.

RAID-DP protection

RAID-DP provides double-parity disk protection when the following conditions occur:

There are media errors on a block when Data ONTAP is attempting to reconstruct a failed disk.
There is a single- or double-disk failure within a RAID group.

The minimum number of disks in a RAID-DP group is three: at least one data disk, one regular parity disk, and one double-parity (or dParity) disk.

If there is a data-disk or parity-disk failure in a RAID-DP group, Data ONTAP replaces the failed disk in the RAID group with a spare disk and uses the parity data to reconstruct the data of the failed disk on the replacement disk. If there is a double-disk failure, Data ONTAP replaces the failed disks in the RAID group with two spare disks and uses the double-parity data to reconstruct the data of the failed disks on the replacement disks. The following figure diagrams a traditional volume configured for RAID-DP protection.

How Data ONTAP organizes RAID groups automatically

When you create an aggregate or add disks to an aggregate, Data ONTAP creates new RAID groups as each RAID group is filled with its maximum number of disks. Within each aggregate, RAID groups are named rg0, rg1, rg2, and so on in order of their creation. The last RAID group formed might contain fewer disks than are specified for the aggregate's RAID group size. In that case, any disks added to the aggregate are also added to the last RAID group until the specified RAID group size is reached.

If an aggregate is configured for RAID4 protection, Data ONTAP assigns the role of parity disk to the largest disk in each RAID group.
Note
If an existing RAID4 group is assigned an additional disk that is larger than the group's existing parity disk, then Data ONTAP reassigns the new disk as parity disk for that RAID group. If all disks are of equal size, any one of the disks can be selected for parity.
If an aggregate is configured for RAID-DP protection, Data ONTAP assigns the role of dParity disk and regular parity disk to the largest and second largest disk in the RAID group.
Note
If an existing RAID-DP group is assigned an additional disk that is larger than the group's existing dParity disk, then Data ONTAP reassigns the new disk as the regular parity disk for that RAID group and restricts its capacity to be no greater than that of the existing dParity disk. If all disks are of equal size, any one of the disks can be selected for the dParity disk.

Hot spare disks

A hot spare disk is a disk that has not been assigned to a RAID group. It does not yet hold data but is ready for use. In the event of disk failure within a RAID group, Data ONTAP automatically assigns hot spare disks to RAID groups to replace the failed disks. Hot spare disks do not have to be in the same disk shelf as other disks of a RAID group to be available to a RAID group.

Hot spare disk size recommendations

Keep at least one hot spare disk for each disk size and disk type installed in your storage system. This allows the storage system to use a disk of the same size and type as a failed disk when reconstructing a failed disk. If a disk fails and a hot spare disk of the same size is not available, the storage system uses a spare disk of the next available size up. See Disk failure and RAID reconstruction with a hot spare disk for more information.

Note
If no spare disks exist in a storage system, Data ONTAP can continue to function in degraded mode. Data ONTAP supports degraded mode in the case of single-disk failure for aggregates configured with RAID4 protection and in the case of single- or double- disk failure in aggregates configured for RAID-DP protection. For details see Disk failure without a hot spare disk.

Maximum number of RAID groups

Data ONTAP supports up to 400 RAID groups per storage system or cluster. When configuring your aggregates, keep in mind that each aggregate requires at least one RAID group and that the total of all RAID groups in a storage system cannot exceed 400.

RAID4, RAID-DP, and SyncMirror

RAID4 and RAID-DP can be used in combination with the Data ONTAP SyncMirror feature, which also offers protection against data loss due to disk or other hardware component failure. SyncMirror protects against data loss by maintaining two copies of the data contained in the aggregate, one in each plex. Any data loss due to disk failure in one plex is repaired by the undamaged data in the opposite plex. The advantages and disadvantages of using RAID4 or RAID-DP, with and without the SyncMirror feature, are listed in the following tables.

Advantages and disadvantages of using RAID4:

Factor affected by RAID level	RAID4	RAID4 with SyncMirror
What RAID and SyncMirror protect against	Single-disk failure within one or multiple RAID groups	Single-disk failure within one or multiple RAID groups in one plex and single-, double-, or greater-disk failure in the other plex. A double-disk failure in a RAID group results in a failed plex. If this occurs, a double-disk failure on any RAID group on the other plex fails the aggregate. See Advantages of RAID4 with SyncMirror. Storage subsystem failures (HBA, cables, shelf) on the storage system
Required disk resources per RAID group	n data disks + 1 parity disk	2 x (n data disks + 1 parity disk)
Performance cost	None	Low mirroring overhead; can improve performance
Additional cost and complexity	None	SyncMirror license and configuration Possible cluster license and configuration

Advantages and disadvantages of using RAID-DP:

Factor affected by RAID level	RAID-DP	RAID-DP with SyncMirror
What RAID and SyncMirror protect against	Single- or double-disk failure within one or multiple RAID groups	Single-disk failure and media errors on another disk. Single- or double-disk failure within one or multiple RAID groups in one plex and single-, double-, or greater disk failure in the other plex. SyncMirror and RAID-DP together cannot protect against more than two disk failures on both plexes. It can protect against more than two disk failures on one plex with up to two disk failures on the second plex. A triple disk failure in a RAID group results in a failed plex. If this occurs, a triple disk failure on any RAID group on the other plex will fail the aggregate. See Advantages of RAID-DP with SyncMirror. Storage subsystem failures (HBA, cables, shelf) on the storage system
Required disk resources per RAID group	n data disks + 2 parity disks	2 x (n data disks + 2 parity disks)
Performance cost	Almost none	Low mirroring overhead; can improve performance
Additional cost and complexity	None	SyncMirror license and configuration Possible cluster license and configuration

Advantages of RAID4 with SyncMirror

On SyncMirror-replicated aggregates using RAID4, any combination of multiple disk failures within single RAID groups in one plex is restorable, as long as multiple disk failures are not concurrently occurring in the opposite plex of the mirrored aggregate.

Advantages of RAID-DP with SyncMirror

On SyncMirror-replicated aggregates using RAID-DP, any combination of multiple disk failures within single RAID groups in one plex is restorable, as long as concurrent failures of more than two disks are not occurring in the opposite plex of the mirrored aggregate.

For more SyncMirror information

For more information on the Data ONTAP SyncMirror feature, see the Data Protection Online Backup and Recovery Guide.

Larger versus smaller RAID groups

You can specify the number of disks in a RAID group and the RAID level of protection, or you can use the default for the specific storage system. Adding more data disks to a RAID group increases the striping of data across those disks, which typically improves I/O performance. However, with more disks, there is a greater risk that one of the disks might fail.

Configuring an optimum RAID group size for an aggregate requires a trade-off of factors. You must decide which factor--speed of recovery, assurance against data loss, or maximizing data storage space--is most important for the aggregate that you are configuring. For a list of default and maximum RAID group sizes, see Maximum and default RAID group sizes.

Advantages of large RAID groups

Large RAID group configurations offer the following advantages:

More data drives available. An aggregate configured into a few large RAID groups requires fewer drives reserved for parity than that same aggregate configured into many small RAID groups.
Small improvement in storage system performance. Write operations are generally faster with larger RAID groups than with smaller RAID groups.

Advantages of small RAID groups

Small RAID group configurations offer the following advantages:

Shorter disk reconstruction times. In case of disk failure within a small RAID group, data reconstruction time is usually shorter than it would be within a large RAID group.
Decreased risk of data loss due to multiple disk failures. The probability of data loss through double-disk failure within a RAID4 group or through triple-disk failure within a RAID-DP group is lower within a small RAID group than within a large RAID group.

For example, whether you have a RAID group with fourteen disks or two RAID groups with seven disks, you still have the same number of disks available for striping. However, with multiple smaller RAID groups, you minimize the risk of the performance impact during reconstruction and you minimize the risk of multiple disk failure within each RAID group.

Advantages of RAID-DP over RAID4

With RAID-DP, you can use larger RAID groups because they offer more protection. A RAID-DP group is more reliable than a RAID4 group that is half its size, even though a RAID-DP group has twice as many disks. Thus, the RAID-DP group provides better reliability with the same parity overhead.