CompTIA A+ guideCore 1 skills
RAID and Storage for A+: RAID 0, 1, 5, 6, 10 and Drive Types
Work out usable capacity and fault tolerance for every RAID level on 220-1201, then tell NVMe, SATA, M.2, mSATA, and SAS apart.
Short answer
RAID 0 stripes data with no redundancy and gives you all the capacity, while RAID 1 mirrors, so a two-drive mirror gives one drive of space and survives one failure. RAID 5 needs at least three drives, gives up one drive's worth of space to parity, and survives one failure; RAID 6 needs at least four, gives up two, and survives any two. RAID 10 mirrors pairs of drives and stripes across them, needs at least four, and gives you half the space. With four 2 TB drives that is 8 TB for RAID 0, 6 TB for RAID 5, and 4 TB for RAID 6 or RAID 10, and none of them replaces a backup.
Where RAID and drives sit on Core 1
Two 220-1201 objectives cover this page. Objective 3.4 compares storage devices: hard drives, solid-state drives and their interfaces and form factors, RAID 0, 1, 5, 6, and 10, removable storage, and optical drives. Objective 5.2 is the troubleshooting side: drive and RAID symptoms such as a missing array, a RAID failure, audible alarms, and S.M.A.R.T. warnings. Both sit in the two largest Core 1 domains, Hardware (25%) and Hardware and Network Troubleshooting (28%).
Most RAID questions come down to two calculations: how much space you get, and how many drives can fail before data is lost.
The five levels in one table
The capacity column assumes every drive is the same size, s, and that there are n drives.
| Level | How it works | Minimum drives | Usable capacity | Drive failures it survives |
|---|---|---|---|---|
| RAID 0 | Stripes data across drives. No redundancy | 2 | n × s | None. One failure loses the array |
| RAID 1 | Every drive holds the same data | 2 | s | All but one drive |
| RAID 5 | Stripes data with parity spread across all drives | 3 | (n − 1) × s | One |
| RAID 6 | Like RAID 5 with two parity blocks per stripe | 4 | (n − 2) × s | Any two |
| RAID 10 | Mirrors pairs of drives, then stripes across the pairs | 4 | (n ÷ 2) × s | One per mirrored pair |
The Linux md(4) manual page describes the same behavior: RAID0 "has zero redundancy," each RAID1 device "contains exactly the same data," RAID5 distributes parity across all devices and survives one failure, and RAID6 "can handle the loss of any two devices" and "requires N+2 drives to store N drives worth of data."
Worked example: four 2 TB drives
| Level | Calculation | Usable | What happens if two drives fail |
|---|---|---|---|
| RAID 0 | 4 × 2 TB | 8 TB | Array lost. It was lost after the first failure |
| RAID 1 (two of the drives) | 2 TB | 2 TB | A two-drive mirror is lost if both drives fail |
| RAID 5 | (4 − 1) × 2 TB | 6 TB | Array lost |
| RAID 6 | (4 − 2) × 2 TB | 4 TB | Array survives, running degraded |
| RAID 10 | (4 ÷ 2) × 2 TB | 4 TB | Depends on which two |
RAID 6 and RAID 10 give the same 4 TB here, so the question becomes which failures you need to survive. Label the RAID 10 drives A1 and A2 (first mirror) and B1 and B2 (second mirror).
- A1 and B1 fail: each mirror still has one good drive. The array survives.
- A1 and A2 fail: the first mirror is gone, and the stripe needs it. The array is lost.
RAID 6 survives either pair, because it survives any two failures. RAID 10 is the usual choice when write speed matters more, since RAID 5 and RAID 6 have to calculate and write parity on every write.
Worked example: six 4 TB drives
| Level | Calculation | Usable |
|---|---|---|
| RAID 5 | (6 − 1) × 4 TB | 20 TB |
| RAID 6 | (6 − 2) × 4 TB | 16 TB |
| RAID 10 | (6 ÷ 2) × 4 TB | 12 TB |
As the drive count grows, parity levels waste a smaller share of the space, while RAID 10 always keeps exactly half.
Mixed sizes. If an array mixes a 2 TB drive with 4 TB drives, most RAID implementations treat every member as the size of the smallest. md(4) states it plainly for mirrors: "All devices in a RAID1 array should be the same size." A question that mixes sizes is usually testing whether you notice the wasted space.
Failure, rebuild, and why RAID is not a backup
When a drive in a redundant array fails, the array keeps running in a degraded state on the remaining drives. md(4) describes the driver marking the device faulty and continuing, then rebuilding onto a spare drive if one is available. Replacing the failed drive, or letting a hot spare take over, starts a rebuild that reads the surviving drives. On RAID 5, a second failure before the rebuild finishes loses the array.
RAID protects against a drive failing. It does not protect against a user deleting a folder, ransomware encrypting files, or a fire. Those changes are written to every member at once. That is why Core 2 tests backups separately, in the backup types guide.
RAID can be built by a hardware controller or by the operating system. On Windows, Microsoft lists Dynamic Disks as no longer being developed and points to Storage Spaces as the replacement. On Linux, md provides software RAID.
Drive types that 3.4 expects you to tell apart
The most common confusion is treating a form factor, an interface, and a protocol as the same thing.
| Term | What it is | What to remember |
|---|---|---|
| HDD, 2.5-inch or 3.5-inch | Spinning platters, rated by spindle speed in RPM | Higher RPM generally means faster access. 3.5-inch is the desktop size, 2.5-inch the laptop size |
| SATA | Interface used by HDDs and many SSDs | The same data cable fits a SATA HDD or a 2.5-inch SATA SSD |
| PCIe | The expansion bus that fast SSDs connect to | Also carries graphics, network, and other add-on cards |
| NVMe | A protocol for talking to non-volatile memory | The NVM Express specifications define how host software communicates with storage across transports such as PCIe. NVMe is not a connector |
| SAS | Serial Attached SCSI, used mainly in servers | Expect it in server and storage scenarios, not home PCs |
| M.2 | A small card form factor and slot | An M.2 slot can be wired for SATA, for PCIe NVMe, or for both. Check the motherboard manual |
| mSATA | An older, smaller SATA card form factor | Found in older laptops. It is not the same slot as M.2 |
| Flash drives and memory cards | Removable storage | Covered with optical drives under removable media |
A typical exam trap: a technician installs an NVMe SSD in an M.2 slot that supports only SATA, and the drive never shows up. The form factor fits, but the slot does not speak the drive's protocol.
Symptoms to connect to this page
Objective 5.2 lists the drive and RAID symptoms, and the hardware and network troubleshooting guide gives each one a first check, with a worked "array missing" case. Two ideas from this page carry straight into those questions. An alarm from the RAID controller means a redundant array is running degraded, so identify the failed member before you change anything. And a missing array is a configuration question before it is a dead-drive question: check the controller or firmware settings, including the storage mode, before you assume the drives failed.
Motherboard connectors, RAM, and power supplies are in the hardware components guide.
Run the four-drive table from memory once more, then practice A+ Core 1 hardware questions, including RAID capacity, with Passy.
Trust the source
Official sources
Exam policies can change. Use these primary sources for the most current details.