https://blog.coffeeordeath.dev/tags/tech/ Recent content in Tech on Coffee or Blog Hugo -- gohugo.io en Mon, 21 Apr 2025 00:00:00 +0000 https://blog.coffeeordeath.dev/posts/zfs-raid-types/ Mon, 21 Apr 2025 00:00:00 +0000 https://blog.coffeeordeath.dev/posts/zfs-raid-types/ If I didn&rsquo;t have to spend a bunch of time moving virtual machines around, this probably wouldn&rsquo;t matter very much. But, alas, I did find myself building zfs raid arrays a few times and couldn&rsquo;t seem to remember what I wanted where. 🧂 Take with a conservative grain of salt. RAID Type Min Disks Usable Capacity Fault Tolerance Performance Notes Notes Striped 1+ 100% of total None Fastest read/write, no redundancy Equivalent to RAID0 Mirror 2+ 50% of total 1 disk per mirror vdev Excellent read, good write, fast recovery Equivalent to RAID1 in traditional RAID RAIDZ1 3+ N - 1 1 disk failure Slower write, decent read Similar to RAID5 RAIDZ2 4+ N - 2 2 disk failures Slower write, decent read Similar to RAID6 RAIDZ3 5+ N - 3 3 disk failures Slowest write, decent read For high fault tolerance ZFS RAID10 4+ (even) 50% of total 1 disk per mirror vdev Best balance of performance + redundancy Stripe of mirrors (manual mirror vdevs) dRAID 3+ Varies (RAIDZ-like) Parity-based (configurable) Improved resilver vs. <p>If I didn&rsquo;t have to spend a bunch of time moving virtual machines around, this probably wouldn&rsquo;t matter very much. But, alas, I did find myself building zfs raid arrays a few times and couldn&rsquo;t seem to remember what I wanted where.</p> <p>🧂 Take with a conservative grain of salt.</p> <table> <thead> <tr> <th>RAID Type</th> <th>Min Disks</th> <th>Usable Capacity</th> <th>Fault Tolerance</th> <th>Performance Notes</th> <th>Notes</th> </tr> </thead> <tbody> <tr> <td><strong>Striped</strong></td> <td>1+</td> <td>100% of total</td> <td>None</td> <td>Fastest read/write, no redundancy</td> <td>Equivalent to RAID0</td> </tr> <tr> <td><strong>Mirror</strong></td> <td>2+</td> <td>50% of total</td> <td>1 disk per mirror vdev</td> <td>Excellent read, good write, fast recovery</td> <td>Equivalent to RAID1 in traditional RAID</td> </tr> <tr> <td><strong>RAIDZ1</strong></td> <td>3+</td> <td>N - 1</td> <td>1 disk failure</td> <td>Slower write, decent read</td> <td>Similar to RAID5</td> </tr> <tr> <td><strong>RAIDZ2</strong></td> <td>4+</td> <td>N - 2</td> <td>2 disk failures</td> <td>Slower write, decent read</td> <td>Similar to RAID6</td> </tr> <tr> <td><strong>RAIDZ3</strong></td> <td>5+</td> <td>N - 3</td> <td>3 disk failures</td> <td>Slowest write, decent read</td> <td>For high fault tolerance</td> </tr> <tr> <td><strong>ZFS RAID10</strong></td> <td>4+ (even)</td> <td>50% of total</td> <td>1 disk per mirror vdev</td> <td>Best balance of performance + redundancy</td> <td>Stripe of mirrors (manual mirror vdevs)</td> </tr> <tr> <td><strong>dRAID</strong></td> <td>3+</td> <td>Varies (RAIDZ-like)</td> <td>Parity-based (configurable)</td> <td>Improved resilver vs. RAIDZ, good balance</td> <td>Requires newer ZFS versions</td> </tr> <tr> <td><strong>Special VDEV</strong></td> <td>1+</td> <td>Not for main storage</td> <td>Depends on layout</td> <td>High IOPS for metadata/small files</td> <td>Used for performance, not redundancy</td> </tr> </tbody> </table> <blockquote> <p>N = Number of disks in the vdev</p> </blockquote>