RAID 60 Capacity Calculator for Home Servers

So you start out with your set of drives and know everything should be usable storage. You plug in the drives and format the array, but then you find there’s a massive drop in usable storage. What happened? It didn’t get stolen. It got sacrificed for your data safety.

That’s where RAID 60 comes into play. It’s the safety net, basically combining several sets of RAID 6 arrays. Each of those RAID 6 groups has two drives dedicated for parity protection. You’ll see how much it sacrifices in terms of capacity on the calculator.

How to Build a Safe Storage System

Capacity can’t be judged by drive count alone; it’s also about drive grouping. A drive added to an unbalanced grouping doesn’t always give you more usable space. Symmetry is best in RAID 60. Two groups of four drives might comprise eight drives in all. But because each group has two drives dedicated to parity, only half your total space are usable.

That’s how this type of drive are set up, but people don’t think about it. They think: how much do I have? They do not ask how it is laid out. More is better, within reason. More drives and fewer groups provide greater storage efficiency different than fewer drives and more groups. You still lose parity for every two drives in a group, but the cost is spread out over more data drive.

You can see how this would work with the layout tool which help you visualize the design before computing your storage requirements. It’s a system you’re building, not just storing stuff into a box.

There should of be a storage reserve. You can save between ten and twenty percent of your space for that with the calculator. In case big files gets uploaded or the snapshots eat all your space, this will give you some breathing room. That buffer is the difference between a healthy system and one that locks up during a scrub. You don’t need it, so it’s like an insurance policy. And it comes at the price of storage, but protects against locking up the system.

Heat matter in an array because of the density. The tool calculates how much heat each drive will emit at what speed it run. If it spins faster, it heats up more and finishes rebuilding sooner. If it spins slower, it doesn’t heat up as much but takes longer to rebuild. Air flow trumps storage capacity in a crowded chassis. Any drive going down is bad news. A drive overheating while rebuilding are worse.

Rebuild cost and speed depend on the drive type. Shingled magnetic recording drives is slow and cheap, while solid-state drives are quick and expensive but fail different. Tool has a reference table to pair the workload with the drive type. Don’t put virtual machines on a slow drive. Don’t use fast drives for cold storage. Match the media to the job.

Extra disk protection comes with hot spares. Extra drives assigned as spare is waiting in case one of your drives fails and needs replacing. This is useful if you have large numbers of disks. If you have a small number, it’s a safety net. If you don’t have a spare and you lose a drive while rebuilding, you will lose all data in that group when a second drive fails.

It’s trying to balance speed with safety, heat and cost. It takes your numbers (in either decimal or binary) and translates them into other systems. This ensures the OS reports back the correct amount of space. You won’t have to wonder if the manufacturer’s specs are accurate or if your system matches them.

First, plan out how you want the group laid out. Then decide on what kind of media to use to make it go, leaving some room for maintenance. Ignoring parity costs will come back to bite you in the future. Verify the efficiency and ensure that the heat production doesn’t exceed the cooling capacity. Finally, fill up the rest.