RAID 0: what it is and how it works

Learn what RAID 0 is, how striping works, how its capacity is calculated, the risks involved, when to use it and what to do if the array fails.

Is your computer too slow for what you need? It is one of the most common complaints, especially among people who run heavy programs that put a lot of demand on the system. But improving your hard drive’s performance doesn’t always mean spending a fortune. One option is RAID 0, which combines two or more hard drives or SSDs into a single, faster volume.

With RAID and NAS, a rebuild or an initialization without a diagnosis can reduce the chances of recovery. Before touching anything, see Bot’s RAID and NAS data recovery, which starts with a free diagnosis.

RAID is a technique developed in the 1980s that is still widely used today. That is partly because the basic principle of how hard drives work hasn’t changed much since then, even though capacity and speed have grown enormously. RAID usually has one of two goals: making a storage system faster or making it more reliable.

When it comes to speed, RAID 0 is the best choice. RAID 1, on the other hand, is designed to improve reliability. The two can also be combined in what is known as RAID 10. In this article, we focus on the first one. Read on to see how this technique can help you get more out of your drives.

What is RAID 0?

RAID stands for Redundant Array of Inexpensive (or Independent) Disks. Put simply, RAID levels are different ways of combining storage drives so they work together.

RAID 0 is one of these levels. It combines two or more drives and splits the data written to them across all of the drives in the array. Because each drive only has to write part of every file, the array as a whole is significantly faster.

In practice, with two drives, it is as if each drive only had to handle half of the file, which greatly speeds up writing (and reading).

How does RAID 0 work?

Still not sure how the data is split? Let’s take a closer look. RAID 0 uses a technique called striping: data is divided into fixed-size blocks (stripes), and these blocks are written alternately to each drive in the array. Since the drives read and write their blocks at the same time, overall read and write speeds increase.

Here is a simple comparison: it is like reading a thousand pages on your own versus splitting them with a friend so that each of you reads half at the same time. You will get to the end much faster.

Capacity

The storage capacity of a RAID 0 array depends on the drives in it. With identical drives, simply multiply the capacity of one drive by the number of drives. For example, two 1 TB drives give you 2 TB.

But be careful: if you combine a smaller drive with a larger one, the smaller drive sets the limit, because every drive must hold an equal share of the data. A 2 TB drive paired with a 4 TB drive, for example, gives a total of only 4 TB (twice the smaller drive), and the remaining 2 TB of the larger drive cannot be used by the array.

Speed

As we have seen, the main advantage of RAID 0 is speed. With two drives, sequential read and write speeds can come close to double those of a single drive. However, the gain depends on the type of workload: with many small, random accesses (as on some shared, multi-user systems), the improvement may be much smaller. That’s why it is important to check whether RAID 0 really suits your needs.

Data safety

Another factor to consider is data safety. Because the data is split across two or more drives, a failure in just one of them can make everything inaccessible. For this reason, this type of array is not recommended for storing important data, or any data that isn’t backed up.

What are the risks of using RAID 0?

As we explained, RAID 0 has no redundancy: it neither mirrors data nor stores parity. This means that if one of the drives fails permanently, recovering your files becomes much more difficult. That is the main risk of this type of array.

In RAID 1, by contrast, the data is copied from one drive to another. It doesn’t deliver the speed gains of RAID 0, but it does improve reliability. So it is important to decide what you need most before choosing one of these setups.

Can you build a RAID 0 array with different drives?

You may have a spare hard drive at home and want to try RAID 0. But be careful: it isn’t always a good idea. When you mix drives of different brands, models or capacities, the array is held back by the weakest member: it runs at the pace of the slowest drive and only uses the capacity of the smallest one, multiplied by the number of drives.

In addition, different drives may tolerate heat, vibration, physical shocks and other common causes of failure differently, and in RAID 0 a single failure is enough to lose access to the whole array.

What is RAID 0 mainly used for?

As you have seen, speed is RAID 0’s greatest advantage, while reliability is not its strong point. That is why it is best suited to situations where performance matters more than data safety.

Examples include professional video and audio editing, rendering, post-production, scratch and temporary files, and other demanding workloads where you want high performance at a low cost. What matters is keeping up-to-date backups of these files.

How do you recover data from a RAID 0 array?

As you have seen, hard drives can fail. And if one of the drives in the array starts to fail, all your data may become inaccessible. In that case, the first step is to find out whether a drive has simply been disconnected or whether it is faulty.

To do this, shut down the computer and check that every drive is properly connected (data and power cables). Then turn it back on and see if the problem is solved. If the error persists, one of the drives is very likely failing. Do not rebuild, reinitialize or recreate the array, as this can overwrite the information needed to recover it. Instead, seek out a professional data recovery service as soon as possible.

Recovery software, especially on a RAID 0 array, can end up putting your files at even greater risk. Steer clear of quick fixes from the internet and never open a drive yourself: outside a cleanroom, this only causes more physical damage, which is very difficult to reverse.

Companies like Bot offer safe, professional data recovery, with more than two decades of experience, more than 200,000 cases and a 96% success rate. A specialist laboratory has the equipment to image each drive and rebuild the array without writing to the originals, so your chances are much better with experienced professionals.

Want to know more about Bot? Follow us on Instagram, Twitter and Facebook to learn more about our work!

Need to recover the files? Bot runs a free diagnosis in its lab: see our RAID recovery service.

About the author
Filipe BragaGestor de laboratório · Laboratório Bot

Filipe Braga is an engineer and runs the Bot lab. He works on the most complex data recovery cases, mainly RAID systems and servers, where every drive counts and the order of steps matters. Day to day, he organises the bench work so that each device follows the right procedure. On the blog, he writes about RAID and difficult cases. Outside the lab, he enjoys cycling and a good wine.

All articles by Filipe Braga →

Need to recover your files?

The diagnosis identifies the fault and whether recovery is technically viable. You receive a quote before any work begins.

Start diagnosis
Share