TechKnowSurge
TechKnowSurge Catalog Videos Architecting RAID
VideoSecurityFree

Architecting RAID

Selecting the right RAID configuration depends on balancing performance, redundancy, and workload requirements. This content covers practical RAID selection criteria and a common real-world server storage architecture.

⚑ Complete this video to capture a CTF flag worth 1 point.

About this video

Choosing a RAID configuration requires weighing performance demands against redundancy requirements, and the right answer varies depending on the workload. RAID 0 uses pure striping to maximize read and write speeds but provides no fault tolerance, which makes it a poor choice in most enterprise contexts where data availability is a priority. RAID 1 mirrors data across drives for strong redundancy but does not improve performance, limiting its usefulness for high-throughput applications like databases. RAID 5 and RAID 6 are the more commonly deployed options in production environments because they combine striping with parity, delivering both performance and fault tolerance. RAID 6 extends this further by supporting two simultaneous drive failures, making it appropriate for mission-critical storage systems where continuous availability is essential. A practical and widely adopted server architecture separates the operating system from application data entirely. The OS is installed on a small RAID 1 mirror, typically using modest-capacity drives, so that a single drive failure does not take the system offline. Recovery in this scenario requires only a drive replacement rather than a full restore from backup. Data volumes are then provisioned separately, either carved from additional RAID 1 capacity for lighter workloads or deployed on dedicated RAID 5 or RAID 6 arrays when performance and scalability matter. Keeping the OS and data on distinct volumes also prevents runaway data growth from consuming the system partition, which can destabilize or completely disable the operating environment. For demanding workloads such as database servers or storage area networks, expanding the number of disks in a RAID 5 or RAID 6 array directly increases throughput, giving administrators a straightforward path to scaling performance alongside capacity.

What you'll learn

What's covered

RAID Configuration Strategy

Key terms

RAID 0
A RAID configuration that stripes data across two or more drives to improve read and write performance, but provides no fault tolerance — losing any one drive loses all data.
RAID 1
A RAID configuration that mirrors identical data simultaneously across two drives, providing full redundancy so that the array continues operating if either drive fails.
RAID 5
A RAID configuration that stripes data and distributed parity across three or more drives, providing read performance gains and the ability to survive the failure of any single drive.
RAID 6
A RAID configuration that stripes data across four or more drives with two independent sets of parity, allowing the array to survive the simultaneous failure of any two drives.
Redundancy
The duplication of critical components or systems to increase reliability and availability.
Striping
A RAID technique that splits data across multiple drives sequentially to improve read and write performance.

Topics

Raid Storage Architecture Data Redundancy Server Storage Disk Performance Storage Management

Transcript

There are so many different RAID configurations that we could set up, so which one do we use? Some of that depends on our use cases.

Choosing a RAID Level

We have a lot of options when it comes to RAID. However, RAID zero only gives us performance, because it just does the striping. So I find that we don't use it a whole lot on enterprise networks, because usually we want some sort of redundancy within our data.

RAID one I do find that we use, but not for really high volume traffic like databases, because it doesn't have that striping and so there's no performance increase.

Usually I'm setting things up with RAID five if I just want one parity disk, or RAID six if I want to make sure things are both performing well and also have some sort of redundancy within that data and being able to self-heal. So typically what I would do is a RAID five or RAID six with many disks, depending on how much speed I'm trying to get out of the reads and writes — for instance, database servers or storage area networks. I have a lot of disks that are involved with it.

A Common Server Setup

Maybe this is best illustrated with a common setup that I use, and that is when I am setting up a server, I will have the main operating system set on a mirrored drive. I don't need a bunch of space for this, so maybe I'll get a couple of 256 or 512 drives, just enough to handle the operating system on here, and then I'll install this operating system on this RAID one configuration, so it's mirrored.

What this guards against is to make sure that if something were to happen to the server itself and one of these drives dies, then I can easily bring it back up and running without going to backup. I have the operating system already on the drive, and then I could just replace one of these drives to bring it back up.

Keeping Data Separate from the Operating System

From the data perspective, if this is something that's simple, then maybe I'll just carve out an area of this RAID one drive right here — create another volume on it that is specifically for data. I like to keep my data separate from the operating system, because if the data overruns the drive and then brings my operating system down, that can be really painful. It's really painful just to fill up a drive in general, but it's even more painful when I have a hard time even working and bringing up the operating system to do anything. So I really do like to keep the data separate from the operating system.

Now let's say we have something like a database server. I'm probably going to do maybe even a RAID six or something even more powerful than this. But if it's just data that I want to store that's significant, but I want some redundancy, some performance with it, then here's a RAID five. I've got the striping that's happening with the par bit, so I have redundancy involved with it. This is a very common setup that I set up in a lot of my servers.

About TechKnowSurge

TechKnowSurge builds IT and cybersecurity professionals through hands-on, concept-first training built around real understanding — not memorization. Free interactive tools, structured programs, and 25+ years of real-world experience, all in one place.

Explore free tools and programs →