Storage Solutions in Windows
This is a guide on storage solutions in Windows
Compact Disk Storage
To start with, we will take a look at the compact disk storage medium. This was released in the early 1980's as a portable storage medium for video, audio, and other data. This allowed you to record whatever data you wanted, then store it, and play it back.
One of the advantages, at least compared to something like a floppy disk, was the amount of storage. There was much more room on a compact disk than there was on a floppy. We will come back to that in a moment. The essence of how a cd-rom worked was to use a laser. As long as the disk itself had two different ways to reflect that laser, then this is how you could translate those two different reflective states to a 1 or 0.
Now, despite the fact that they did have a much greater capacity than something like a floppy disk, they were still prone to damage. The surface was quite easily scratched. It could stand up against a few small scratches without too much difficulty, but if there was a very deep scratch, then usually that disk became unreadable. So you had to treat them fairly gently.
In terms of the dimensions and the design, it was a pretty small size. Hence the name compact disk. It was about 4.7 inches in diameter or 120mm. It was very light, only 15 to 20 grams. As mentioned, the storage was far greater than what you could put on a floppy disk. Floppy disks back in those days were barely 1.5 Mb of storage. So now, you can have 6550 to 700 Mb. So, it was a tremendous increase at that point in time as compared to a floppy disk.
There were several different types of compact disks. The original was a cd-read only memory, or cd-rom. Now, this came with the data already on it. For example, if the disk contained a software installation, it was already there and you could not change it. It was in that state for its entire lifespan. If it was a music cd, the music was already on it. So, you could not put anything on that disk, you could just read it.
Then we saw the introduction of the cd-recordable. This essentially was a blank disk. This allowed you to write the information to the disk and read it many times. You could only write it one time. So once you filled up that disk, you could not overwrite it with anything else. Now, that of course, was a drawback. Compared to something like a floppy disk, you could write as many times as you wanted. Again, the capacity was a tremendous advantage over a floppy disk.
Then we saw the cd-rewritable, which did in fact allow you to write something and then overwrite it again. So, now we have the same capability as a floppy disk, where we can save new information over top of what was already there. Still, of course, you could read it as many times as you wanted. For both the recordable and the rewritable, of course what you needed was what we called a cd burner. It used the latter to alter the chemical state of the disk itself to essentially write or rewrite the 1 or the 0. So, the laser, having made that change to the chemical disk, could quite literally reflect the light differently.
The cd-rom extended architecture just allowed for multiple types of data to be on the same disks. So audio, video, and data, could all be on the same disk. Then, there was also a specific cd for video and cdv. This was typically implemented along with the mpeg file format. It had higher resolution and allowed the video to be stored in an uncompressed state, but still with the capacity of only 650-700 Mb in total. That did not allow for very much video. It was only a few minutes of uncompressed video.
DVD Storage
Now, in this section, we will talk about the digital versatile disc, or dvd. It can be audio, video, or data.
So this is still another type of optical disc storage that was developed in 1995. Like its predecessor, cd-roms, it also had a few variations in the format. You might see dvd-r or dvd+r. Now the r stood for recordable. Then below that, you can see the -rw and the +rw. Which is recordable and writable or rewritable. The dash versus the plus denotes a difference.
Now, when this format was first coming out, they were not compatible with each other. So, if you had a -R device, you needed -R discs. Of course, the same goes for +R. So you could not mix and match them. The only difference was in how they determined the placement of the laser on the disc. There really was not any difference in the performance or capacity. It was literally just how they determined where to put the laser. Later revisions came out with, in many cases, what you would see as both. A couple different ways, you might see the + sign over top of the dash. You might also see a question mark in between, which meant that it supported either one. So most later players and readers did support both.
Then finally, dvd-ram, that does stand for random access memory, but that does not mean it was used like the ram in your computer. It was simply how the data was written. In the previous formats of R and RW, the data was written in one long spiral track, very much like an old vinyl record for those who remember those. Whereas the dvd-ram was much more like a hard drive. It had concentric rings or tracks. You could easily go from track to track. The difference was that for the R and RW's, you needed specific software installed on your computer to be able to work with those formats. With dvd-ram, you did not. It was treated really just like a hard drive if you will, in terms of accessing the data. So you in fact did not need any specific software.
Now, this was adopted fairly early by the movie and home entertainment industry as a replacement for VHS. You certainly could store a lot more information on dvd's than you could on cd's and they were certainly more portable than VHS tapes. You would often find gaming consoles using them as well. Again, because of the increased capacity.
You could also store many different kinds of data on the dvd. Again, that comes back its name, digital versatile disc. So, plain old data, video, audio, could all be stored. It did not have the capacity to store high definition video. Now, some of you may think I have a high definition dvd's at home right now. You certainly may, but that is a specific type. That is an HD DVD. In this case, we are just talking about the original dvd, which did not support high definition video.
Now, like many technologies when they first came out, they were a little more expensive, but they were adopted quite quickly. So the manufacturing costs did drop significantly. They are still cheaper than some of the newer technologies such as Blue-Ray. They also shared the same physical dimensions as compact discs. They were the exact same size. In fact, most dvd players can still read compact discs as well.
As for the capacity, the physical size again, was the exact same as cd-roms. They were 12 cm or 120 mm in diameter. The standard capacity for a dvd was 4.7 Gb. So that is significantly larger than its predecessor. Cd-roms, which were only 650-700 mb. So greatly increased capacity. In fact, you could also get dual layer dvd's, whereby different intensities of the laser would penetrate more deeply into the discs. So you could vary the intensity of the laser to read layer one and layer two. In which case, you could effectively double the capacity.
As far as the data rates were concerned, they did range quite a bit as technology improved. Overall, 1.4 to 32.2 Mb/s when reading or writing. It really was that increased capacity compared to cd-roms that allowed dvd's to essentially replace that medium.
Blu-Ray Discs
In this section, we will talk about the Blu-Ray disk format. This is another optical disk data storage format that draws its name from the fact that it quite literally uses a blue laser, compared to a red laser, which was used for cd's and dvd's.
Now, if you want to get particular about it, it actually is a little closer to violet, but Blu-Ray is a much better name. That is actually important, because blue light has a higher frequency compared to red. That translates into a more tightly focused beam and in turn, more data on the disc.
So this effectively became the replacement for dvd's. It supports high definition and ultra-high definition content. It is primarily used for movies and video games, but it does share the same physical size as dvd's and cd's in terms of its dimensions. It is still 12 cm in diameter and 1.2 mm in thickness.
Now like the predecessors, there are variations in the format. There is a Blu-Ray Disc recordable or R. This, of course, just means that you can overwrite the existing data. There is a mini-Blu-Ray disc which is physically smaller. It is only 8 cm in diameter as opposed to 12. So, of course, less data overall. It is even more compact.
Now both BD9 and BD5 were somewhat backward compatible if you will, in that they allowed Blu-Ray information to be written to dvd media. At the time, it was still a little bit cheaper. So you could in fact use a Blu-Ray with a dvd physical disc. Now, neither of those caught on particularly well because Blue-Ray overtook dvd so quickly that you will not see too much of those. The BD9 was able to work with the 8.5 Gb size specification and BD5 only the 4.7 Gb.
The IH-BD is intra-hybrid. That has one layer that is only readable and another layer that is writable. Then BDXL, for the most part, you will generally hear referred to as extra-large because it really just is increased capacity. It introduced a third and a fourth layer to get even more storage.
So in terms of the capacity with just a single layer Blu-Ray disc, you can get up to 25 Gb. So again, that is already significantly larger than dvd. Then with the dual layer, you can get up to 50 Gb. Now by the time you get to the XL3, the triple layer essentially, it goes all the way up to 100 Gb.
If the previous layers were essentially 25 Gb each, why does the three layers go up by 50 Gb? Well, it was a slightly different format. Like anything, there are always improvements as these later revisions come out. So in fact, XL3 does go all the way up to 100. By the time you get to XL4, you are having to penetrate so far down into the disc, that you cannot get another 50. Up to 128 Gb with the XL4. So again, far more storage than you can have on a dvd.
Solid State Drives
In this section, we will talk about solid-state drives. This is a persistent memory storage type that essentially uses the same kind of memory as what we think of as the ram in our computers. There is one significant difference and it is that word persistent.
The ram in your computer to this day still uses what is known as dynamic ram or volatile ram, which means it needs power to retain any data. When the power is lost, whatever is in memory is lost. Persistent memory, quite simply, does not need power. It is sometimes also referred to as non-volatile. This is in and of itself the same type of memory, it is just a different interface. A flash drive, of course, would just use USB. Solid-state drives can use the same interfaces as traditional hard drives do, but because it is persistent, we just do not need power to keep the data on the storage device.
So the data is stored on an integrated circuit, not a magnetic spinning disk. So essentially there are no mechanical components. There is no spinning disk. There is no read-write arm that is moving back and forth. So this ultimately offers much faster access compared to traditional hard drives and in some cases better resistance to accidents. For example, if you were to drop a laptop, the laptop itself might survive. The shock might damage the hard drive because of that read write arm or the spinning disk. Since they are not present in solid-state drives, then it can in some cases be more durable as well.
Ultimately, it is the performance that we tend to favor when it comes to solid-state drives. So in every aspect of getting to the data or reading and writing the data, booting the computer, launching applications, and transferring files, really everything is much faster. We simply do not have to locate the appropriate place on the disk to get to that data. It is just written on a chip somewhere. So in almost every instance, data access is much faster compared to traditional hard drives.
Now, looking at some other key characteristics, they also tend to require less defragmentation. The spinning disk of a traditional hard drive has its fastest access around the outside edge. So most data prefers to be written as close to that outside edge as possible. Ultimately, it has to go somewhere wherever there is free space. So you might end up with some of the file having been originally written close to that edge. Then subsequent additions to the file, for example, may just not have had any place to go. So they end up being further into the disks. So they end up being fragmented. Really, the location, the address of the data does not matter with solid-state drives. So you end up with less fragmentation.
There is very little operating noise; virtually silent because there are no moving parts. They tend to be smaller than traditional hard drives as well and they also require less power to operate.
There are still some limitations to solid-state drives. They are still a bit pricey compared to traditional hard drives. The cost per gigabyte would still be lower than that of a solid-state drive. Most commonly used solid-state drives have capacities under 1 Tb. So they tend to be smaller. Now you can certainly find larger ones but they would be more expensive. Generally, it is the performance that we are concerned with, not so much the capacity. So they do still tend to be favored in many cases.
They provide little warning of impending failure. A lot of the traditional hard drives had monitoring capabilities. Where if it noticed that it was having difficulty writing to a particular area of the disk, it would actually tell you. That does not happen with solid-state drives. Ultimately, they can have a shorter life cycle than Hard Disk Drives. It is a little harder on the device, when it comes to the wear and tear of the read-write process. Every time it writes new data, it has to erase whatever was there. Then it has to write the new information. Traditional hard disks did not have to do that. It would just overwrite what was there. So it may not be suitable for long-term archiving.
Now that is a bit of a relative term. Let us say, for example, if you have a laptop and you get 5 years out of the laptop, then your solid-state drive should certainly last that long. If you are talking about needing to store something for 20 years then chances are a traditional magnetic hard drive would have a longer lifespan than a solid-state drive. So there are still particular scenarios where a traditional hard drive would be favorable. However, if it is performance you want, then definitely go with a solid-state drive.
Magnetic Hard Drives
In this section we will look at magnetic hard disk drives, which is an electromechanical data storage device.
Now, really that is just a fancy term for something that has moving parts. It uses rapidly rotating magnetic platters. You cannot really tell as it is sort of a top down view, but in almost every hard drive there is more than just one platter. There are usually at least two, perhaps maybe as many as four. Plus, each platter has a top surface and a bottom surface. So for every platter you have two storage surfaces. Then it also has this actuator arm that moves back and forth and applies the electrical charge to the magnetic platter. This is how you read and write data.
Now, that data can be randomly accessed. So, in other words, you can jump around from one place to another. The arm has to move back and forth to the inside versus the outside. The disk of course, has to spin to the correct location to access that data. So, it typically does not perform as well, particularly compared to something like solid-state drives. It is a non-volatile storage mechanism, meaning that it does not require power to retain its data. Of course, that has been the case really since hard drives came out. Really that is their purpose, to be able to store data while the system is turned off.
As for some of the key characteristics, it is a very mature technology. Hard drives have been around for decades and with respect to just storing the data, they are very reliable. Now, they are susceptible to mechanical failure. Once the data is written to a magnetic hard disk, it is usually there for 10 to 15 years very reliably. You could certainly still get up to 20 or even 25 years. So reliable in that regard, in that it just does not degrade very rapidly. They are still very cost effective. Again, compared to something like a solid-state drive, magnetic disks are much cheaper. They generally have a much higher capacity as well.
Now these days, the most common interface is likely a serial ata, at least for an internal hard drive. I would not be surprised to still find some Ide. Of course, if it is a server system, you certainly might find a scsi interface too. Most desktops would probably have a serial ata interface.
Now, as far as performance, you cannot really state definitively that a magnetic hard drive will have X level of performance. It will vary greatly, depending on the platter rotation speed, the internal cache, and the interface type.
Now, as far as the rotation, the most common values there are 5400 versus 7200 rpm. Clearly, when you are trying to access something on a spinning disk, the faster it spins the faster you can locate it. 5400 was common in some older laptops but not seen much anymore. So I would say that most traditional magnetic hard drives would probably be 7200 rpm.
The internal cache is really up to the manufacturer. It is up to the make and model if you will. The idea here is that it uses an actual little bit of memory to just act as a temporary storage location while data is being written or read. That cache is standard memory. It is not a spinning disk. So it certainly helps to improve the overall performance.
As mentioned, it would depend on all of those to determine what the performance is.
In virtually every case, a magnetic hard disk drive will underperform a solid-state drive. It is just a completely different type of technology. There are no moving parts. There is no spinning disk in a solid-state drive. So even the best performing magnetic hard drive would not get you the performance of a solid-state drive.
They do have higher capacity. So in most cases, they are still preferred when you need to store a very large amount of data. Really it just comes down to the cost at this point. To get the same capacity in a solid-state drive would simply be much more expensive than a traditional magnetic hard drive.
2.5 Inch Versus 3.5 Inch Drives
In this section, I will compare and contrast some of the features of the two most common physical sizes for hard drives.
Now, beginning with the larger, 3.5 inch, this is more commonly used in desktop computers. They are typically very easy to install. Once you get the cover off a desktop, the drive bay is usually very accessible. It is just a matter of affixing a couple of screws, usually two per side, and then attaching the data cable and the power cable. Really, that is it. They tend to be very cost effective. Again, hard drives have been around for a very long time, so the price has done nothing but come down.
In general, and I stress that generalization, you will typically find a higher capacity compared to the 2.5 inch drive. So 2-4 Tb might be very common. For example, in a 3.5 inch drive, you can get much larger. You can get smaller, but that would just be a common value. Again, in general, they might have a higher data transfer rate compared to the 2.5 inch drive. You would likely see a faster platter rotation speed and more cache, simply because there is more physical space inside the unit itself.
Now, if we contrast that with the 2.5 inch drive, this is visibly smaller. Of course, it is much more commonly used in laptops, where space is the main concern. Now that said, you can use the smaller drive in a desktop, but it can be a little challenging to install because you would need some type of adapting bracket for it to fit in the larger drive bay. They do exist and they do actually make less noise. Everything is a little more tightly packed in the 2.5 inch drive. As such, it is actually a bit quieter. This can make a difference depending on where you work.
If, for example, you were in a recording studio, noise might make a difference. So, you might want the smaller drive. Again, in general, compared to the larger drives, they tend to have a smaller capacity. Under 2 Tb might be very common for the 2.5 inch drive. You would likely find a lower data transfer rate as well, due to a slower platter rotation speed and less cache.
Now again, I want to stress that those are just very general statements. There would certainly be some overlap. If you imagine these two as a venn diagram, you could get a 2.5 inch drive that would outperform the 3.5 inch drive. Generally, you will get better performance from the larger drive.
Hybrid Drives
In this section, we will talk about a hybrid drive, which is a combination of a solid-state and a traditional magnetic hard drive. These came about really when solid-state drives themselves were fairly new. As such, they were quite expensive.
So what this gives you is a sort of happy medium, a transition if you will, that does offer better speed because of the solid-state component, but the higher capacity of a traditional hard drive, at a middle group in terms of cost. Now on that note, the cost of a solid-state drive themselves has also dropped quite significantly in recent years. So I would say it is probably not all that common to find a hybrid drive in today's marketplace, but it is certainly not out of the question.
Essentially, in terms of how it worked, it had both the solid-state drive and the traditional hard drive within a single device. Typically, you had a smaller section that was the solid-state drive where the operating system and the most frequently accessed data would be stored. Then in the traditional hard drive, you would find more infrequently accessed data like large media files or large amounts of documents.
Now with respect to which files went where, there is a software caching algorithm, which can quite literally keep track of the most used resources. This caching algorithm resides right in the hybrid drive's firmware. In other words, the operating system does not see two separate sections of the drive. It just sees a drive. This algorithm automatically allocates to the solid-state section or the hard drive section, depending on how often you use it.
Now the solid-state drive aspect itself was nonvolatile. So again, you did not need power to store data. It effectively acted as a cache which, in fact, is not new. Traditional magnetic hard drives have had a cache for quite some time. This is just a lot more of it. There really was not this advanced algorithm if you will, that tried to keep track of which files were accessed most commonly. It just happened to be whatever was being worked on at any given point in time.
So ultimately, with this algorithm, the speed will improve over time. In other words, the very first time you use it, the drive really does not know what is going to be the most commonly used data. So it would act as a traditional hard drive at that point in time. As it learns, then it starts to get faster. So ultimately, you do end up with a drive that is faster than a traditional hard drive, but still slower than a solid-state drive. Again, you get that happy medium at a middle ground cost.
Flash Drives
Now in this section, we will take a look at flash drives, which use a type of memory known as electrically erasable programmable read only memory. Which is a bit of a fancy term for just saying that you can write to it, and then you can erase that, and rewrite it, and then read many times from it. It also has the characteristic of being nonvolatile which means that we do not need to supply constant power to the device for it to retain its data. I am sure many of you are familiar with the USB flash drive. Of course, it does not need to be continually plugged into the computer for it to retain its data.
Now, because it is just a collection of memory chips, it offers very fast access times. It is relatively shock resistant and durable. If you just drop a USB flash drive, chances are the data is going to be fine. If you drop a hard drive, there is a much better chance of losing that data due to damage. Of course, they are very cheap by today's standards. So, even if you did damage to it, it can be replaced quite inexpensively.
So some of the key uses of flash memory include short term data storage. Now, that is a bit of a relative term and it is not to suggest that you cannot store data for a long time. It is just that they are also very portable. Now, that is nice in many cases, but because of that, it is very easy to lose. They are usually very small, compact devices. So they are easy to lose- they get left behind. So just in that regard, it might not be the best idea to implement backups or long term archive storage using flash memory. It can certainly store data for a long time.
In some systems, it can be used for ram enhancement. You can quite literally insert the flash drive and tell the operating system to use that as ram for a system that maybe does not have enough. In some cases, you can also implement multi-boot configurations. If your system supports booting from USB, you can install an operating system on the flash drive and boot from that. So you can take it out and boot to what we will call your default operating system. Then you can shut it down, insert it, reboot, and boot to the operating system on the flash drive.
So some common types of devices include the USB flash memory or memory stick or thumb drive, you might also hear. Of course, these are very common. Many people use these again, for just being able to carry data around very easily. Compact flash is generally implemented as a card. You will often see these in cameras or printers. This also goes for the Secure digital or SD card. In fact, in many cases, the SD card has superseded the compact Flash card. SD cards also come in two different sizes: MiniSd and MicroSD. For smaller and smaller devices such as small digital cameras, and cellphones, and maybe a GPS. This again, has been largely replaced by the SD cards.
Again, the idea is that they are all very small. They are easy to carry around and they can be rewritten at any time. So you can just transfer the data onto a hard drive, for example, and then just reuse that device as many times as you want. They tend to have a limited write capability, but usually they will last several years before you start to encounter any errors. Whatever is still on the device can probably still be read from. You might just encounter a situation where it will no longer write to the device, but again, in most cases, you will get several years out of the service, as long as it is cared for properly.
How Swappable Drive Media
In this section we will talk about hot swappable drives, which refers to the fact that you can quite literally replace a physical hard drive without powering the system down.
Now, what you have here is a hot swappable bay. It is the bay that is permanently attached to the computer. It is mounted on the computer . The power and the data connections attach to the back of the bay. Then you quite literally just slide the hard disk into the bay and it makes the connections at the back of the bay. As soon as it does, it is usable again without having to power down.
Now, hot swapability itself is really not all that foreign a concept. Virtually every USB device is hot swappable. You do not have to run off your computer to plug in a USB device or to remove it. We tend not to think of that as something that is very common for internal drives, but that is exactly what this is.
Now, it is far more common in server systems than it is in desktops because quite simply, servers tend to have a lot of hard drives. As such, they can maybe fill up or just fail a little more often than a system with one or two hard drives. Because the more you have, the more likely at least one of them is to fail. So it makes it very easy to swap out a failed drive without having to take that server down and interrupt all of its other functionality.
Now, it is supported for both solid-state and traditional hard disk drives, but it is dependent on the interface. So, for example, scsi, serial attached scsi, and serial ata, all do support hot swap ability, but the ide interface does not. So it is something that you need to be mindful of. Again, servers tend to have more scsi interfaces than a desktop computer. You can still do it on a desktop computer, if you have an interface that supports it.
Some scenarios might be, if you have a completely different operating system or different applications that you want to run in the same physical computer. You can just swap out one hard drive, put in the other one, and you have got that different operating system. Now, you would reboot in that case anyway, but it is still very simple and very easy to access the drive. You quite literally just open the door, slide it out, put in a different one, and reboot. So again, it is much more common in servers, but it is supported on desktops.
So, looking at some features of hot swappable drives, this is an ideal scenario for storage and archiving data. If you have a lot of data that just needs to be archived for a very long time, then you can essentially just fill up one drive, swap it out, fill up another one, and keep going. So those full drives can simply be swapped out once they do get full. Again, you do not have to interrupt anything.
If it is a matter of a failed drive, then any kind of faulty drive can also be swapped out, again, without shutting anything down. So this really helps provide high availability and high reliability because that failed drive can be replaced on the fly. So the server stays up and running and all of the services that it provides remain available for the users to access.
Types of Raid
In this section, we will take a look at raid, which stands for a redundant array of independent disks. This is a hard drive configuration that allows you to have multiple physical drives, operating as a single logical unit.
Now, if you imagine a system that has three physical drives. In a normal configuration, they might be your D,E, and F drives. With raid, those three physical disks can all operate as a single logical drive letter. So, as far as the operating system is concerned, it is just a D drive, even though there are multiple physical drives.
Now, there are various raid levels that we will take a look at shortly, that allow you to configure both redundancy and even non-redundant configurations, despite the fact that the R stands for redundant. This is meant to provide for reliable data storage for data that is in use. In other words, if this data needs to be accessed, and let us just say it is fairly important data, then you want to ensure that it is always accessible.
Now, you can protect the data of course by just using backups. This is meant to be able to continue to service that data, even though one of your physical disks may have failed. If you have a system that has multiple hard drives, you are more likely to lose one of those hard drives just due to normal failures. So let us just say, for argument's sake, you had ten hard drives. Well, you are ten times more likely to lose one of those drives. With raid configurations, you can simply discard that failed drive, replace it, and all of the data that was there is still intact, without having to restore from a backup.
Ok now, we will talk about that in greater detail in a moment as well. This is a fairly low cost configuration that you can implement on just about any system. It is certainly much more common on systems with multiple hard drives in the first place, such as servers. This is also much more commonly used with hot swappable configuration. So that if one of those drives fail, we can literally just slide it out, put in a new one, and all of the data is actually still there.
Now, there are also two main categories of raid known as software based and hardware based. Software based means that you are just using the hard drives that are physically there in the computer, along with the inherent management interface of the operating system itself. Most operating systems do have some kind of disk management utility where you can partition and format your hard drives. So if you do have multiple physical drives, they can be configured into a raid array, just using that inherent administrative interface. A hardware based raid means that you go out and actually purchase a physical controller card and all of the hard drives connect to that card. Then there is usually software that comes with the card that you install separately. That is how you manage the devices.
The hardware based raid is preferable, because in almost every case, there is a lot of cache memory on the card that can help to speed up access to the drives. They usually support more physical drives than what you might have inherently in the computer. Software based is, of course, a little bit cheaper. You are not purchasing anything extra. Hardware based is preferable and usually offers a little more functionality than software based.
So then looking at the levels, we will talk about the most commonly supported levels of 0, 1, 5, and a relatively new one of 1 plus 0. Now, there are other levels, but these ones are widely supported by most operating systems. So raid is known as striping. The 0, in this case essentially indicates that there is actually no redundancy. So it is still that R for redundancy, but the 0 in fact says no, there is no redundancy. So what you have here is a file that is being written across multiple physical disks.
So imagine all of these blocks,1 through 8, as just representing a single file. It just takes that file and breaks it up into chunks. Then a part of it is written to Disk 1. Then the next part is written to Disk 2. Then back to Disk 1. Then back to Disk 2. It just keeps bouncing back and forth, until the entire file has been written.
So what this does for you, is it allows you to write to both disks simultaneously and to read from both disks simultaneously. Now, since there is no redundancy, this actually does not protect you from a single disk failure. If Disk 2 were to fail, then that entire file becomes unreadable. So you would still certainly still need backups in this case. So that begs the question, why would you even do this? The answer is performance. Because you can write to and read from both disks simultaneously, you get very good performance with striping.
Now, raid 1 is referred to as a mirror. As its name indicates, it completely duplicates the files on one physical disk over to another one. So you can see in this case, blocks 1,2,3, and 4. Again, imagine that represents just a single file that is being written entirely to Disk 1, but then it is exactly duplicated onto Disk 2. So this definitely provides redundancy. If either disk were to fail, the data is perfectly safe and sound on the other disk. So you can replace the failed disk and you can regenerate the mirror. You still have all of your original data.
Now, raid 5 is what is known as striping with parity. Parity, in this case, means that there is some kind of error correction. So it is still striping, where we see any given file is chunked up into these blocks. So block a1, a2, and a3, again, might represent a file. That gets you good performance, but then the parity information allows you to actually recover data if it is lost.
Now, this is somewhat simplistic, but parity works by using a calculation if you will. Let us just actually use the numbers. So for block a1, a2, a3. If you were to add up the numbers 1 plus 2, plus 3, that is 6. So you can think of the parity information for parity A, as being equal to 6. With that calculation known, if I were to say, lose drive 2, then block A2 would be lost. I would be left with 1, plus an u known, plus 3, equals 6. Well, I know it equals 6. I know that I still have a 1 and a 3. Therefore the missing value has to be 2.
So you can quite literally regenerate all of the data on the entire device. Because as you can see, the parity information itself, is also spread across every drive, just like the files are. So it does not matter which drive fails. Any one of them can be regenerated from the surviving disks and the parity information. So you can quite literally just replace the failed drive with a new one and you can regenerate the data. Again, you did not need to go to your backups.
Now finally, 1 plus 0 sometimes referred to as raid 10, is a stripe of mirrors. So what we see here, are two individual mirror sets that are then striped. So recall that striping in of itself does not give you redundancy, but it does give you very good performance. So by adding in the mirror sets, then we have redundancy.
So, again we see block a1 is mirrored to its partner on disk 2. There is redundancy but then the rest of the file, block a2, is over on the other set. That is also mirrored. So now we have incorporated both: the benefits of striping and its good performance, along with the protection of mirroring. Again, that one is relatively new. It is still supported on just about any system. Typically again, you would probably only find this on server systems because it starts to get a little more expensive because we have a lot more disks. The more you have in terms of the disks, the more it is going to cost of course. It is still something that you can implement on a desktop system, if you feel it is appropriate, but much more commonly in the server world.