Digital storage basics, Part 1: Internal storage vs. memory

CNET editor Dong Ngo goes over the basics of digital storage devices for home users.

An example of digital storage diversity: Two system memory sticks (for desktop and laptop), an SSD, a laptop hard drive, and a desktop hard drive.
An example of digital storage diversity: Two system memory sticks (for desktop and laptop), an SSD, a laptop hard drive, and a desktop hard drive. Dong Ngo/CNET

Editors' note: This post is part of an ongoing series. For the other parts, check out the related stories.

It's not the locker room type of storage we're talking about here. Instead it's something much more important and often underrated: the place where information is stored.

When it comes to computer storage, judging from many questions friends and readers send me, there's quite a bit of confusion among general users as to what it actually is. And it's not your fault; digital storage can be as messy as my desk. This is the reason for this series, where I sort out the basics and more, in layman's terms.

That said, some information in this might be too basic for advanced users. Home and novice users, however, give yourself some uninterrupted time and dive in. You'll survive.


1. Understanding the units
No matter how boring this is, you can't grasp digital storage without know its measurement unit, which is byte.

Byte (symbol: B): Byte is generally the smallest unit in digital storage. You can think of 1 byte as one character in a document. For example, we actually need to use 4 bytes to store just the word "byte." In real life, we use larger units, including kilobyte, megabyte, gigabyte, and terabyte.

Note: Technically, there's another smaller unit called bit (symbol: b), which is a single binary unit that represents the state 0 or 1, which encodes digital information. A byte is a sequence of bits, and generally 1 byte equals 8 bits. Bit is more commonly used to show the data being transferred, especially over a long distance, such as the speed of the Internet, which is measured in bits per second. Byte is more commonly used to show the amount of storage or in situations you can move a large amount of data. When it comes to storage space, it's better to use byte, much like it's more practical to count the number of cows than counting the number of feet and then divide by four.

Kilobyte (KB or kB): By general definition, one kilobyte is 1,024 bytes. In many cases, for the sake of simplicity, 1 kilobyte is understood as 1,000 bytes.

Megabyte (MB): By general definition, 1 megabyte is 1,024,000 bytes. Similarly, it can also be understood as 1,000,000 bytes.

Gigabyte (GB): By general definition, 1 gigabyte is 1,000,000,000 bytes.

Note: There's another unit called a gibibyte (GiB), with 1 GiB equaling 1,073,741,824 bytes. The JEDEC memory standard also defines 1 gigabyte as 1,073,741,824 bytes, which happens to be the definition that Microsoft uses and hence is used by the Windows operating system to report storage device capacity. This causes confusion since all storage devices now appear to offer less storage space than their advertized capacity. For example, a 500GB drive, once formatted by Windows, will report a capacity of only around 465GB. This is just a matter of interpretation.

Terabyte (TB): By general definition, 1 terabyte is 1,000,000,000,000 bytes, or 1,000GB.

Currently, the largest 3.5-inch hard drive (commonly found inside a desktop computer) offers 4TB of storage space. Most computers come with drives with capacities of somewhere between 120GB and 2TB. Most mobile devices, such as tablets or smartphones, offer between 8GB and 120GB of storage space.

Note: Generally, a typical photo taken by the iPhone 4 takes up about 2MB of storage space. A digital song uses about 5MB. A compact disc (CD), which has the capacity of 700MB, can hold about 350 iPhone photos or some 140 songs. The actual size of digital content varies a great deal, however, depending on the format and the compression level. The common rule is the richer (and/or higher quality) the content, the larger storage space it requires. A 10-minute audio podcast needs anywhere between 4MB and 10MB, but a 10-minute high-def movie requires a few hundred megabytes or even a gigabyte of storage space.

A 2.5-inch hard drive next to a 3.5-inch hard drive.
A 2.5-inch hard drive next to a 3.5-inch hard drive. Dong Ngo/CNET

2. Storage vs. memory
These are two terms that are often mistakenly used for each another, though they are two very different things.

Storage, in a nutshell, is where the information (such as Word documents, photos, movie clips, programs, and so on) is stored. In a computer, the whole operating system itself, such as Windows 7 or Mac OS, is also stored on the internal storage device. Storage is nonvolatile, meaning that the information is still there when the host device (a computer, for example) is turned off and is readily accessible when the device is turned back on. It's like a book or a paper notebook that's always there, ready for you to read or write on.

Memory (aka system memory, random access memory, or RAM), on the other hand, is where information is being processed and manipulated. Data in the system memory is volatile, meaning that when the computer is turned off, it's gone; the memory becomes blank, as if nothing has been there before. It's somewhat like the short-term memory part of your brain, where images or ideas are being formed and processed when you read a book -- those that disappear the moment you stop reading.

When you turn on the computer, most of the boot time is when the operating system is being loaded from the computer's main storage unit -- likely a hard drive -- to the system memory. The computer is fully loaded and ready to do other tasks when this process is done.

Despite their differences, there's a strong relationship between system memory and storage. The Word document that you're working on, for example, is in the computer's memory. When you save it, a copy of it now resides on the computer's storage. When you close Microsoft Word completely, the document now only resides on the hard drive (storage) and is no longer in the memory, until you open it again.

System memory is a lot more expensive than hard-drive storage, gigabyte-to-gigabyte.
System memory is a lot more expensive than hard drive storage, gigabyte to gigabyte. Dong Ngo/CNET

All this means is that you generally don't actually experience storage. Everything that's presented to you on a computer's screen or via the speakers actually takes place in the system memory. Before it gets there, however, it needs to be loaded from the computer's storage device into the system memory. So the larger and faster system memory the computer is equipped with, the more quickly the information becomes ready and the more you can do with a computer at one time (multitasking). You generally need far less memory than storage. Most new computers come with somewhere between 2GB to 8GB of memory, and you don't need more than that. This is a good thing, too; gigabyte to gigabyte, memory is much more expensive than storage.

Of course, memory is just one of many factors in a computer's performance. Another factor is the storage itself, which is either a hard drive (aka hard disk) or a solid-state drive (SSD).

3. Hard drive vs. solid-state drive
The hard drive has been the most common storage device for decades, dominating since the early 1960s. Solid-state drives, however, are relatively new and have been getting more and more popular in the last three years. In most case, they can be used interchangeably, and both have pros and cons.

A standard SSD (left) next to a standard laptop hard drive.
A standard SSD (left) next to a standard laptop hard drive. Dong Ngo/CNET

A. Hard drive (or HDD): While the hard drive has evolved a lot since its inception, the basics remain the same: it's a box that contains a few magnetic disks (known as platters) attached to a spindle, very similar to a spindle of blank CDs or DVDs. Each of the platters has a reading/writing head hovering on top. As the spindle spins, the head moves in and out to write or read data to and from any part of the platter, on a tiny information-recording unit called the "data track." This type of access to information is called "random access," as opposed to the inefficient "sequential access" found in the old and obsolete types of storage, such as tape.

While the concept is rather simple, the inside of a modern hard drive is a world of advanced nanotechnology. This is because as hard drives' storage capacities increase while their physical sizes remain the same, the density of information written on the platters becomes so great that we need to use nanometers to measure it. One nanometer is 1 billionth of a meter (a meter is about 3.3 feet).

Perspective: Inside a regular 2.5-inch laptop hard drive, the WD Scorpio Blue , for example, the gap between the recording head and the platter is just a few nanometers. The two can never touch each other -- or else the drive will be "bricked" -- and when the hard drive is at work, its platters spin at 5,400rpm. (Desktop and high-end laptop hard drives spin even faster at 7,200rpm or 10,000rpm.) To put this in context, if we enlarged the Scorpio Blue by 13,000 times, the platter would look like a circular race track about 3.3 miles in diameter; a data track would be about 0.4 inch in length, and the recording head would be about the size of a go-kart. When the hard drive is in operation, this go-kart would be flying on the track less than the thickness of a human hair above it, at a speed of about 3.4 million miles per hour.

Hard drives generally come in two physical designs: 3.5-inch (for desktops) and 2.5-inch (for laptops). The laptop hard drives can also come in different thicknesses, such as 9.5mm (standard), or 7mm (ultrathin). A hard drive is connected to a host using a connecting interface standard.

Connection interface: This is the standard that determines how a hard drive (or a standard SSD) is connected to a host (such as a computer) and how fast the data rate is between the storage device and the host. There have been a handful of interface standards for storage. Currently, most if not all consumer-grade drives use the serial ATA (or SATA) standard. This standard is available in three generations: SATA I, SATA II, and SATA III, which offer a speed cap of 1.5Gbps, 3Gbps, and 6Gbps, respectively. The latest generation of the SATA standard is backward compatible with the previous generations, in terms of usability. In terms of performance, you'll need use those of the same SATA generation for optimal speed.

Pros of hard drives: Generally, hard drives offer the largest amount of storage per unit (currently up to 4TB for the 3.5-inch design, or 2TB for the 2.5-inch design). They are also very affordable, costing just a few cents per gigabyte. For this reason, hard drives are still the most popular form of computer storage and are used in most storage applications.

Cons of hard drives: Since these are mechanical devices, hard drives suffer from wear and tear, just like any other machine with moving parts. They also use significantly more energy (compared with SSDs), generate heat, and are much slower. Hard drives also require some time to spin from being idle or turned off, which makes the host computer take longer to boot. Generally, a typical hard drive, in common use, lasts about five years.

Inside an SSD (left) and a laptop hard drive. By opening them up, I have basically destroyed the hard drive. The SSD, which has no moving parts, however, still works fine, even without its cover.
Inside an SSD (left) and a laptop hard drive. By opening them up, I have basically destroyed the hard drive. The SSD, which has no moving parts, however, still works fine, even without its cover. Dong Ngo

B. Solid-state drive (SSD): Unlike a hard drive, an SSD has no moving parts. Similar to system memory, SSDs are microchips designed to store information. However, these are nonvolatile memory chips that can retain information the way hard drives do. Most standard SSDs come in the 2.5-inch design, and on the outside, they look just like a regular 2.5-inch hard drive. Standard SSDs work in any cases in which hard drives of the same connection interface are used. Since there are no moving parts, SSDs can be made in many different (and sometimes proprietary) physical shapes and sizes, making them the best choice for mobile devices, such as smartphones or tablets. Generally the lifespan of an SSD depends on how much information is being written on it (the less, the better) and how large its capacity is (the larger, the better).

Pros of SSDs: Much faster than regular hard drives, much more energy-efficient, more durable, much cooler, and quieter. Upgrading a computer from using a hard drive to an SSD as its main storage offers the single biggest incentive in terms of performance. Most SSDs last much longer than five years; some could even last hundreds of years.

Cons of SSDs: The biggest catch with SSDs is the price. Currently SSDs are between 7 and 50 times more expensive than hard drives in terms of cost per gigabyte, depending on the capacity. SSDs also have limited capacities, offering just about 512GB or less before getting too expensive to be practical. SSDs also suffer from a finite time of writing, called "write endurance." In other words, an SSD can be written on a limited number of times before it becomes unreliable. Before you can rewrite on a part of the drive, you'll need to first erase the information already stored on that part. This is why the write endurance rating is also known as the program/erase (PE) cycles. In reality, this is not a big deal because for most situations, an SSD would likely be replaced for other reasons way before its PE cycles end.

The best use for SSDs is as the main storage unit of a computer that hosts the operating system; it will improve the computer's overall performance a great deal, compared with a hard drive. In desktops, you can also use an SSD as the main drive and another regular hard drive as a secondary drive to house data. This hybrid solution is actually the best practice that balances performance, cost, and storage space. Or you can also opt for a hybrid drive.

The Seagate Momentus XT hybrid hard drive offers either 500GB or 750GB of storage space and costs far less than any SSD of the same capacity.
The Seagate Momentus XT hybrid hard drive offers either 500GB or 750GB of storage space and costs far less than any SSD of the same capacity. Dong Ngo/CNET

C. Hybrid drive: As the name suggests, a hybrid drive is one that uses both regular platter-based storage and solid-state-based storage in one box. Hybrid drives come with a built-in algorithm that automatically moves the frequently accessed files, such as the those of the operating system, to the solid-state part, and leaves the more static data, such as photos or movies, on the hard-drive part. This offers SSD-like performance without the high price tag and the limited storage space. Currently, the most popular hybrid drive is the Seagate Momentus XT. Apple also recently announced its own hybrid storage drive called the Fusion Drive .

In real-life testing, hybrid drives indeed help boost a computer's performance, compared with hard drives, but they are in no way as fast as SSDs.


That's it for now. If you still have questions, put them in the comments section or send it to me via Twitter or my Facebook page. Check back again for Part 2 , where I'll talk about external storage devices.

Read the full CNET Review

WD VelociRaptor WD1000DHTZ (1TB, SATA-600)

The Bottom Line: WD VelociRaptor WD1000DHTZ makes an excellent storage solution for housing a large amount of data that's frequently accessed and manipulated. / Read full review

Read the full CNET Review

Seagate Momentus XT (second generation, 750GB)

The Bottom Line: Similar to the original, the Momentus XT (second generation) makes an excellent host drive for any computer, both laptops and desktops. It offers great value in terms of performance and cost, especially when compared to SDDs. / Read full review

About the author

CNET editor Dong Ngo has been involved with technology since 2000, starting with testing gadgets and writing code for CNET Labs' benchmarks. He now manages CNET San Francisco Labs, reviews networking and storage products, and also writes about other topics from online security to new gadgets and how technology impacts the life of people around the world.

 

ARTICLE DISCUSSION

Conversation powered by Livefyre

Don't Miss
Hot Products
Trending on CNET

Hot on CNET

CNET's giving away a 3D printer

Enter for a chance to win* the Makerbot Replicator 3D Printer and all the supplies you need to get started.