Digital storage basics, Part 2: External drive vs. NAS server

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

Play

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

Now that you have some ideas about internal storage vs. memory after reading Part 1 , here in Part 2 I'll talk about the two main types of external storage devices: direct-attached storage (DAS) and network-attached storage (NAS).

DAS and NAS share a few things in common. They both use one or more internal drives (hard drives or SSDs) on the inside, and they utilize the same methods to combine drives' storage space (when multiple drives are used). But before we go any further, let's find out some common terms.

Since external drives such as the Seagate Backup Plus and the Pegasus R6 are representative of all consumer-grade DAS devices, for the sake of simplicity, DAS devices will be called external drives in this post.

Note that, as with Part 1, the information in this post is intended for novice/home users.


Understanding the common terms

Here are some common terms used when talking about external drives and NAS that you'll likely run into when reading or shopping for storage devices.

Single-volume (single-drive)
This means that the storage device comes with just one internal drive inside. A single-volume external storage device generally has the maximum capacity of its single internal drive, which for now is 4TB or 2TB, depending on the type of internal drive being used, 3.5-inch or 2.5-inch, respectively.

Multiple-volume (multiple-bay or multiple-drive)
This is the type of external storage device that uses more than one internal drive as its storage. These devices range from using two drives (dual-volume), such as the WD VelociRaptor Duo, to using many, such as the six-bay Pegasus R6 from Promise.

Hosting more than one internal drive not only increases the total storage space but also gives you the option to use RAID.

RAID
RAID stands for redundant array of independent disks and requires at least two internal drives. RAID allows for combining multiple drives' storage space into a single volume. Depending on the type of array, a RAID configuration can offer faster speeds, more storage space, or both. Generally, you'll need to use drives of the same capacity in a RAID array. Below are the most common RAID arrays.

RAID 1: Also called mirroring, RAID 1 requires at least two internal drive drives. In this setup, data is written identically to both drives simultaneously, resulting in a mirrored set. A RAID 1 setup continues to operate safely even if only one drive is functioning (thus allowing you to replace a failed drive on the fly). The drawback of RAID 1 is that no matter how many drives you use, you get the capacity of only one. RAID 1 also suffers from slower writing speeds and is mostly used in dual-drive storage devices.

RAID 0: Like RAID 1, RAID 0 requires at least two internal drives. Unlike RAID 1, however, RAID 0 combines the capacity of each drive into a single volume to deliver maximum capacity and bandwidth. The only catch is that if one drive dies, you lose information on all devices. So while more drives in a RAID 0 setup means higher bandwidth and capacity, it also means a greater risk of data loss. RAID 0 is also used mostly for dual-drive storage solutions. Should you chose RAID 0, backup is a must.

RAID 10: This array is used mostly in storage devices that have four internal drives. It's basically a combination of RAID 1 and RAID 0, offering benefits for both performance and data safety.

RAID 5: This array requires at least three internal drives, but it distributes data on all drives while keeping the array safe against the failure of one. Though a single drive failure won't result in the loss of any data, performance will suffer until you replace the broken device. Still, because it balances storage space (you lose the capacity of only one drive in the RAID), performance, and data safety, RAID 5 is the preferred setup.

RAID 6: This array is similar to RAID 5 but now the array can survive the case that two of its internal drive fails at the same time. RAID 6 is generally used in storage devices that have 5 internal drives or more. In a RAID 6, you lose the capacity of two internal drives.

If the above seems complicated, don't feel bad; RAID is quite complicated. The good news is that all RAID-capable storage devices come with the RAID setup preconfigured; most of the time, you don't need to configure it yourself.

Hot-swap: In storage, hot-swap is a feature only available in devices that use a redundancy RAID configuration (i.e. not applicable to RAID 0) and means that you can change one of the internal drives of the array without having to power down the storage device. This is a great feature in case you need to replace a failed drive while still making the storage device available to the host or the network.

In addition to RAID, multiple-drive storage devices usually also offer the JBOD option.

JBOD
This is the acronym for just a bunch of disks; in a JBOD setup, each drive works as an individual volume to the host. There's no relationship between them. In other words, if you plug a dual-drive storage device in JBOD to a computer, it will appear as though you plugged in two separate portable drives of the same capacity at the same time.


In addition to the things external drives and NAS devices in common, they also have a few distinctive differences, in terms of connectivity, performance, features, and diversity. These differences give them their own pros and cons.

Two multiple-bay external drives next to three portable drives.
Two multiple-bay external drive next to three portable drives. Dong Ngo/CNET

A. External drives

The reason external drives are called direct-attached storage (DAS) devices is because they connect to a host -- be it a computer, a server, or a media player -- directly and become a storage extension of the host. There are two main types of external drives: portable and desktop.

Portable drive (aka portable external drive, or laptop external drive)
These are external drives that are based on the 2.5-inch (laptop) internal drive, either a hard drive or a solid-state drive (SSD). They are generally compact, lightweight, and easy to carry. Portable drives are almost always single-volume drives and most, if not all, of them have the capability for bus power. A bus-powered drive uses its data cable to draw power from the host, sparing you from having to use a separate power adapter, making the drive truly portable. Portable drives are designed for people who need to carry data with them, such as laptop users, or to easily move data from one computer to another.

The Elgato Thunderbolt SSD is the first Thunderbolt bus-powered portable drive; most if not all USB portable drives are bus-powered.

There's another type of portable storage called a thumbdrive, though some also call them a jump drive, USB drive, USB key, or just simply USB. A thumbdrive is basically a small flash memory chip (similar to those of SSDs) connected to a USB connector head that you can plug directly into a USB port of a host, such as a computer. You can think of a thumbdrive as a portable drive that's based on a nonstandard SSD. Thumbdrives are very popular thanks to their key-size design and affordable pricing. These tiny devices offer up to 128GB of storage space, such as the Lexar Echo MX. Most of the time, they come in a variety of capacities that range from 500MB to 16GB.

Desktop external drive
A desktop external drive uses the 3.5-inch (desktop) internal hard drive as its storage and can be a single-volume or multiple-volume drive. A desktop external drive always requires a separate power adapter because none of the existing peripheral ports provides enough juice to power a 3.5-inch hard drive. For this reason, they are intended to be used in one place, being connected semipermanently to a host. Some of them are, in fact, too big or heavy to be carried around, such as the Pegasus R6.

Inside an external drive, you'll find a single or a few internal drives.
Inside an external drive, you'll find a single or a few internal drives. Dong Ngo/CNET


Pros of external drives
External drives are generally the fastest way to add more storage space to computer, both in terms of performance and convenience. In fact, Thunderbolt storage devices are now the fastest storage devices on the market, even when compared with internal drives. This is because Thunderbolt's speed caps at 10Gbps, which is much faster than SATA 3's speed of 6Gbps.

External drives are also versatile and come in many different capacities, shapes, and sizes. They are the best way to carry data with you or to quickly make backups of your important data.

Cons of external drives
The main disadvantage of external drives is their inability to share data or unused storage space with other computers or devices, other than the one to which it's connected. Portable drives are also prone to data loss because they can be dropped, get lost, or even get stolen.


Connection standards for external drives

No matter how big or small, external drives use one of the following four peripheral standards to connect to a host.

1. USB or universal serial bus
Living up to its name, USB is by far the most universally used peripheral standard among computers and electronic devices. There are now two main USB standards: USB 2.0, which caps at 480Mbps, and USB 3.0, which caps at 5Gbps. The former is available in virtually all computers made in the last decade, while the latter has gained popularity in just the past three years.

USB 3.0 is backward-compatible with USB 2.0, meaning that a USB 3.0 external drive will work with a USB 2.0 port and vice versa. For best performance, however, both the host and the external drive need to support USB 3.0. The easiest way to tell which USB standard a device supports is via the color of the port or the connector: blue means USB 3.0.

Two USB 3.0 ports (note the color blue) and two FireWire 400 ports (bottom) on the back of a desktop computer.
Two USB 3.0 ports (note the color blue) and two FireWire 400 ports (bottom) on the back of a desktop computer. Dong Ngo/CNET

On the host, the USB port remains physically the same (this port is called A-female), which allows it to host USB devices of any standard. On the other hand, USB devices, those that connect to a host using an USB cable, use a few USB port standards (this port is called B-female), which determines the type of USB cable required for them to work.

A USB cable comes with two connector ends. The end that goes into a host (called A-male) remains the same regardless of USB standards. The other end (called B-male) varies depending on the type of B-female port the device has.

The most popular types of USB cables are the following:

Standard USB cable (A-male to B-male): This is the standard cable used for most printers and USB 2.0 desktop external hard drives.

Standard Mini-USB cable (A-male to mini-B-male): This is a standard USB 2.0 cable to be used with a device that has a Mini-USB port. Older portable drives and smartphones use this type of cable.

Three popular USB cables: Micro-USB (left), standard USB 2.0, and Micro-USB 3.0 (note the blue of its A-male connector).
Three popular USB cables, from left: Micro-USB, standard USB 2.0, and Micro-USB 3.0 (note the blue of its A-male connector). Dong Ngo/CNET

Standard Micro-USB cable (A-male to micro-B-male): This is the most popular USB 2.0 cable, currently used in the majority of smartphones and many portable drives.

Standard USB 3.0 cable (A-male to B-male): This is the regular-size USB 3.0 cable used in most USB 3.0 desktop external hard drives. Both of its connector ends are blue.

Standard Micro-USB 3.0 cable (A-male to micro-B-male) : This is the cable used in most portable USB 3.0 drives. Generally, only the the A-male connector of this cable is blue.

All USB storage devices come with their own cables. But should you lose that cable, don't worry; you're not out of luck. You can use USB cables interchangeably among devices that share that particular type of cable. You can also buy standard cables on the market; they don't cost very much. Just make sure you buy the right type for your device.

2. FireWire, or IEEE 1394
This is an older peripheral standard used mostly for older computers, especially Macs. It comes in two speed standards, FireWire 400 and FireWire 800, that offer speeds of 400Mbps and 800Mbps, respectively. FireWire lets you daisy-chain multiple storage devices together, so you can add more storage to a host that has only one FireWire port. FireWire 400 and FireWire 800 use a cable type of their own and also come in regular and mini sizes. FireWire, however, is becoming obsolete and isn't supported by newer computers.

3. eSATA or external SATA
As mentioned in Part 1 , SATA is the interface standard for connecting internal drives to a computer's motherboard. eSATA allows this interface to work from outside the computer, similar to USB or FireWire. eSATA allows for the same speed as SATA, which currently caps at 6Gbps. Though it's very fast, eSATA is not universally available; most of the time you'll need an add-in card to support this standard. There's only one type of eSATA cable, with both of its ends the same.

A FireWire 800 cable (left) next to a Thunderbolt cable.
A FireWire 800 cable (left) next to a Thunderbolt cable. Dong Ngo/CNET

4. Thunderbolt
Thunderbolt is the latest and the fastest peripheral connection to date. Introduced back in January 2011 and initially made available to Macs only, the new Thunderbolt standard allows for speeds of up to 10Gbps (twice that of SATA 3). On top of that, up to seven devices can be daisy chained to one another without degrading the connection speed. Thunderbolt supports more than just storage; it also can deliver audio and video signals, making it the most versatile and powerful peripheral standard thus far.

Thunderbolt is not perfect, however. For now it's still mostly used in Macs, and it's also very expensive. A Thunderbolt drive costs about $150 more than a drive of the same capacity that uses USB 3.0. A Thunderbolt cable itself costs about $50, and many Thunderbolt devices don't include a cable. The Thunderbolt cable, however, is the first active cable, which means that it's a device all by itself, which allows for consistency in performance and other features. There's currently only one standard Thunderbolt cable and port type.


Coupling a NAS-enabled router and a portable drive is an example of a NAS solution.
Coupling a NAS-enabled router and a portable drive is an example of a NAS solution. Dong Ngo/CNET

B. Network-attached storage

A NAS solution negates the biggest drawback of external drives: it connects to a network and make its storage space available to all devices in the same local network. For home and small-business users, there are two main types of NAS: a dedicated NAS server and a NAS-enabled router.

NAS server
A NAS server is an external hard drive that connects to a network instead of a computer. A NAS server generally connects directly to a router, or a switch, using a network cable. It's similar to a real server that hosts lots of storage space, except it doesn't have a mouse, keyboard, or a display. Instead it can be controlled via a Web interface. To take advantage of a NAS server, you need to connect it to the network using a Gigabit wired connection.

Depending on the configuration, a NAS server can do a lot more than just provide storage. It can work as a streaming server, broadcasting digital content to network media players, host storage for remote users to access via the Internet, and even run apps designed for NAS servers. Currently Synology offers by far the most robust NAS servers on the market that offers pretty much everything you'd expect from a storage device and more.

NAS-enabled routers
These are routers that comes with USB ports, built-in storage, or a drive bay to host a hard drive. Two examples of those with built-in storage are Apple's Time Capsule and the My Net N900 Central from WD. Currently most high-end Wi-Fi routers come with one or two USB ports that you can connect an external drive to.

This type of router generally offers two main network storage features: data sharing and digital-media streaming. They provide the NAS feature as an option for those who just want to use network storage casually. Compared with dedicated NAS servers, these routers offer fewer features and much slower data rates. But for casual usage, some of them, such as the Asus RT-N66u or the Linksys EA4500, are more than fast enough.


Pros of NAS
NAS solutions offer a lot more features than external drives and allow multiple clients to access its storage and resources at a time. For those that come with Internet-based features, the data is also available to remote users without the risk of losing it or hardware damage caused by travel. For many tasks, such as downloading a huge amount of data from the Internet, a NAS server lets you do so without needing to have your computer on.

Cons of NAS The biggest disadvantage of NAS is its limited data speed, which caps at that of the Gigabit Ethernet connection. In the best-case scenario, 130MBps is the fastest data connection you can get from a NAS server, but most servers are slower. Basically, a NAS server's speed varies between the speed of USB 2.0 and the speed of USB 3.0, and is much slower than the speed of Thunderbolt, which gets up to about 700MBps. Note that 130MBps is in no way slow; in fact it's very fast, even faster than the real-world speed of many internal drives. However, having that speed ceiling means that for now you can't use NAS for heavy-duty applications such as HD movie editing.

On top of this, NAS is generally harder to set up than an external drive and is more suitable for advanced or professional users than for novices.


That's it for now. If you have questions, feel free to send them my way via Twitter or my Facebook page (though I can't promise I'll be able to address all of them individually). Check back soon for Part 3 , where I'll talk about backup and redundancy.

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.

 

Join the discussion

Conversation powered by Livefyre

Don't Miss
Hot Products
Trending on CNET

HOT ON CNET

MacBook Pro running slow?

Speed up your MacBook by adding more RAM with these quick and easy steps.