Chapter 9: Hard Drives and Other Storage Media

When the CPU is not working with them, your computer stores programs and data files on magnetic disks, optical disks, and other media. Unlike RAM, which loses data when power is turned off, the system's storage media can retain programs and files, even if the computer is turned off.

This chapter describes the various types of storage media commonly used in personal computers, and the interfaces that the motherboard uses to exchange data with them. It also explains how to install disk drives into a drive bay and how to connect external storage devices through a USB or FireWire port.

As a category, these devices that use storage media are known as mass storage because they can hold extremely large amounts of data. As far as the CPU is concerned, there's no difference among the different types of mass storage; the chipset handles the timing and the specific instructions necessary to find an address on each device.

Hard Drives

A hard drive (or more accurately, a hard disk drive) is a device that contains one or more magnetic disks, along with the mechanical components and electronic circuits required to spin the disk, read and write data, and convert between the magnetic impulses on the disc and digital data that the chipset can exchange with the CPU. Most new desktop computers come with one hard drive, and space for more. A typical desktop computer motherboard can support four or more hard drives (or other mass storage devices). In a laptop computer, there's usually space for just one relatively small hard drive.

IBM introduced the ancestors to today's hard drives in 1973. Because they contained two 30-megabyte disk spindles, they were widely known as Winchester drives, inspired by the famous Winchester 30–30 rifles. Since then, the industry has multiplied the amount of data that can fit in the same space many times, so the capacity of a modern hard drive might be as much as 750 gigabytes or more.

How hard drives work

Hard drives are digital recorders that use a recording technology related to an audio or video tape recorder: they use a thin coat of iron oxide (that's a fancy name for rust) or other thin-film metal media over a carrier to store millions of tiny magnetic bits. In a hard drive, the carrier is a polished metal platter.

A hard drive has several parts:

• One or more oxide-covered disks that hold the data.

• A motor that spins the disks at a constant rate of speed.

• A read/write head that hovers over each spinning platter and converts between magnetic bits and electronic signals.

• An arm for each disk that moves the read/write head to the physical location of each address on the disk.

• A sealed aluminum case that keeps dust, dirt, and moisture away from the disks and heads.

• A circuit board attached to the case that sends and receives instructions from the chipset and converts them to control information for the arms, and exchanges data between the read/write heads and the chipset.

There's an impressive amount of activity going on inside the case. The disks are rotating at up to 160 rotations per second or more, and the arms carrying the heads can change position hundreds of times per second as they move from one part of a disk to another. Unfortunately, you must to take my word for this because opening the case would contaminate the platters and heads, and effectively destroy the data stored on that drive. Figure 9.1 is a photo supplied by a drive manufacturer that shows a hard drive with its cover removed.

Figure 9.1: The read/write heads inside the drive are at the end of the arms.

When the drive receives an instruction from the CPU, that instruction includes the address where the program or data file is stored. The logic circuits on the drive's circuit board translate that information into the physical location on one or more of the disks and instructs the arm to move the read/write heads to each location. When a head reaches its target location, it reads the data stored at that address or writes new data.

If you examine a hard drive, you may notice that the screws holding the case together require a special screwdriver, and some of the screws might be covered with labels that say "Warranty void if this label is removed." The drive manufacturers are serious about discouraging users from opening up a case. Don't take this as a challenge-you really will destroy the drive and lose all your files if you try to open it.

The disk drive uses tracks (concentric circles around each disk) and sectors (short segments of each track) to define addresses (see Figure 9.2). Before the drive leaves the factory, the manufacturer performs a low-level format routine that sets the physical locations of each track and sector. When you (or the computer maker) install the drive in a computer, you must run a second, high-level formatting routine that sets the file structure. The section Formatting and Partitioning a Drive later in this chapter offers more information about this.

Figure 9.2: The dark circle is a track; the light gray block indicates a sector.

Most drives have a separate head for each platter surface (front and back, or top and bottom), so the number of heads is double the number of platters.

The other unit of measure related to disk drives is a cylinder. You can think of each cylinder slicing through all the platters inside the drive, although it's purely a way to calculate the capacity of the drive, so there aren't any physical cylinders involved. The computer's BIOS uses the number of cylinders and the number of heads (one for each surface) to determine the physical location of each data address.

Choosing a hard drive

There are three reasons to buy a new hard drive:

• As part of a new computer

• As a replacement for a drive that has failed

• As an additional drive to increase your computer's storage capacity

In a new computer, you can be confident that the drive is fully compatible with the rest of the computer, so the only important thing to decide is the drive's capacity. You can never have too much storage space inside your computer, so a bigger drive is usually a better choice. But remember that you can always add a second drive to your desktop computer when the first one fills up, and the cost per megabyte a year or two from now will almost certainly be less than it is today.

If you're buying a laptop computer, look for the largest capacity that you can afford because you can't add a second drive inside the computer.

When you're replacing a damaged drive or adding a second (or third, or …) drive to an existing computer, the new drive's interface should be compatible with the motherboard (that is, if your motherboard only has IDE sockets, you must use an IDE drive; if the motherboard includes SATA sockets, you can use an SATA drive). Unless you really need an enormous capacity, look for the sweet spot that offers the lowest cost per bit. For example, you might find three drives on sale at your local retailer: a 100GB drive for $60, a 250GB drive for $90, and a 400GB drive for $175. In this case, the 250GB drive is the best choice.

Hard drives are bigger, faster, and more reliable than they were just a few years ago, and it's a safe bet that next year's models will be even better. All the major brands sell excellent quality products that should meet the needs of most users.

Size

The type of computer dictates the physical size of the hard drive. Drives for desktop computers are 4 inches wide and about 5.75 inches deep. Most desktop drives are 1 inch high, but some high-capacity drives with more platters inside are 1.63 inches high. Because these drives fit the same drive bays as a 3.5-inch floppy disk drive, you might see desktop drives described as 3.5-inch drives.

Most laptop computers use smaller drives that are 2.75 inches wide by 3.94 inches deep. The most common laptop drives are 0.75 inches (19 mm) high, but some very compact computers might use drives that are only 0.67 inches (17 mm), 0.49 inches (12.5 mm) or even 0.37 inches (9.5 mm) high. Based on the size of the platters inside the drive, these are known as 2.5-inch drives. Figure 9.3 shows the relative sizes of a laptop drive and a desktop drive.

Figure 9.3: Laptop drives (right) are smaller than drives for desktop computers (left).

Even smaller hard drives also exist, for use in sub-notebook computers, mobile telephones, and other portable devices. These tiny drives are not used as internal drives inside personal computers, but you might find one that fits a laptop computer's PC Card socket as a removable storage device.

Data capacity

The capacity of a hard drive depends on the number of platters inside the drive, the physical size of each platter, and the density of the data stored on the platters. As magnetic storage technology has improved, the maximum amount of data that a platter (and therefore a drive) can hold has increased, and the cost of storage has dropped.

Unfortunately, this increase in capacity has happened more quickly than the BIOS companies had expected. Because of the way the BIOS specifies addresses, there's a practical limit to the number of cylinders that it can use, even if the number of cylinders on the drive is greater than that maximum. In order to overcome this limit, you may have to either update the BIOS or use special translation software supplied with the drive. If the computer shows a significantly smaller capacity than you expect, look in the Support section of the drive manufacturer's Web site for instructions for bypassing the size barrier.

Note 

Don't let the difference between decimal kilobytes (equal to 1,000 bytes) and binary kilobytes (equal to 1,024 bytes) confuse you. Some drive manufacturers use the decimal capacity, but Windows and the BIOS utility reports that same capacity in binary, so a 200GB (decimal) drive might appear to have only about 180GB.

Speed

The specifications for a disk drive usually include several speed ratings; the two most important values are the average seek time needed for the drive to move a head from one address to another and the rotational speed of the platters.

Seek time is important because it reflects the amount of time needed for the CPU to request and receive data from the drive. Fast as it is, a drive with a seek time of nine milliseconds (ms) is exchanging data with a CPU that can perform millions of processes in that time. So a faster average seek time makes a huge contribution to the computer's overall performance.

As the name suggests, rotational speed is the rate at which the platters spin inside the drive, expressed in rotations per minute. The most common rotation speeds for new drives are 5400 RPM, 7200 RPM, and 9600 RPM, although some high-performance drives offer speeds up to 15,000 RPM. If you transfer a drive from an older computer, you might find one as slow as 3600 RPM, but those drives are probably at the end of their useful lives. Faster rotation is better, especially in servers and other high-demand systems, but it might not always be noticeable in home and office computers.

Cache buffer

The cache buffer is a RAM chip built into the disk drive circuitry that it uses as a buffer between the fast CPU and chipset and the relatively slow drive. It typically holds copies of the data that the drive has just transferred to the CPU or from the CPU to the buffer most recently, and one or more sectors adjacent to the one that moved most recently.

Even a small cache buffer improves the performance of a drive, but increasing the size of a buffer doesn't always provide a significant additional improvement. A 2MB buffer is enough for most single-user computer applications except games, streaming video, and multimedia editing, and an 8MB buffer should be entirely adequate for everything but high-usage servers. The difference in cost between a small cache and a larger one is often insignificant.

IDE, SATA, and SCSI Interfaces

The hard drive exchanges data with the CPU and chipset through a socket on the motherboard or on a plug-in expansion card. Most current computer models use either the IDE (Integrated Device Electronics) interface or the newer SATA (Serial ATA) interface. Some servers and older desktop computers might use the SCSI (Small Computer Systems Interface) system, but they are not common in new desktop and laptop machines.

CROSS-REF 

For more information about the CPU, chipset, motherboard, and expansion cards, go to Chapter 4.

IDE

For many years, the most common type in desktop and laptop systems has been the IDE interface. IDE is a parallel interface that transfers 16 bits (two 8-bit bytes) at a time, using control circuits located on each drive. The IDE standard is also called ATA (Advanced Technology Attachment), PATA (Parallel ATA) and EIDE (Enhanced IDE). Figure 9.4 shows a pair of IDE sockets.

Figure 9.4: This motherboard has two IDE sockets. The smaller connector at the bottom is for floppy disk drives.

The computer's motherboard normally has two IDE sockets that can accept cables from hard drives, CD or DVD drives, and other types of mass storage. Each IDE socket can support two disk drives or other IDE devices through a single cable, usually identified in the BIOS as Master and Slave. The position of a jumper on each drive identifies it as either a Master or a Slave. Figure 9.5 shows the connectors on an IDE drive. Table 9.1 lists the signals on each pin in an IDE connector.

Figure 9.5: IDE drives use jumpers to identify themselves as Master or Slave.

Table 9.1: IDE Connector Pins
Open table as spreadsheet

Pin #	Signal Function	Pin #	Signal Function
1	Reset	2	Ground
3	Data 7	4	Data 8
5	Data 6	6	Data 9
7	Data 5	8	Data 10
9	Data 4	10	Data 11
11	Data 3	12	Data 12
13	Data 2	14	Data 13
15	Data 1	16	Data 14
17	Data 0	18	Data 15
19	Ground	20	Key
21	DMARQ	22	Ground
23	DIOW-	24	Ground
25	DIOR-	26	Ground
27	IORDY	28	CSEL
29	DMARK-	30	Ground
31	INTRQ	32	IOCS16-
33	DA1	34	PDIAG-
35	DA0	36	DA2
37	CS1FX-	38	CS3FX-
39	DASP-	40	Ground

SATA

The SATA interface has been introduced as a replacement for IDE, but the computer industry is still in a transitional period when both interfaces are widely used. Most new motherboards have sockets for both types.

Instead of the 40-pin IDE connector and cable, SATA drives use a 7-pin data connector. Each SATA socket supports just one drive, so there's no need for jumpers to identify Masters and Slaves. Table 9.2 lists the SATA pin connections.

Table 9.2: SATA Connector Pins
Open table as spreadsheet

Pin No.	Signal Name	Description
1	GND	Ground
2	A+	Transmit+
3	A−	Transmit−
4	GND	Ground
5	B−	Receive−
6	B+	Receive+
7	GND	Ground

Figure 9.6 shows the connectors on an SATA drive.

Figure 9.6: SATA drives don't need jumpers.

SATA is a serial interface, with a new design that offers several advantages over the older IDE drives:

• SATA drives are easier to install because they don't require jumpers.

• SATA drives don't produce as much heat.

• SATA drives use smaller cables that don't interfere with airflow inside the computer case.

• SATA drives can support faster data transfer than IDE.

That faster data transfer doesn't make much difference yet. In today's computers, IDE and SATA drives with the same specifications have similar performance. Up until now, there has been no reason to provide chipsets that could support a faster data transfer rate than the IDE drive design could handle. But SATA eliminates that bottleneck, so the next generation of motherboards and chipsets will support faster data transfer.

SCSI

The SCSI standard is a parallel interface that uses a single controller and a daisy chain of up to 15 internal or external devices. SCSI devices can include input and output devices (such as scanners) as well as storage devices. Each device in a SCSI chain has a unique ID number. A new, faster Serial Attached SCSI (SAS) standard is also available.

SCSI drives permit faster data transfer, and the interface can support more devices than IDE or SATA, so they're widely used in servers. However, they're not often used in modern PCs for home or office use. Most desktop and laptop computers don't work with SCSI drives unless you install a SCSI controller in an expansion slot or a PC Card socket.

Installing a new hard drive

Adding or replacing a hard drive inside your computer might seem like a major project, but it's not as difficult as it might appear. The drive bays in a desktop system have been designed to facilitate easy access.

Jumper settings

Each IDE controller can support two drives through the same cable: a Master and a Slave. Remember Figures 9.5 and 9.6? Before you install an IDE drive, you must configure the drive as either a Master or a Slave by moving the jumpers on the edge of the drive to the correct set of pins. SATA drives don't require any jumper settings.

Most drives have labels that show the different jumper settings that apply to that drive. If your drive has no label, consult the manual or the manufacturer's Web site.

The boot drive is normally the Master drive on the Primary IDE cable. However, each of the other drives in your system can be either a Master or a Slave, as long as you don't create a Slave drive on a cable that doesn't have a Master drive.

If you're replacing an existing drive, the new drive must have the same setting; if you're adding a drive, you need to know how the drives that are already in place are configured.

There are a couple of ways to identify a drive's settings. The easiest way is simply to look at the jumpers on the edge of the drive and compare them to the label on the top or bottom of the drive. Different brands of drives use different jumper arrangements, so there's no universal set of standard jumper settings.

The software CDs supplied with most new hard drives include a utility program that scans your existing drives and identifies their settings. If you don't have a CD, you can download the software from your drive manufacturer's Web site.

As an alternative, you can use the BIOS Settings utility to tell you which drives are assigned to each IDE channel and which are Masters and Slaves. This can often be easier than removing each drive to check the jumpers. To view the assignments, follow these steps:

1. Restart the computer and press the appropriate key to open the BIOS Settings utility.

2. Go to the screen that shows all of the drives in this computer. Depending on the type of BIOS, this might be the Main Menu, the Standard CMOS Features screen, the Drive Settings screen, or it might have some other name, but it has a list that includes:

   • Primary Master

   • Primary Slave

   • Secondary Master

   • Secondary Slave

   Or possibly:

   • IDE Channel 1 Master

   • IDE Channel 1 Slave

   • IDE Channel 2 Master

   • IDE Channel 2 Slave

3. The list identifies the type of drive connected to each channel, including hard drives, the CD or DVD drives, and other IDE devices.

If you're replacing an existing drive, find that drive on the list in the BIOS utility and note whether it's a Master or a Slave. Set the jumpers on the new drive to match the old one.

If you're installing a new drive, look for a channel with no drive in place. The BIOS utility might show a blank space for one or more channels, or it might say "No Drive Detected" or something similar. Note whether the unassigned space is a Master or a Slave and set the jumpers on the new drive to fill that space.

When you have set the jumpers on your drive, press Esc or F10 to leave the BIOS utility without making any changes. Let the computer restart and immediately use the Windows shut down routine to turn it off.

In a desktop

It's always helpful to read the installation instructions supplied with a new drive or any other computer part. If you discover that the manufacturer's instructions contradict the instructions in this book, follow the maker's advice.

Tip 

Seagate (one of the major drive manufacturers) offers several videos that provide detailed step-by-step instructions for installing a disk drive. Even if you're using some other brand of drive, these video guides can help you understand how to do it. Look for the Seagate Online Multimedia Install Guide at http://www.seagate.com/support/howto/install_guide.

To install a disk drive into a desktop or tower case, follow these steps:

1. Turn off the computer and unplug it.

2. Remove the cover from the case. If you have an anti-static grounding wrist strap, put it on and ground yourself to the case.

3. If you're replacing an existing drive, disconnect the cables from that drive and remove the drive.

4. Find an empty drive bay. If you're installing a hard drive, look for an internal drive bay; for a drive with removable media, use one of the spaces in the front panel. Choose a space that allows air to flow around the drive, and make sure your data and power cables can easily reach the connectors on the new drive.

5. Examine the mounting arrangement of the drive bay you plan to fill. In some cases, you slide the drive into the bay, but in others, the drive mounts in a removable frame.

6. Use at least four of the screws supplied with the drive to attach the drive to the frame. If necessary, mount the frame inside the computer.

7. Connect a power cable from the power supply to the power connector on the drive. If you're installing an SATA drive and the power supply doesn't have SATA connectors, you need a power cable adapter (available from a computer retailer).

8. If you're adding a Slave to a channel that already has a Master, find the unused connector on the data cable already connected to the Master and plug it into the new drive.

9. If you're installing a Master drive on the Secondary Channel, use the data cable supplied with the drive or buy a round replacement IDE cable. Plug one of the data connectors into the drive. The cable connectors probably have labels that show which ones go to the Master, the Slave, and the motherboard.

10. If it's not already connected, plug the other end of the cable (the motherboard connector) into the empty IDE or SATA connector on the motherboard.

11. Turn on the computer and immediately press the key that runs the BIOS Settings utility.

12. In the BIOS utility, move to the page that lists the IDE and SATA channels.

13. Choose the channel for the drive you just installed. A new screen appears with details about the drive.

14. Select the Auto-Detect line in the menu and press Enter. This instructs the utility to identify the drive and reads the details into the BIOS.

15. If the BIOS doesn't recognize the drive, make sure you have connected the cables and set the jumpers correctly.

16. When the BIOS has accepted the drive details, use Esc or F10 to save the changes to the BIOS and restart your computer.

17. Replace the cover.

18. Run the Drive Installation program supplied with the new drive to partition and format the drive. There's more information about partitioning and formatting later in this chapter.

In a laptop

Installing a drive in a laptop is a different process because the computer has only one drive and it usually plugs directly into a connector on the motherboard. However, every laptop make and model is different, so there's no common method for opening the case and removing or installing a drive. Look on the computer maker's Web site for step-by-step instructions.

The designs of different laptop computers vary widely; in some computers, the disk drive is easy to remove and replace, but in others it's extremely complicated. If the instructions on the Web site or the Service Manual sound like they are excessively difficult, let a professional handle the project for you.

Any time you work on a laptop computer, remember to disconnect external power and remove the battery before you try to open the case.

Read the instructions supplied with the drive for information about jumper settings. The single drive inside a laptop computer is always the Master, but some drives still provide jumpers in case you want to install the same drive in a desktop machine (using an adapter that converts the laptop pin arrangement to the ones used in a desktop drive bay).

Don't try to remove or replace a laptop drive without specific instructions. In many computers, you might have to push, pull, or squeeze the drive or a mounting bracket in a way that's not obvious. If you have to force the drive to move it, you're probably doing something wrong.

Formatting and partitioning a drive

In addition to the low-level formatting that sets the locations of tracks and sectors on the drive's platters, a disk drive also requires a separate high-level formatting process that creates the drive's file structure and writes the boot files onto certain sectors. Because several different files structures are possible, the drive makers leave high-level formatting to the computer builder or user who installs the drive. In most cases, formatting a drive means performing a high-level format.

A new drive always requires formatting, and formatting can also be a last-resort method for repairing a drive with corrupted sectors or files. This corruption might be caused by a virus or by random corruption that happens for no apparent reason. Remember that formatting a drive can destroy all of the files stored on that drive, so it's always good practice to try other recovery techniques first. Some computer users (and some incompetent tech support advisors) suggest reformatting as a quick and easy way to fix an unknown problem, but there's almost always another, less destructive approach.

Partitions are separate sections of a physical drive that appear to Windows as different drives. For example, a disk drive with two partitions might show up in Windows as the D: drive and the E: drive. You can split a new drive into partitions when you format it.

As you know, Windows holds individual programs and blocks of data as files, which it stores in folders or directories. The file system is the structure that specifies the format of file names, and the way a drive stores and organizes those files. The operating system uses the file system to find and read files and folders. In Windows XP, the preferred file system is NTFS ([Windows] NT File System), but it can also recognize the older FAT16 (16-bit file allocation table) and FAT32 (32-bit FAT) systems.

The size of a partition was limited in some older file systems, so it was sometimes necessary to split large physical drives into two or more logical drives. This is not a problem with NTFS, so most users might want to create just one partition on each drive. However, separate partitions might be useful when two or more users share a single computer as a method for isolating each user's data files. Other users like to keep programs on one partition and data files on another, and still other use separate partitions to load more than one operating system (such as Windows and Linux) on the same drive.

The formatting tools supplied with new hard drives and the utility included in Windows XP offer all three files systems, but the NTFS system is almost always the best choice. The exceptions might be computers that can load more than one operating system (through a startup menu).

All of the major drive manufacturers supply formatting software with new drives unless they're sold in bulk OEM (original equipment manufactuer) packages. If you don't have a software CD for your drive, you can download the formatting program and detailed instructions from the manufacturer's Web site.

If you're adding a new drive to an existing system, you can use the Disk Management program supplied with Windows XP to format and partition the drive instead of the program supplied with the drive.

To format a drive with the Windows Disk Management tool, follow these steps:

1. Complete the drive installation process described earlier in this chapter.

2. From the Start Menu, select Programs Administrative Tools Computer Management. You can also open Administrative Tools from the Control Panel.

3. Choose the Disk Management item at the bottom of the list in the left pane of the Computer Management window. The screen shown in Figure 9.7 appears.

   
   Figure 9.7: The Windows XP Disk Management tool can format and partition hard disks.

4. Right-click the drive you want to format in either the list at the top right of the window or the bar graph in the lower right section.

5. Choose Format from the pop-up menu.

6. Follow the instructions as the program walks you through the formatting process.