Hard Drives

Hard drives are mass storage devices. Virtually all computers today have at least one hard drive. Early hard drives were capable of storing only a small amount of data (10 MB), they were large, and expensive compared to today’s standards. At the time, the capacity was not a problem as the size of programs, and the work requirements put on them, were small by today’s standards. The first hard drive was about 4 inches tall, six inches wide; eight inches long and weighed almost ten pounds. A new hard drive can fit into your pocket and hold as much as 60 GB (with capacities of new drives increasing all the time). In this section, we will study hard drives from the early versions to today’s monsters.

Technology of Hard Drives

The first means of mass storage was the magnetic tape drive. Tapes are a good media for storing large amounts of data, however, as we said earlier, accessing data was slow and linear. Floppy drives resolved some of the problems with speed and data access, but had limited storage capacity. Hard drives were invented to provide a storage media that not only held large amounts of data, but were fast and allowed easy (random) access to that data. Using the floppy as a model, designers created a disk drive that could hold up to 10 Megabytes of data. At the time, this was a phenomenal amount of data. Since the diskette was not removable, or flexible, it became know as a “fixed disk” or hard disk drive (HDD).

The first IBM drives came out in the late 1970s and early 1980s and were code named “Winchester.” The original design concept included two 30 MB units in one enclosure: 30-30 (hence “Winchester”). The PC-XT was the first personal computer to include a hard disk. They were called “fixed disks” because they were not removable. The Winchester technology is the direct ancestor of all PC fixed disks.

A hard disk operates on the same principal as a floppy disk. It has a spinning platter with a pair of read/write heads that traverse perpendicular to the rotation of the platter. What makes the hard drive different is that the platters are made of an aluminum alloy and have a thin magnetic media coating on both sides. These ridged disks are responsible for the higher storage capacities available on hard drives. First, designers can install more than one platter per drive, providing for more surface area and thus more storage. Second, they can spin the disk faster (3,600-10,000 rpm for hard disks as compared to only 300- 600 rpm for floppy disks) and have better control over the read/write heads. This increased accuracy, higher speeds and more surface area allow for capacities in the gigabyte range instead of the 1.44 megabytes of a floppy. In addition, unlike the floppy drives, a hard drive assembly is housed in a sealed case, which prevents contamination from the surrounding environment. Keeping the inside air clean and all contaminates out is essential to long life of the drive. The tolerance or space between the surface of the platter and read/write heads is so small that the mere presence of a fingerprint can cause damage to the drive.

Read/Write Heads

The purpose of a hard drive is to achieve fast, random access to data stored on a flat surface. This is accomplished by using motion in two directions. The disk spins and the read/write heads are moved across the platter perpendicular to the motion of the disk. It should be obvious that the heads should not (and in fact do not) touch the surface of the platters while running. This would no doubt cause a catastrophic failure commonly known as a head crash. When this happens, the only solution is to throw the drive away and start over. Based on the Bernoulli principle (in any horizontally moving fluid the pressure increases as the velocity of flow decreases), the heads “ride” on a cushion of air. The interaction between the air in the chamber and the surface of the moving disk creates an interface of disturbed air that the heads ride on. Some designers in the hard drive business call this cushion of air an air bearing.

A critical element in hard drive design is the speed and accuracy of moving the heads across the platen. The read/write heads are mounted on an actuator arm that pivots so that the heads will traverse perpendicular to the rotation of the platen. Two designs have been employed that meet these requirements. They are stepper motor actuators and voice coil actuators.

Early hard drives used a stepper motor to move the actuator arms in fixed increments or steps. A stepper motor is a very precise motor that doesn’t spin continuously like we think motors should. It moves in discrete steps. They operate on pulsed electricity; each pulse will move the motor one step. For example, if you had a 360-stepper motor, it will rotate one degree per pulse or step. If you want to rotate the motor one half of a rotation, you would send it 180 pulses. If you were to grab the end of a stepper motor that was turning, you will be able to feel the pulsing of each step. Stepper motors are commonly used in modern robots and animatronics to create precise movements of mechanical arms.

While very accurate for these larger applications, stepper motors have some problems when used in hard drives. The biggest problem comes with heat. As the drive warms up, the physical expansion of the mechanical parts and the platter can change the positioning. Since the stepper can only move in predetermined steps, errors can occur. This is one reason why you should never try to run a hard drive (floppy either) that has been left out in the cold without giving it time to warm up to room temperature. The second cause of problems is time and physical deterioration of the components. A third problem, read/write heads need to be “parked” when not in use with actuators. Remember, the heads do not touch the surface of the disk when it is rotating. However, when it stops rotating, they will contact the surface. This can cause damage to data. Therefore, when the drive is shut down, it must park or move the heads to an area of the disk that does not contain data.

The stepper motor drives of the past have pretty much been replaced with voice coil actuators. They are called a voice coil motor because they are designed using the same technology as the voice coil found in a stereo speaker. This principle uses a permanent magnet and a coil that is connected to the speaker’s paper cone, or in the case of a hard drive, the actuator arm. By passing electrical current through the coil, it generates a magnetic field that moves the actuator arm into the proper position. These motors have proven to provide a greater degree of accuracy, performance and reliability over their predecessor. The accuracy is achieved by providing a feedback signal from the drive to determine the actual position and make adjustments. Usually, one platen is used to provide this feedback data. This way, when heat causes expansion of the platen, the motor can compensate.

The biggest advantage of the voice coil motor is that you no longer need to park the heads. When the drive is shut down, (the power is removed from the coil) the actuator arm (which is spring loaded) moves back to its initial position thus eliminating the need to park the head. In a sense, they are self-parking.

Magnetically Storing Data

As we learned previously, computer data is stored in BINARY language. In the case of physical memory, it is the presence of voltage that represents the binary one in each cell of the memory array and the lack of voltage that presents binary zero. This works well, as long as there is electricity. For permanent storage, it will either have to stay connected to a battery or find some other way of recording ones and zeros. To the rescue is magnetic media. With magnetic media, the ones and zeros can be stored as either magnetic or non-magnetic areas on the drive surface. It is based on the principal of electromagnetism. This states that any time electric current flows through a conductor; a magnetic field is generated around the conductor. In addition, if a conductor is passed through a changing magnetic field, current is generated. The direction of the current is based on the polarity of the magnetism. Using these principles, magnetized and nonmagnetized positions are created on the surface of the platter. The binary data is not stored as magnetic poles (plus or minus) but as flux reversals. The term flux defines a magnetic field that has a specific direction. A flux reversal is the change in polarity. Each bit written on the drive creates a pattern of positive-to-negative or negative-to-positive flux reversals on the medium.

The read/write heads are used to read data from the magnetic surface or alter the surface for storage of data. They are made of U-shaped electrically conductive materials. By wrapping the conductive material with coils of wire, and passing current through the coils, a magnetic field can be produced that will write data to the disk. Alternatively, by passing the read head over the magnetic fields on the surface of the disk, a current is generated thus reading data from the disk. The process of reading and writing to a disk is called encoding.

The first method of encoding (back in the 70s) was called FM (Frequency Modulation). This method was commonly used on single density floppy disk (single density encoding). In order to achieve greater storage capability, an upgraded version called MFM or Modified Frequency Modulation was implemented. This technique allowed twice as much data to be stored on the same disk, thus the double density disks (double density encoding). MFM is the standard for floppy disk encoding today.

The next standard encoding method for hard drives was called RLL (Run Length Limited). Due to its increased performance and ability to store more data on a disk than either of its predecessors, this method soon became the standard for use on PC hard drive.

Two new technologies have been developed that have become today’s standard for data encoding on modern hard disks. These are PRML (Partial Response, Maximum Likelihood) and its successor EPRML (Extended Partial Response, Maximum Likelihood).

These methods are a complete departure from earlier technologies in the way that data is read and decoded on a disk. Instead of reading flux reversals, a controller employs sophisticated digital signal sampling, along with processing and detection algorithms to manipulate the analog data stream on the disk (the partial response). It will then determine the most likely sequence of bits this represents (maximum likelihood). EPRML is simply an improved way of doing this.

This sounds like a strange and unreliable way to read data on a hard disk, but is the standard used currently.

Hard Disk Architecture

In order for data to be stored and retrieved on the surface of a hard disk, the data must be organized. This organization will determine two things. First, is the maximum amount of data that can be stored, and second, how that data is retrieved. As we learned, hard drives are composed of one or more disks or platters on which data is stored. The physical organization of data on these platters if often referred to as the logical geometry of a hard drive. It is this logical arrangement of data that determines the maximum storage capacity of the drive.

The logical arrangement of data on a hard drive is based on five numerical values.

Write Precomp and Landing Zone are obsolete, but are often seen on older drives.

Knowledge of the physical layout of data is required in order to properly install and configure a hard drive. The BIOS needs to know how to find the data. If you look at any documentation that comes with a new hard drive, you will find information regarding the cylinders, heads and sectors. These values are commonly called CHS values (Cylinders, Heads, Sectors). Also, many drives will have this information printed on the label.

The following is an example of the documentation for a Seagate drive.

Now let’s look at the individual components and see how they affect storing data.

Cylinders

Data is stored in concentric circles on the surface of each platter. Each concentric circle is called a track. A set of tracks (all tracks of the same diameter) through a stack of heads, is called a “cylinder.” The number of cylinders is the logical number used to describe the drive (not the number of tracks). BIOS limitations set the maximum number of cylinders (for calculation purposes) to 1024.

Heads

We have already talked about hard drives containing one or more platters. Each side of a platter is considered a head. Therefore, the number of heads equals the total number of sides of all the platters used to store data. If a hard drive has 6 platters, it could have up to 12 heads. Hard drives using voice coil motors for actuator arms must reserve a head for navigation (accuracy of the arm position). Since platters are two sided it is logical to always have an even number of heads. However, if one is used for navigation, the reported number of heads will be an odd number. Again, the BIOS limitations set the maximum number of heads to 16 (for calculation purposes).

Remember, CHS numbers are logical values. This has allowed some hard drive manufacturers to use a technology called sector translation to create more than one

logical head on a single side of a platter. Sector translation is a software conversion of the actual values into values that the BIOS will accept. Regardless of the methods used to manufacture a hard drive, BIOS limits the maximum number of heads (for calculation purposes) to 16.

Landing Zone

As we mentioned, early drives using stepper motors required a place to park the heads when the drive is not running. A landing zone designates one cylinder for safely parking the read/write heads. A landing zone is only used on older drives that have stepper motors to drive their actuator arms.

Sectors

The third element of CHS values is Sectors. A hard drive is cut (figuratively) into small arcs (like a pie). Each arc is called a sector, and holds 512 bytes of data. Each sector will contain, in addition to the data, header information that includes the sector ID or number and trailer information that includes a checksum to insure data integrity. As with the other CHS values, the BIOS sets a limit on this one too. BIOS limitations set the number of sectors at 63.

Write Precomp

If you look at the graphic of a sector or a slice of pie, it is obvious that the outside regions are larger than the inside regions. This is the same with sectors; the outside ones are physically larger than the inside ones. Since all sectors can logically store only 512 bytes of data, the ones on the outside have space left over while the ones on the inside are cramped. The technology of early drives needed a method of compensating for this anomaly. This method was called “write-precomp.” It literally means a point (the cylinder) at which the drive needs to compensate for the size difference of sectors. As we saw in our example of the Seagate drive, write precomp is no longer used.

Calculating the Size of a Drive

When installing a hard drive, it is important that you provide the computer’s BIOS the information it needs to recognize the drive so that it can address all the sectors that contain data. If the computer does not know these values or has the wrong values, it will not recognize the drive. Once you know the three CHS values of a drive, you can calculate the capacity.

Lets look at calculating the maximum hard drive size when considering the BIOS limitation of CHS. The BIOS limits are:

1024 cylinders

16 heads

63 sectors

512 bytes per sector

Therefore:

1024 x 16 x 63 x 512 bytes/sector = 528,482,304 bytes

This value can be rounded to 528 megabytes or if you divide the total number of megabytes (1024 x1024 or 1,048,576) into the total number of bytes, you will get 504 MB. Keep these two numbers in mind as they are often used interchangeably.

Based on these values, the largest hard drive size recognized by the BIOS is 504 MB.

Today’s standard drives far exceed these early limits. In fact, it would be difficult to find a computer with less than a 2 GB hard drive. However, as a computer technician, you may encounter one of these older drives. Larger drives manage to exceed this limitation in two different ways. We mentioned sector translation as a method of converting logical values into values that the BIOS can understand. Other translation methods can be employed that allow the BIOS, the CPU and the drive controller to report and see the CHS values that they need to run. The other method is to replace the BIOS. (More on this to follow).

If you purchased one of the original personal computers, you will remember that hard drives were not included. As hard drives were introduced into the personal computer market, a large variety of drives became available, each with their own CHS values. To simplify installation, IBM created standard “types” of drives. To configure a drive, all the technician had to do was enter the drive number into the BIOS setup. The correct CHS values would then be properly configured. This system worked for some time, but as increasingly drives entered the market, increasingly drive types had to be added to the list. This created an additional problem in that BIOS’s were outdated because they did not contain the latest list of types. So, at type 46, the decision was made to stop adding new types and use type 47 as “User”. From this point on, the BIOS did not need to be updated, but the technician had to select type 47 and manually configure the CHS value.

Today, the process is even more simplified. A 48th type has been assigned. That type is ”Auto”. With auto, the BIOS information is stored on the hard drive itself. When running the BIOS setup, a ROM chip on the hard drive will be queried and the following data collected:

· Buffer type indicating sector buffering or caching capabilities.

· Cylinders in the current translation mode.

While this is simple, as an A+ technician, you will need to be familiar with all three methods of configuring BIOS for a hard drive. Those methods are:

Enter the drive type number.

Use type 47 “USER” and enter the values manually.

Use the auto function and let the BIOS collect the information from the drive.

Hard Drive Interfaces

The development of hard drives followed two paths. The first and most obvious was to be able to store more and more data in a smaller and smaller place. This growth was based on the technology of the read/write heads and how accurately information could be stored and retrieved. The second path was around the interface. A hard drive interface is the standard that determines how the drive communicates with the motherboard and CPU. In fact, hard drives are actually categorized by the interface. Let's look at hard drive interface technology as it progressed from the early drives to today’s behemoths. With each new interface came many advantages. As a computer technician, you will have to know these differences.

ST506/412

Seagate introduced their ST-506/412 drive in 1983. This drive required that the ROM BIOS chip be installed on the controller that was a separate component at that time. These full-height 5-¼ inch drives have taken their place in the annals of computer history books alongside the Hollerith card. It was not until the introduction of the AT system that the BIOS was placed on the motherboard. These drives are easily identified, as unlike hard drives of today, these drives require a separate cable for data (20-pin) and the controller (34-pin). In addition, like its predecessor the floppy drive, the controller cable could connect to two drives and uses a twist to select between the two different drives.

ESDI

The first attempt at hard drives produced a workable, but low performance mode. It became obvious very early that hard drives were here to stay and performance

improvements would be required. In 1983, the Maxtor Corporation introduced the next generation hard drive. This drive was called ESDI or Enhanced Small Device Interface. Performance enhancement was achieved by integrating the controller functions directly onto the hard drive itself. This approach greatly improved data transfer speeds. In addition to the controller, some ESDI controllers offered enhanced command sets. One of the features of the new command set was to support auto sensing of the drives.

The installation of ESDI drives was no problem as it is almost identical to the installation of ST-506 drives. As with their cousin the ST-506/412, these drives are now part of computer history.

IDE

The replacement to the ESDI came on scene in the early 1990s. This new drive was called the IDE or Integrated Drive Electronics (pronounced Eye-Dee-E). The name IDE is derived from the fact that the drive has a built-in (integrated) drive controller. These drives incorporated the benefits of both of its predecessors. IDE quickly became the standard for hard drives in personal computers.

The installation of the original IDE was quite different from its predecessors. These drives were designed as part of an expansion card and were installed into a free 16-bit ISA slot on the motherboard. Because of this, they were often called hardcards. From these first hardcards was developed what is now considered the standard 40-pin connector on motherboards. These connectors are actually a compressed version of the standard 98-pin standard 16-bit ISA bus slot. The reduction of pins is simple because all the pins not used by the controller were simply eliminated. Western Digital and Compaq developed the first ATA IDE drives and were the first to establish the 40-pin IDE connector specification that is now standard in all computers. ANSI standards committees accepted the standards as the Common Access Method (CAM) AT. The official name for these drives is now ATA/CAM (AT Attachment/Common Access Method). The terms IDE and ATA/CAM are often used interchangeably. With this standard, drives started using a single cable with no twist. Since two drives can be connected to one cable, each drive must be individually configured as either a master or a slave. Some drives have stand-alone and auto configuration as well. Switches or jumpers on the drive are used to complete the configuration. This is now a standard method for installing drives.

ATA

ATA is an implementation that integrates a controller right on the disk drive making the device more intelligent. It functions as the “go-between” for data transfer. ATA can transfer data up to 6.67 MB/s or if utilizing a cache-hit burst, up to 10 MB/s. It is somewhat limited in its ability to attach devices. It can only handle two channels with two drives per channel. This particular implementation is the common one, especially for off the shelf units.

This version is also known as IDE. It has a 16-bit interface and can support up to two hard drives as well as PIO (Programmed Input/Output) modes 0, 1, and 2.

ATA-2 (Fast ATA/Enhanced IDE [EIDE])

This version supports multiword DMA modes 1 and 2. It supports PIO modes 3 and 4, which are faster than PIO modes 0, 1, or 2. It also supports LBA (Large Block Addressing).

ATA-3

This version held little change from ATA-2, but it introduced SMART (Self-Monitoring Analysis and Reporting Technology).

ATA-4 (Ultra-DMA/ATA-33/DMA 33)

This versions supports multiword DMA mode 3, which runs at 33 MB/s. It also introduced CRC (Cyclic Redundancy Check) error checking.

ATA-5 (ATA/66 / Ultra DMA/66 / Fast ATA-2)

This version doubles the throughput to reach 66 MB/s, because it doubles the burst rate of ATA-33. This helps with data-transfer bottlenecks. It also uses CRC (Cyclic Redundancy Check) error checking. It uses a 40-pin cable connector.

Ultra-ATA 100 (Ultra ATA DMA Mode 5)

This version has a maximum data transfer rate of 100 MB/s. It also uses CRC (Cyclic Redundancy Check) error checking.

EIDE

The logical successor to the very popular IDE or ATA drive is today’s standard known as the Enhanced IDE (EIDE). It is also called ATA-2 and ATA-3. It has several improvements.

The details of how EIDE increased hard drive capacities and transfer speeds will be covered later in this section (Programmed I/O and DMA). The other two improvements really made a difference in how we perceive our computers today. The original controllers allowed for only hard drives (and only two of them). As we have seen, these drives are connected on the same cable. The addition of two channels (two cables) allows a computer to be expanded further, allowing four drives to be added. This allows the computer’s storage capacity to be physically expanded. This capability was further enhanced with the acceptance of the ATAPI (AT Attachment Packet Interface). The interface was developed by an independent industry group to allow non-hard drives (CDROMs and high-speed streaming tape units) access to the ATA interface. Now almost every computer sold has a standard CD-ROM installed on the second IDE channel for their motherboard.

The most important consideration when installing any drive is to correctly install the cabling. The figure below shows a standard IDE drive. Note the alignment of pin

number 1 and the power connector.

Unlike the twisted cable provided with floppy drives to identify one drive from the other, hard drives use a flat cable with no twist. Therefore, each drive must be designated as either the master or slave so that the BIOS can identify them on the cable. The documentation supplied with the drive should provide the necessary information for configuration. Often this information is printed on the label on the drive. Both drives must be properly configured before starting the system. If the drives are not properly jumpered, they won’t work. This is a very important concept for the A+ Exam!

There are two different drive-naming conventions. The one we are most familiar with is the A:, B:, C:, D:, etc. These names apply to how the drive is formatted, how it is seen by the operating system and its cable orientation. At this point, we are not interested in these names; we are still working at the hardware level where the drives are numbers. If only one hard drive is installed, it must be configured as drive 0 or Master. A second drive on the same cable is installed as hard drive 1 or Slave. If you use two controllers with two cables and four drives, each controller will have a master and a slave. One controller will be designated as the primary controller and one as the secondary controller. A single drive installed on a computer must therefore be connected to the primary controller and designated as the master or drive 0.

Partitioning is used to divide a physical hard drive into logical units. You will need at least one partition on every drive, but you may want to create more. If you intend to use the drive with more than one operating system such as Windows 98 and Windows NT, you will need a separate partition for each. They are not compatible. In addition, if you have a large hard drive (larger than your system will support), you can divide the drive into smaller logical units. Your system will see each of the smaller units, thus allowing you to use the entire drive. First, let us consider our choices for file systems. Each is based on how files are stored and indexed for retrieval.

FAT (File Allocation Table – also known as FAT16) – This is the standard file system for DOS. Subsequently, Windows 3.1, Windows 9x, Windows NT, and Windows 2000 will run on a FAT 16 drive. FAT uses 12 (floppy disk) and 16 (hard disk) bit numbers to identify clusters. Clusters are groups of sectors. By grouping sectors into clusters, drives as large as 2 gigabytes can be recognized. More on clusters in the following section. You can only create two FAT partitions on a single drive. One is the primary and the other is the extended. The extended partition can be further divided into as many a 23 logical drives (D-Z). When a file is stored, it is broken up into smaller units and distributed through as many sectors as needed. The FAT table is simply an index that keeps track of which part of the file is stored in which sector.

FAT32 – This is supported by Windows 95 OSR2, Windows 98, Windows 2000 and WindowsXP . This system uses 32-bit numbers to identify clusters, which results in a maximum volume size of 2 terabytes (a terabytes is a gigabyte times 1024 or 1024 x1024 x 1024 x1024 bytes).

NTFS (Windows NT File System) - This is the native file system for Windows NT. It will support up to 16 exabytes (an exabyte is a terabyte times 1024). This file system also supports many advanced security features that are applicable to networking operations.

A File allocation table is an index that keeps track of two pieces of information. The first is the cluster ID that is a hex number that identifies the location of the cluster. The second piece of information is a status code.

When a file is saved - DOS starts at the beginning of the FAT, looks for the first empty cluster, and begins to write the data. Once the cluster is full, (512 bytes) it will enter the location of the next open cluster in the status field for that cluster and continue to write data. After all the data is written, it will use the code FFFF to indicate that this is the end of the file. This creates a map from cluster to cluster identifying all the clusters that contain a specific file.

One of the anomalies of this type of file storage is fragmentation. Fragmentation is the scattering of files all over the drive. As a file is written to sectors (clusters), it is placed in the first available location. As files are deleted an opening occurs, and they are filled with the next file to be saved. With time, this process will begin to scatter files all over the drive. This is an acceptable way to operate and causes no problems for the computer. However, it slows down the hard drive since it has to access two or more places to retrieve a file. It is possible for a file to be fragmented into hundreds of pieces; forcing the read/write heads to travel all over the hard drive to retrieve a single file. The elimination of fragmentation will improve the speed of the hard drive dramatically. The details of defragmentation are covered later in this chapter under “Maintaining a Hard Drive”.

Before we get back to partitioning, we must first understand the term clusters. As we have seen, the CHS values limit the maximum size of a hard drive to 504 MB. Since FAT uses a 16-bit binary number to identify locations, it can address 2l6 or 65,536 locations. To round this off, divide by 1024 bytes per kilobyte and you get 64K. Therefore, the size of a hard drive partition should be limited to 64K x 512 bytes/sector or 32 Megs. With this limitation, how are larger hard drives used?

There are two answers to this problem. The first method, used with earlier drives (under 100 Meg), was to use partitioning to break the drive up into multiple partitions, each being less than 32 Megs.

The second method is called Clustering. Clustering involves combining of a set of adjacent sectors into one logical unit. The logical unit being called a cluster or group of sectors. The number of sectors in each cluster is determined by the size of the partition, since there can never be more than 64k clusters. To determine the number of sectors in a partition, divide the number of bytes in the partition by 512 (bytes per sector). Then divide the number of sectors by 64,000 (maximum allowable clusters). The following table shows how this works.

The main reason for partitioning came about because of limits imposed by DOS. As drive technology exceeded the development of DOS, it was necessary to have a means of recognizing larger and larger drives. DOS could only recognize up to 32 Megabytes of a drive. With the release of DOS 3.3 and partitioning, a drive could be sub-divided into logical units small enough for DOS to handle. With time, and improvements in DOS, larger partitions could be recognized but the practice of naming the partitions stayed the same. As stated earlier, a DOS drive can have only two partitions, a primary and an extended.

A PRIMARY partition is a bootable partition. This is where the operating system is stored. When a computer boots from a hard drive, it looks for a special sector in the primary partition called the Boot Sector. This sector will tell it where the operating system is located and how to get started. The name of the Primary partition is “C:”

The EXTENDED partition is for the rest of the hard drive. The extended partition can be one drive or be logically partitioned into several drives. Each of the drives will be assigned a drive letter starting with the letter “D:” and progressing until drive letter “Z:” is created (remember, “A:” and “B:” are reserved for floppy disk drives).