OVERVIEW

In Chapter 3, we discussed the role of RAM, the computer's main memory.   RAM temporarily holds program instructions, data, and output until they are no longer needed by the computer.  As soon as the computer finishes with any given program and its data, those items are erased from RAM.  Consequently, if programs, data, and processing results are to be preserved for future use, a computer system needs more permanent storage.  Storage systems fill this role.

We begin this chapter with a discussion of characteristics common among storage systems.  Then we cover one of the most important kinds of storage systems in use today-those based on magnetic disks.   While this part of the chapter is primarily about floppy disk drives and hard disk drives, we also look at other common magnetic storage devices, such as Zip drives.  From there, we study optical discs, namely CDs and DVDs, and then turn to other types of storage systems, such as magnetic tape, flash memory, online storage, and smart cards.   The chapter concludes with a summary and comparison of the storage devices covered in the chapter 9.


PROPERTIES OF STORAGE SYSTEMS

Several important properties characterize storage systems.   In this section, we consider some of the most significant, including the two physical parts of a storage system, the non-volatility property of storage media, the ability to remove storage media from many storage devices, and the methods used to access and represent data.

Storage Devices and Media

There are two parts to any storage system: a storage device and a storage medium.  The storage medium is where the data is actually stored (such as a floppy disk or CD); a storage medium needs to be inside the appropriate storage device (such as a floppy drive or CD drive) to be read from or written to.  Often the storage device and medium are two separate pieces of hardware, though with some systems -- such as a hard drive -- the two parts are permanently sealed together to form one single piece of hardware.

Storage devices can be internal (located inside the system unit), external (plugged into an external port on the system unit), or remote (located on another computer, such as a network server).  Internal devices have the advantage of requiring no additional desk space and are often faster than their external counterparts.  External devices can be more easily used with multiple computers, or added to a PC that has no room left inside its system unit.  Remote devices are accessed over a network, such as a home network, a company network, or the Internet.  Regardless of how they are connected to the computer, letters of the alphabet and/or names are assigned to each storage device, so the devices can be identified when they need to be used (see Figure 4-1).







Non-Volatility

Storage media are non-volatile.  This means that when you shut off power to a storage device, the data stored on that device's storage medium will still be there when you turn the device back on.  This feature contrasts with RAM, which is volatile.  As discussed previously, data held in RAM is erased once it is no longer needed or the power to the computer is turned off.

Removable vs.  Fixed Media

In many storage systems, although the storage device is always connected to the computer, the storage medium used with that device can be inserted and removed.  These are called removable-media storage systems.   Floppy disks, CDs, and DVDs are examples of removable media.   On the other hand, fixed-media storage systems, such as most hard drive systems, seal the storage medium (such as the hard disk) inside the storage device (such as the hard drive) and users cannot remove it.

Fixed-media devices generally provide higher speed and better reliability at a lower cost than removable-media alternatives.  Removable-media devices have other advantages, however, including the following.

  • Unlimited storage capacity -- You can insert a new medium into the storage device to replace one that has become full.


  • Transportability -- You can easily share media between computers and people.


  • Backup -- You can make a duplicate copy of valuable data on a removable medium and store the copy away from the computer, for use if the original copy is destroyed.


  • Security -- Sensitive programs or data can be saved on removable media and stored in a secured area.



Virtually all desktop, notebook, and tablet PCs include both removable-media and fixed-media storage systems.


Random vs.  Sequential Access

When the computer system receives an instruction that requires data or programs located in storage, it must go to the designated location on the appropriate medium and retrieve the requested data or programs.  This procedure is referred to as access.  Two basic access methods are available: random and sequential.

Random access, also called direct access, means that data can be retrieved directly from any location on the medium, in any order.  With sequential access, however, the data can only be retrieved in the order in which it is physically stored on the medium.  Most of a PC's storage devices -- including hard disk drives, floppy disk drives, and CD/DVD drives -- are random access devices.  They work like audio CDs or movie DVDs—the user can jump directly to a particular selection or location, as needed.  One type of PC storage device that uses sequential access is a tape drive.  Computer tapes work like audio cassette tapes or videotapes -- to get to a specific location on the tape, you must play or fast-forward through all of the tape before it.  Media that allow random access are sometimes referred to as addressable media.   This means that the storage system can locate each piece of stored data or each program at a unique address, which is determined by the computer system.


Logical vs.  Physical Representation

Anything (such as a program, letter, digital photograph, or song) stored on a storage medium is referred to as a file.  Data files are also sometimes called documents.  When a document that was just created (such as a memo or letter in a word processing program) is saved, it is stored in a new file on the storage medium that the user designates.  During the storage process, the user is required to give the file a name, called a filename; that name is used when the user requests to see the document at a later time.

To keep files organized, related documents are often stored inside folders located on the storage medium.  For example, one folder might contain memos to business associates while another might hold a set of budgets for a specific project (see 
Figure 4-02 ).

To further organize files, you can create subfolders within a folder.   For instance, you might create a "Letters" folder that contains one subfolder for letters sent to friends and a second sub-folder for letters sent to potential employers.  In  Figure 4-02 ), both Budgets and Memos are subfolders inside the My Documents folder.

Although both the user and the computer use drive letters, folder names, and filenames to save and retrieve documents, the way a user perceives this process differs from the way a computer implements this process.   Typically, a user views how data is stored (what we have discussed so far in this section and what appears in the Windows Explorer screen in  Figure 4-02 ) using logical file representation.  That is, we view a document stored as one complete unit in a particular folder on a particular drive.  In contrast, the physical way data is stored and organized on the storage media (as viewed by the computer) is called physical file representation.  For example, the ABC Industries Proposal Memo file shown in  Figure 4-02 is logically located within the Memos folders inside the My Documents folder on the hard drive C, but the data in these folders could be physically stored in many different pieces scattered across that hard drive.  When this occurs, the computer keeps track of the various locations used and the logical representation (filename, folder names, and drive letter) that is being used to identify that file.  Fortunately, we don't have to be concerned with how files are physically stored on a disk, because the computer keeps track of that and retrieves files for us whenever we request them.


MAGNETIC DISK SYSTEMS

Speedy access to data, relatively low cost, and the ability to erase and rewrite data make magnetic disks the most widely used storage media on today's computers.  With magnetic storage systems, data is written by read/write heads magnetizing particles a certain way on a medium's surface.   The particles retain their magnetic orientation until the orientation is changed again, so files can be stored, rewritten to the disk, and erased, as needed.   Storing data on a magnetic disk is illustrated in Figure 4-3.   The most common type of magnetic disk is the hard disk; another common type of magnetic disk is the floppy disk.







Floppy Disks and Drives

Over the years, most PCs have been set up to use a floppy disk -- sometimes called a diskette or disk -- to accommodate removable storage needs.   Floppy disks are a removable medium and very inexpensive, so they are handy for such tasks as backing up small amounts of data, sending small files to others, and sharing data between two computers— such as a computer at home and one at school.  Floppy disks are written to and read by floppy disk drives (commonly called just floppy drives).   Because floppy drives are relatively slow and their capacity is relatively small compared to newer removable storage options, some manufacturers refer to the floppy drive as a legacy drive and are no longer automatically including one as part of their computer systems.   Instead of having an internal floppy drive, some PC buyers are opting for an external portable floppy drive that they can move from PC to PC, as needed.


Floppy Disk Characteristics


A floppy disk consists of a round piece of flexible plastic coated with a magnetizable substance.  The disk is protected by a square, rugged plastic cover lined with a soft material that wipes the disk clean as it spins (see Figure 4-4).






The surface of a floppy disk is organized into circular rings called tracks, and pie-shaped sectors.   On most PC systems, the smallest storage area on a disk is a cluster -- the part of a track that crosses a specific number (always two or more) of adjacent sectors (see  Figure 4-05 ).  Data is stored along the tracks of the disk; tracks, sectors, and clusters are numbered so that the computer can record where data is stored and can retrieve it at a later time.  To accomplish this, the PC keeps a directory -- called the file directory or file allocation table (FAT) -- of where each file is physically stored, its size, and what filename the user has assigned to it.  When the user requests a document (always by filename), the computer uses the FAT to retrieve it.   A cluster is the smallest addressable area on a disk; consequently, everything stored on a disk always takes up at least one cluster of space on the disk.  When a file takes up more than one cluster of space, each cluster contains directions pointing to the next cluster used, so the computer can retrieve all pieces of the file in the proper order when it is needed.

Most floppy disks in use today measure 3 1/2 inches in diameter (small enough to fit into a shirt pocket) and can store 1.44 MB of data, which is sufficient to store about 500 or so pages of double-spaced text created using a common word processing program.  Digital photographs, music files, or documents containing a lot of images usually require a higher capacity removable storage media—such as a CD, DVD, or high-capacity magnetic disk— which are discussed shortly.


Using Floppy Disks


To use a floppy disk, it must first be inserted into a floppy drive (with the label area facing up and closest to the user, as illustrated in Figure 4-6).





When it is completely inserted, the disk clicks into place, the metal shutter is moved aside to expose the surface of the disk, and the eject button on the front of the drive pops out.   Because the drive openings for some other types of removable disks (such as Zip disks, discussed shortly) are similar in size and appearance to a floppy drive opening, be careful when inserting a floppy disk to ensure that you are using the proper drive.   If the disk does not fit into or doesn't "click" into place inside the drive opening, you are likely inserting it into the wrong drive.

Before a floppy disk can be used, it must ^formatted to prepare it for use.  Most floppy disks sold today are already formatted for either IBM or Macintosh computers and, therefore, are ready to use.   Formatting a disk that already contains data erases everything on the disk.  Although in the past users would reformat floppy disks if they became unreliable, today users typically discard floppy disks when they become unreliable because of their very low cost.  The formatting process is sometimes used, however, to quickly erase a floppy disk for reuse.

When the floppy disk needs to be accessed, the drive begins to rotate the disk within its plastic cover.  The drive's read/write head can read (retrieve) data from or write (store) data onto the actual surface of the disk while the disk is spinning.  The read/write heads move in and out, allowing the read/write head access to all tracks on the disk.  While the disk is spinning, the drive's indicator light goes on—don't remove the floppy disk while this light is on.   To remove the disk, wait until the light goes off, and then you can press the eject button to remove the disk.


High-Capacity Removable Magnetic Disks and Drives

A number of higher-capacity removable magnetic storage media -- sometimes called superdiskettes -- have emerged in recent years, either as replacements for standard floppy disks or as supplemental storage solutions.   Although some of these systems have a large installed base and are still widely used at the present time, that may not be the case in the future as recordable optical disc technology improves.  High-capacity removable disks include Zip disks and SuperDisks.

Zip disks, introduced by Iomega Corporation in 1995, are high-capacity magnetic disks that can be read from and written to only with Zip drives.   Zip disks are similar in size and appearance to floppy disks (see Figure 4-7) but have a capacity of 100, 250, or 750 MB.&





Zip drives are downward compatible, meaning the higher-capacity Zip drives can read any Zip disks at their designated storage capacity or lower.  For instance, the Zip 750 drive can read all three sizes of Zip disks (although it can only write to Zip 250 and Zip 750 disks), while the Zip 100 drive can only be used with 100 MB Zip disks.  Zip disks cannot be used in a conventional floppy disk drive, and none of the Zip drives can read standard floppy disks.   Zip drives are most appropriate for users who need to back up large files or transfer large files between PCs or other users that have a Zip drive.   Because Zip drives were one of the first high-capacity removable storage solutions, they enjoy widespread use.

SuperDisk drives, originally made by Imation and more technically called laser servo (LS) drives, are similar to Zip drives in that they accept disks -- called SuperDisks or LS-120 or LS-240 disks, depending on their capacity -- with larger capacities (120 or 240 MB) than standard floppy disks.   While SuperDisk drives are slower than Zip drives, they have the advantage of being able to read from and write to standard floppy disks, in addition to SuperDisks.  A regular floppy drive, however, cannot read a SuperDisk.

SuperDisk drives are no longer being manufactured by Imation, although other LS drives are available.  It is expected that other types of high-capacity media -- such as optical discs and flash memory media -- will eventually replace both the conventional floppy disk and high-capacity magnetic disks.


Hard Disk Drives

With the exception of computers designed to use only network storage devices (such as network computers and some Internet appliances), virtually all PCs come with a hard disk drive (commonly referred to as hard drive) that is used to store most programs and data used with that PC.  Hard drives are typically located inside the system unit and are not designed to be removed, unless they need to be repaired or replaced.  In common practice, the terms hard disk, hard disk drive, hard disk system, and hard drive are used interchangeably.


Hard Drive Characteristics


Similar to floppy drives, hard drives store data magnetically; their disks are organized into tracks, sectors, and clusters; and they use read/write heads to store and retrieve data.  However, the hard disks used with a hard drive are made out of metal and are permanently sealed (along with the read/write heads and access mechanisms) inside the hard drive.  One drive may contain a stack of several hard disks, as shown in
Figure 4-08 .   Hard drives are typically fixed-media systems in which the storage media (the hard disks) are not removable from the storage device (the hard drive); one exception is a hard drive that uses a removable hard disk cartridge, as discussed later in this chapter.

Hard drives are faster than removable-media systems and can store a great deal more data.  The capacity of a typical internal hard drive for today's desktop PCs ranges from 40 to 300 GB.   Internal hard drives for notebook computers are also getting larger -- up to 80 GB.  Most hard drives for desktop PCs use 3 1/2-inch hard drives, although a switch to 2 1/2-inch hard drives is expected in the near future; most notebook computers use a 2 1/2 inch hard drive.  

Even smaller hard disk systems are becoming available for systems requiring tiny drives, such as the 1-inch Microdrive developed by IBM, who is now partnered with Hitachi for their hard drive systems.  Hitachi recently announced that a 4 GB version of the Microdrive will be available by the end of 2003.  The increased capacity is due to a new storage technology developed by IBM called Pixie Dust, which sandwiches three atoms of the precious metal ruthenium between two magnetic layers.   This technology enables data to be stored at much higher densities on magnetic media than previously possible.

Like floppy disks, hard disk surfaces are divided into tracks, sectors, and clusters when formatted, but include many more of each.   A new hard drive is typically formatted for use at the factory before it is sold, so it is ready for software and data as soon as it is installed.   Because reformatting a disk erases everything on the disk, hard drives are rarely reformatted.  This task is only performed if errors are preventing the hard drive from operating properly and there is no other option.

In addition to tracks, sectors, and clusters, hard drives use the concept of a cylinder.  A cylinder is the collection of one particular track on each disk surface, such as the first track or the tenth track on each disk surface.  In other words, it's the area on all of the hard disks inside the hard drive that can be accessed without moving the read/write access mechanism, once it has been moved to the proper position.  For example, the four-disk system in  Figure 4-09   contains eight possible recording surfaces (using both sides of each disk), so a cylinder on that system would consist of eight tracks, such as track 13 on all eight surfaces.  Hard drives are commonly organized into anywhere from a few hundred to a few thousand cylinders.   The number of tracks on a single disk is equal to the number of cylinders in the disk system.

Most hard drives are hermetically sealed units.  This precaution keeps the disk surfaces completely free of contamination, enables the disks to spin faster, and limits causes of operational problems.   Hard disks typically spin between 5,400 and 15,000 revolutions per minute (rpm), depending on the type and size of the drive.   In addition to spinning faster than most other types of storage systems, the hard disk constantly rotates when your computer is turned on instead of only rotating when it needs to be accessed.  This feature eliminates the delay of waiting for the drive to come up to the correct speed.  (Most PCs can be set up to go to sleep and turn off the hard drive after a specified period of inactivity to save power; in this case, touching the keyboard or mouse starts the hard disks spinning again.)

To retrieve or store data, most hard drives have at least one read/write head for each recording surface.  These heads are mounted on an access mechanism, similar to a floppy disk, this mechanism moves the heads in and out among the tracks together.  It positions all the heads on the cylinder containing the track from which data is to be read or to which data is to be written.  It is important to realize that a hard drive's read/write heads never touch the surface of the hard disk at any time, even during reading and writing.   If the read/write heads do touch the surface -- such as if the PC is bumped while the hard drive is spinning or a foreign object gets onto the surface of the disk, a head crash occurs, which may do permanent damage to the hard drive.  Because the heads are located extremely close to the surface of the disk -- usually less than a millionth of an inch above the surface -- the presence of a foreign object the width of a human hair or even a smoke particle (about 2,500 and 100 millionths of an inch, respectively) on a hard disk's surface is like placing a huge boulder on a road and then trying to drive over it with your car (see  Figure 4-10 ).   One never knows when a hard drive will crash -- there may be no warning whatsoever -- and this is a good reason for keeping the drive backed up regularly.  Backing up a computer system is discussed in more detail in Chapter 6.  When hard drives containing critical data become damaged, data recovery firms may be able to help out, as discussed in the Inside the Industry box.

In order for a hard drive to read or write data, the following three events must be carded out, all of which may add time to the total disk access time.


  1. Move the read/write heads to the cylinder that contains (or will store) the desired data—called seek time.


  2. Rotate the disks into the proper position so that the read/write heads are located over the part of the cylinder to be used—called rotational delay.


  3. Read the data from the disk and transfer it to memory or transfer the data to be written to the disk from memory and then store it on the disk -- called data movement time.



Typical hard disk access times are from 10 to 20 milliseconds.   To minimize disk access time, drives usually store related data on the same cylinder.  This strategy sharply reduces the seek-time component and improves the overall access time.

In addition to being used with computers, hard drives are increasingly being incorporated into consumer products, such as digital video recorders (DVRs) like TiVo and game boxes like Xbox and PlayStation.  Although growth in the computer storage industry has been slowing, demand for storage products for consumer applications is on the rise.


Partitioning and File Systems


Partitioning a hard drive enables you to logically divide the physical capacity of a single drive into separate areas called partitions.   You can then treat each of the partitions as an independent disk drive, such as a C drive and a D drive, although they are physically still one drive.  At least one partition is created when a hard drive is first formatted; you can change the number and sizes of the partitions at a later time, although this action usually destroys any data in the partitions being changed.  Consequently, you should back up your data located on that drive to another storage medium before you repartition a hard drive, and then copy the data back onto the repartitioned hard drive.  Some operating systems have a limit to the number of partitions that can be used.

Because older operating systems could only address hard drives up to 512 MB, hard drives larger than that limit had to use multiple partitions.   Most newer operating systems allow larger drives, but partitioning a large drive can make it function more efficiently.  This is because operating systems typically use a larger cluster size with a larger hard drive.  When a large cluster size is used, disk space is often wasted because even tiny files have to use up one entire cluster of storage space.  When a hard drive is partitioned, each logical drive uses a smaller cluster size, since each logical drive is smaller than the original drive.  Windows computers using the FAT32 file system are much more efficient than those using the original FAT system since FAT32 systems allow cluster sizes to be as small as 4 KB each, which cuts down on wasted storage space.  Windows NT and Windows XP computers have the option of using the NTFS file system, which can address much larger drives than either FAT or FAT32.

Another reason for partitioning a hard drive is to be able to use two different operating systems on the same hard drive -- such as Windows and Linux.  You can then decide which operating system you will run each time you turn on your computer.  Creating the appearance of having separate hard drives for file management, multiple users, or other purposes is another common reason for partitioning a hard drive.   Some users choose to install their programs on one hard drive (usually C) and store their data on a second drive (such as D).   This system of using separate logical drives for data and programs makes locating data files easier, as well as enables users to back up all data files simply by backing up the entire data drive (program files aren't typically backed up as frequently as data files, if at all).   Operating systems and backing up data are discussed in more detail in Chapter 6.


Disk Cache


A cache (pronounced cash) is a place to store something temporarily.   For instance, in Chapter 3 we learned that cache memory is a group of very fast memory chips located on or near the CPU that are used to store the most frequently and recently used data and instructions.  Because transferring that data and instructions from cache memory to the CPU is much faster than transferring them from RAM or the hard drive, cache memory typically results in faster processing.   Disk caching is similar in concept—it is a strategy for speeding up system performance by storing data or programs that might be needed soon in a designated area of RAM to avoid having to retrieve them from the hard drive when they are requested.  Since retrieving data from RAM is much faster than from the hard drive, disk caching can speed up performance.

The location in RAM where disk caching takes place is called the disk cache.  When a hard drive uses disk caching (as most do today), any time the hard drive is accessed the computer copies the requested program and data, as well as extra programs or data located in neighboring areas of the hard drive (such as the entire track or cylinder), to the disk cache.  The theory behind disk caching assumes that neighboring data will likely have to be read soon anyway (research indicates that there is an 80 to 90% chance the next request will be for data located adjacent to the data last read), so the computer can reduce the number of times the hard drive is accessed by copying that data into RAM early.   When the next data is requested, the computer system checks the disk-cache area first, to see if the data it needs is already there.   If it is, the data is retrieved for processing; if not, the computer retrieves the requested data from the disk (see 
Figure 4-11 ).   Disk caching saves not only time but also wear and tear on the hard drive.  In portable computers, it can also extend battery life.


Hard Drive Standards


Hard drives connect, or interface, with a computer using one of several different standards.  These standards determine performance characteristics, such as the density with which data can be packed onto the disk, the speed of disk access, how large the disk can be, and the way the disk drive interfaces with other hardware.  Some of the most common interfaces are discussed next.

With EIDE, for enhanced integrated drive electronics, the hard drive controller -- the chip that controls the flow of data to and from the hard drive -- is built into the drive.  SCSI, for small computer system interface and pronounced "skuzzy," hard drive controller chips are either attached directly to the motherboard or are located on a SCSI interface card to which the drive is connected.  Both EIDI and SCSI are very fast and can support multiple hard drives.  EIDE has a variety of different specifications, such as ATA, Fast ATA, Fast IDE, or ATA-2, ATA/100, and serial ATA.  EIDE drives are typically less expensive than SCSI drives; consequently, EIDI drives are found more often in desktop PCs.  SCSI is usually faster for server operations with multiple users.  In addition to being used with hard disk drives, SCSI interfaces can also be used to connect some scanners, CD drives, and DVD drives.

Fibre Channel is a newer storage standard that is expected to become widely used with network storage systems, as well as in other high-capacity business storage applications.  Fibre Channel storage devices connect to the host computer using a special Fibre Channel interface card and have the advantage of reliability, flexibility, and very fast data delivery -- up to two gigabits per second.  Because it is more expensive than other standards and is geared for long-distance, high-bandwidth applications.   Fibre Channel is not expected to be widely used with PCs and low-end servers, at least not in the near future.  It is, however, expected to eventually replace SCSI for high-end storage systems.

Some external hard drives today follow none of these standards; instead they connect to the PC using USB or Fire Wire standards through a USB or FireWire port.


Portable Hard Drive Systems


While most hard drive systems are designed to be internal devices permanently located inside the system unit, portable hard drives are available.   Portable hard drives fall into two basic categories: those in which the entire drive is transported from one location to another, and those in which a cartridge containing the hard disk is removed from the hard drive and transported.  When the entire drive is portable -- essentially an external hard drive -- the drive is typically attached to the PC through a USB, FireWire, or PC card port (see  Figure 4-12 ).

Common capacities for external hard drives are 20 to 160 GB.   Portable hard drive systems using removable hard disk cartridges use a hard drive that remains attached to the PC in conjunction with hard disk cartridges that can be inserted into and removed from the drive, similar to a floppy disk system.  Also similar to a floppy drive, the hard drives used with hard disk cartridges can be internal or external, but external devices are more widely used.  Hard disk cartridges can usually store about 20 GB, although larger storage capacities are expected in the near future.  Most removable hard disk cartridges are proprietary, so they can only be used with their respective drives.

Both types of portable hard drive systems are useful for storing and backing up very large files, transporting large files from one PC to another, and for exceptionally secure facilities -- such as government and research labs -- that require all hard drives to be locked up when not in use.  They are also commonly used for complete system backups.  Although their portability has its advantages, portable hard drives generally perform more slowly than conventional fixed internal hard drives.


Storage Systems for Large Computer Systems and Networks


Hard drive systems for large computer systems (such as those containing mainframe computers and midrange servers) implement many of the same standards and principles as PC-based hard drives, but on a much larger scale.  Instead of finding a single hard drive installed within the system unit, you are most likely to find a storage server -- a separate piece of hardware containing multiple high-speed hard drives -- connected to the computer system.  Large storage servers, such as the one shown in  Figure 4-13 , contain racks of hard drives capable of storing a total of 30 TB or more.  These types of storage systems -- also referred to as enterprise storage systems -- usually use fast Fibre Channel connections.   In addition to being used as stand-alone storage for large computer systems, storage servers may also be used in network attached storage (NAS), storage area network (SAN), and RAID storage systems.


Network Attached Storage (NAS) and Storage Area Networks (SANs)


Storage servers are increasingly being used to provide storage for computer networks.  With the huge amounts of data that many companies need to manage and store today -- for instance, Yahoo! needs to store on an on-going basis more than a petabyte of data generated by Yahoo! e-mail users -- network-based storage has become increasingly important.

One possibility is the network attached storage (NAS) device.   NAS devices are high-performance storage servers that are individually connected to a network to provide storage for the computers on that network.  Storage area networks (SANs) also provide storage for a network, but consist of a separate network of hard drives or other storage devices. That storage area network is, in turn, attached to the main network.   The primary difference between NAS and SANs is whether the storage devices act as individual network nodes, just like PCs, printers, and other devices on the network (NAS), or whether they are located in a completely separate network of storage devices that is accessible by the main network (SAN).

However, in terms of functionality, the distinction between NAS and SANs is blurring, since they both provide storage services to the network.  

Both NAS and SAN systems are scalable, so new devices can be added as more storage is needed and devices can be added or removed without disrupting the network.


RAID


RAID (redundant arrays of independent disks) is a method of storing data on two or more hard drives that work in combination to do the job of a larger drive.  Although RAID can be' used to increase performance, it is most often used to protect critical data on a storage server.   Because RAID usually involves recording redundant (duplicate) copies of stored data, the copies can be used, when necessary, to reconstruct lost data.  This helps to increase the fault tolerance—the ability to recover from an unexpected hardware or software failure, such as a system crash—of a storage system.

There are six different RAID designs or levels (0 to 5) that use different combinations of RAID techniques.  For example RAID level 0 uses disk striping, which spreads files over several disk drives (see the leftmost part of
Figure 4-14 ).   Although striping improves performance, since multiple drives can be accessed at one time to store or retrieve data, it doesn't provide fault tolerance.

Another common RAID technique is disk mirroring, in which data is written to two duplicate drives simultaneously (see the rightmost part of  Figure 4-14 ).  The objective of disk mirroring is to increase fault tolerance -- if one of the disk drives fails, the system can instantly switch to the other drive without any loss of data or service.  RAID level 1 uses disk mirroring.   Levels beyond level 1 use some combination of disk striping and disk mirroring, with different types of error correction provisions.

Because using RAID is significantly more expensive than just using a traditional hard ( drive storage system, it has been reserved for use with network and Internet servers.  However, recently RAID has become more popular with PC users looking for increased performance.  One recent test by PC World magazine showed that two RAID-connected drives completed some tasks in 40% less time than one drive of the same type.  To implement RAID on a desktop PC, a RAID expansion card must be used.


OPTICAL DISC SYSTEMS

Optical discs (such as CDs and DVDs) store data optically -- using laser beams -- instead of magnetically, like floppy and hard disks.  Lasers can write and read data at densities much higher than magnetic technology, so the storage capacity of optical discs is much higher than magnetic disks of the same physical size -- usually from 650 MB on up.

Optical discs are made out of plastic with a reflective metallic or otherwise light-sensitive coating.  Data can be stored on one or both sides of an optical disc, depending on the disc.  Most optical discs are 4 1/2 inches in diameter, although smaller discs are sometimes used.  To keep data organized, optical discs are divided into tracks and sectors like magnetic disks, but use a single grooved spiral track beginning at the center of the disc (see 
Figure 4-15 ), instead of a series of concentric tracks.   Because lasers can be very precise, the track can be quite narrow and the spiral can be very tight -- when measured from end to end, the total length of the track on a typical CD is over 3 miles.

The number of sectors used varies depending on the size and type of disc, but a standard 650 MB CD has over 330,000 sectors.  Because the track starts at the center of the disc, optical discs can be made into a variety of sizes and shapes -- such as a heart, triangle, custom shape, or the hockey-rink shape commonly used with business card CDs -- the track just stops when it reaches the outer edge of the disc.  Standard shapes are molded and less expensive; custom shapes—such as those that match a key product or service being sold (such as a soda can, musical instrument, saw blade, candy bar, or house) -- are custom cut and are more costly.  The practice of using optical discs to replace ordinary objects, such as the business card discs shown in  Figure 4-15 , is becoming more common.  For a closer look at business card CDs, see the How it Works box (on page 151 of your textbook).

CD and DVD discs are read by CD and DVD drives.  The speed of a CD or DVD drive is rated as a number followed by the "x" symbol to indicate how fast the drive is compared to the first version of that drive.  For instance, a 52x CD drive is 52 times faster than the original CD drive, and a 4x DVD drive is four times faster than the original DVD drive.  Most optical discs have a title and other text printed only on one side and are inserted into the drive with the printed side facing up (the data is stored on the bottom, non-printed side of the disc).  When inserting a CD or DVD, be careful not to get dirt, fingerprints, scratches, or anything else that might hinder light reflectivity on the disc's reflective recording surface.  The advantages of CDs and DVDs include their large capacity -- typically 650 or 700 MB per CD and 4.7 GB per DVD, although double-sided DVDs currently hold 9.4 GB and are expected eventually to reach 17 GB -- and their small size.  (Because DVD technology uses smaller pits and the tracks are closer together, DVDs hold more data than CDs.)   Another advantage is that optical discs last longer and are more durable than magnetic media, although the discs should be handled carefully and stored in a protective jewel case when not in use to prevent scratches and fingerprints from getting on the disc.   Optical discs are the standard today for software delivery; they are also commonly used for storing and transporting high-capacity music and video files.

There are a variety of types of CDs and DVDs.  Some of the most important types and characteristics of CDs and DVDs are discussed next.

Read-Only Discs: CD-ROM and DVD-ROM Discs

CD-ROM (compact disc read-only memory) discs were the first optical discs of wide acceptance.  Because they are read-only, the data on CD-ROM discs cannot be erased, changed, or added to.  Data on a CD-ROM is stored by burning tiny depressions (called pits) into the disc's surface with a high-intensity laser beam; the parts of the disc that aren't changed are called lands.  The disc is read by a lower-intensity laser beam inside the CD-ROM drive; based on the reflection of light from the disc as it hits the pits and lands, the Is and Os can be determined (see  Figure 4-16 ).   Because the storage process permanently alters the surface of the CD-ROM, the data cannot be erased and no data can be added to the disc.

DVD-ROM (digital versatile disc read-only memory) discs are similar to CD-ROM discs, but they are newer and have a higher storage capacity.  While CD-ROM discs typically hold 650 MB, DVD-ROMs can contain from 4.7 GB to 17 GB, depending on the number of recording layers and disc sides being used.   The DVD was initially developed to store the full contents of a standard two-hour movie, but is now also used for prerecorded music, videos, and software.   DVD-ROM discs are designed to be read by a DVD-ROM drive.

CD-ROM drives can usually play audio CDs, in addition to data CDs.  DVD-ROM drives can typically play data and audio CDs, data CDs, DVD-ROM discs, and DVD movies.

For a look at an emerging issue—copy protection for CDs and other digital media -- see the Trend box (on page 155 of yourtextbook).


Recordable Discs: CD-R, DVD-R, and DVD+R Discs

Recordable discs can be written to, but the discs cannot be erased and reused.   Recordable CDs are referred to as CD-R discs; recordable DVDs are called DVD-R or DVD+R discs, depending on the standard being used (different optical disc and drive manufacturers support different standards).  CD-R, DVD-R, and DVD+R discs are recorded in CD-R, DVD-R, and DVD+R drives, respectively; CD-R discs can be read by most types of CD and DVD drives, and DVD-R or DVD+R discs can be read by most DVD drives.  Recordable CDs are commonly used for backing up files, sending large files to others, and creating custom music CDs from MP3 files legally downloaded from the Internet or from songs on CDs the user owns.  DVD-Rs can be used for similar purposes when more storage space than is available on a CD-R disc is needed, as well as for storing home movies and other video applications since video requires a tremendous amount of storage space.  As shown in Figure 4-17, recordable CDs and DVDs look very similar to their read-only counterparts.  Standard-sized 4 1/2-inch CD-R discs hold 700 MB, 3-inch mini CD-R discs hold about 200 MB, business-card-sized CD-R discs hold 50 MB, and DVD-R or DVD+R discs can store 4.7 GB per side.







Storing data on a recordable disc is similar to the concept illustrated in Figure 4-16, but the discs contain a light-sensitive dye or chemical embedded beneath layers of protective plastic instead of a reflective metallic layer.  The recording laser inside the CD-R or DVD-R drive is less powerful than the one used to create read-only discs, but still makes permanent marks on the disc to represent Os and Is.  The process of recording data onto an optical disc is called burning.  To bum a CD-R or DVD-R disc, special software is needed.  Many commercial programs are available, and burning capabilities are also included in many recent operating systems, such as Windows XP.


Rewritable Discs: CD-RW, DVD-RW, DVD_RW, DVD-RAM, and Blue Laser Discs

The newer rewritable discs can be recorded on, erased, and overwritten just like a magnetic disk.  The most common types of rewritable optical media are CD-RW, DVD-RW, and DVD+RW discs.  CD-RW discs are written to using a CD-RW drive and can be read by most CD and DVD drives.   DVD-RW discs and DVD+RW discs are recorded using a DVD-RW drive or DVD+RW drive, respectively, and can be read by most DVD drives.  An additional rewritable DVD format is DVD-RAM, which requires the DVD disc to be located inside a cartridge (see Figure 4-17) in order for the disc to be used.  

The newest recordable and rewritable technologies use blue lasers instead of infrared (CDs) or red (DVDs) lasers to store data more compactly on the disc.  Blue laser discs based on this technology—developed by Sony and called Blu-ray—can hold 23.3 GB per disc.  A similar, but competing, format developed by Toshiba and NEC is called the Advanced Optical Disc format and is capable of storing up to 36 GB of data on a dual-layer disc.  In contrast, CD-RW discs hold 700 MB, DVD+RW and DVD-RW discs hold 4.7 GB per side, and DVD-RAM discs typically hold between 2.6 and 9.4 GB, depending on the speed of the disc and the number of sides used.

To record and erase rewritable optical discs, phase-change technology is most often used.  With this technology, the recordable CD or DVD disc is coated with a special metal alloy compound that has two different appearances once it's been heated and then cooled, depending on the temperature reached during the heating process.  With one temperature, the surface is reflective; with a higher temperature, it's not.  Before any data is written to a disc, the disc is completely reflective.  To record onto the disc, pits are burned into the surface by creating non-reflective areas; unburned areas (lands) remain reflective.   Just as with other CDs and DVDs, these pits and lands are interpreted as 1 s and 0s when the disc is read.  To erase the disc, the appropriate temperature is used to change the areas to be erased back to their original reflective state.

It is important to realize that the DVD industry has not yet reached a single standard, so there are competing formats that are not necessarily compatible with each other.  Luckily, many DVD drive manufacturers are introducing new drives that support more than one standard, such as one drive from Sony that is compatible with DVD+RW, DVD+R, DVD-RW, DVD-R, CD-RW, and CD-R discs.   Because of the format controversy, recordable and rewritable DVD technology has taken off more slowly than originally anticipated.  However, just as CD-R and CD-RW drives have virtually replaced CD-ROM drives, it is expected that, eventually, rewritable DVD drives will replace CD drives.  About 4 million recordable DVD drives were in use in 2002; that number is expected to exceed 37 million by 2005.



OTHER TYPES OF STORAGE SYSTEMS

Other types of storage systems include magneto-optical discs, flash memory media, magnetic tape, remote storage, and smart cards.  A possibility for the future is holographic storage.

Magneto-Optical Discs

There are a few types of storage systems that use a combination of magnetic and optical technology—the magneto-optical (M-0) disc is one of the most common.  Magneto-optical drives read special M-0 discs, which are usually optical discs inside a rectangular cartridge, similar in appearance to the DVD-RAM disc shown in Figure 4-17.   M-0 discs are available in both 3^-inch and 514-inch sizes and can store up to 9.1 GB per disk.

Flash Memory Media

Unlike magnetic and optical storage systems whose drives have moving parts, flash memory media consists of chips and other circuitry that don't move within the drive as it's being accessed -— called a solid-state storage system.   Because flash memory devices and media are very small, use much less power than conventional drives, and are resistant to shock and vibration since they have no moving parts, they are especially appropriate for use with digital cameras, digital music players, handheld PCs, notebook computers, smart phones, and other types of portable devices (see  Figure 4-18 ).

Today, flash memory is found in the form of rewritable sticks, cards, or drives.  Some computers and many mobile devices contain at least one flash memory port; when an appropriate port is not built into the device, a flash memory card reader or adapter can be used.  Typically, flash memory media is purchased blank, but some flash-memory-card-based software is available, such as games, encyclopedias, language translators and more.   Although flash memory media is relatively expensive per gigabyte, its convenience and universal acceptance makes it an appealing storage option for many purposes.


Flash Memory Sticks


Flash memory sticks were introduced by Sony initially for use with their digital music players.  Since then, however, flash memory use has expanded to digital cameras, PCs, printers, and other applications.  Some newer computers come with a memory stick port built in; if not, an external reader can be used.  Flash memory sticks are about the size of a stick of gum (see  Figure 4-18 ) and hold from 32 MB to 1 GB each; a 128 MB card cost about $50.

Flash Memory Cards


Flash memory cards are the primary removable storage media for handheld PCs, digital cameras, portable entertainment products, and mobile devices.   Flash memory cards come in a variety of formats and are typically inserted into a flash memory port located on the PC or device.   Some flash memory ports accept only one type of memory media; others can read from and write to several types.  Just as with flash memory sticks, external card readers are available that read one or more memory media formats.  Typically, these readers' plug into a PC card slot or USE port.  The main types of flash memory cards are listed next; some are illustrated in  Figure 4-18
and Figure 4-19 .   Of the following flash media formats, CompactFlash and Secure Digital (SD) are the most widely used at the present time.

  • CompactFlash cards are widely used with digital cameras, handheld and portable PCs, digital music players, printers, and other portable devices.  CompactFlash card capacity ranges from 32 MB to 4 GB.


  • Secure Digital (SD) cards are one of the most commonly used cards for handheld PC and smart phone storage.  Also used with digital cameras, digital music players, and digital camcorders, the capacity of the stamp-sized SD card ranges from 32 MB to 1 GB.


  • MiniSD cards are a smaller version of Secure Digital cards that became available in 2003.  Geared primarily for use with mobile and smart phones, capacity currently ranges from 16 MB to 256 MB.


  • MultiMedia cards (MMC) are most commonly used with digital music players and digital camcorders.  Because they are the same size as SD cards, some devices can use these two types of cards interchangeably.  Capacity ranges from 16 MB to 128 MB.


  • SmartMedia cards are frequently used with digital cameras, although other devices may accept them as well.  Larger in physical size than the other types of flash media, SmartMedia cards can hold from 8 MB to 128 MB of data.


  • xD cards (sometimes called xD Picture cards) are one of the newest formats and are designed primarily for digital camera use; capacity of these cards range from 32 MB to 512 MB.




Flash Memory Drives


Flash memory drives (sometimes also called USB mini-drives, removable flash drives, or key drives}, are self-contained storage systems that consist of flash memory media and the drive hardware necessary to write to and read from that media.  Because they have no moving parts, flash memory drives are much more resistant to shock and vibration than conventional drives and are therefore appropriate for harsh environments, as well as for transporting data from one place to another in a briefcase or pocket.  They also have a longer expected life than removable magnetic media.  Although larger flash memory drives exist to replace conventional hard drives in situations where the PC will be subjected to jarring movements, strong vibrations, or other unstable conditions that might harm a conventional hard drive, most flash memory drives are designed to be very portable and so are small enough to fit in a pocket or be carried on a keychain (see Figure 4-20).





To read from or write to a flash memory drive, it is plugged into a PC's USB port and then it is automatically assigned a drive letter by the computer, like any other type of drive attached to a PC.   Files can be read from or written to the drive until it is unplugged from the USB port.  Some flash memory drives have their flash memory media permanently sealed inside; others use standard flash memory cards and can be opened to replace the drive with a new memory card when the original is full or if it becomes damaged.  Flash memory drives today are available in capacities from 32 MB to 2 GB.


Magnetic Tape Systems

Magnetic tape consists of plastic tape coated with a magnetizable substance that is polarized to represent the bits and bytes of digital data, similar to magnetic disks.  Tape was once a prominent storage medium for computer systems, but because of its sequential-access property it has since been replaced by magnetic disks, optical discs, and flash memory media for day-to-day use.  It is still used today for backup and archival purposes because it is a very inexpensive medium.

Most computer tapes today are in the form of cartridge tapes (similar to a video or audio tape), instead of the older reel-to-reel format.   Tapes are read by tape drives, which can be either an internal or external piece of hardware (an internal tape drive is shown in Figure 4-21).





Tape drives contain one or more read/write heads over which the tape passes to allow the drive to read or write data.  Just as with other magnetic storage technologies, the Is and Os stored on magnetic tape are represented magnetically.

There are a variety of sizes and formats of cartridge tapes, such as digital audio tape (DAT), quarter-inch-cassette (QIC), Travan, digital linear tape (DLT), advanced intelligent tape (AIT), Super advanced intelligent tape (S-AIT), and linear tape-open (LTO).  Sizes and formats of tapes are not generally interchangeable, but since magnetic tapes are most often used for backup with a specific tape drive, this incompatibility is usually not a problem.  A typical tape cartridge holds between 4 GB to 240 GB, although some can hold up to 1 TB.  When a larger capacity is needed, some tape drives are designed to be used with multiple tape cartridges, increasing the potential storage capacity to well over 2 TB.


Remote Storage Systems

Remote storage refers to using a storage device that is not connected directly to your PC system; instead, the device is accessed through a local network or the Internet.  Using remote storage devices and media works similarly to using local storage (the storage devices and media that are directly attached to your PC); you just need to select the appropriate remote storage device (typically a hard drive attached to a network server), and then you can store data on or retrieve data from it.  

When the remote device is accessed through a local network, it is sometimes referred to as network storage; the term online storage most commonly refers to storage accessed via the Internet.  Individuals and businesses can use online storage Web sites to transfer files between two computers, to share files with others, and for backup in case of a fire or other disaster.  For some Internet appliances, network computers, and mobile communications devices with little or no local storage capabilities, online storage is especially important.

It is becoming increasingly common for individuals to want to share files -- particular digital photographs -- through the Internet.  Some Web sites dedicated to online storage offer the service for free to individuals; others charge a small fee, such as $5 per month for up to 75 MB of storage space (business accounts typically cost more).  

Although some sites allow access to anyone, most online storage sites are set up to have file access only by password to limit access to just yourself and anyone else you give your password to.  Some sites allow you to e-mail links to others to download specific files in your online collection without having to supply a password.  Other online storage sites contain an automatic back up option in which the files in designated folders on your PC are uploaded to your online account at regular specified intervals.   Two examples of online storage sites are shown in 
Figure 4-22 .


Smart Cards

A smart card is a credit-card-sized piece of plastic that contains some computer circuitry, typically a processor, memory, and storage (see  Figure 4-23 ).   Although the storage capacity of a smart card is fairly small -- usually from a few kilobytes to a few megabytes -- it can be used to hold specific pieces of information that may need to be updated periodically.  

Typically, smart cards are used for payment or identification purposes.   For example, a smart card can store a prepaid amount of digital cash for purchases using a smart card-enabled vending machine or PC; loyalty system information (frequent flyer points, for example); identification data for accessing facilities or computer networks; or an individual's medical history and insurance information for fast treatment and hospital admission in an emergency.

Smart cards are also increasingly being used for national ID cards and student ID cards (see the Campus Close-Up box on page 162 of your textbook).   Although these applications have used conventional magnetic stripe technology in the past, the microprocessor in a smart card protects the integrity of the data on the card (in contrast with the straight data storage capabilities of a flash memory card or CD, for example), and data stored in the card's memory can be added or modified as needed.   For an even higher level of security, some smart cards today contain biometric data—such as a fingerprint—to ensure the authenticity of the user (biometrics is discussed in more detail in Chapter 8).   Many debit and credit cards today are also smart cards.

To use a smart card, it must be inserted into a card reader built into or attached to a PC, vending machine, or other item.  Some keyboards now have a built-in smart card reader to facilitate secure e-commerce applications, such as online shopping.  Once a smart card has been accepted, the transaction -- such as making a purchase or unlocking a door -- can be completed.  While many smart card readers require direct contact between the smart card and the reader, contactless smart card systems using wireless technology allow the card to be read when it is within a particular distance of the reader without physical contact.   E-commerce is covered in more detail in Chapter 11.

One new smart card application is a combination smart card/magnetic disk.   This emerging product, such as the StorCard, has a flexible magnetic disk housed inside a tiny cavity created between the top and bottom layers of the smart card.  A proprietary reader is able to access the disk via a shutter, similar to a floppy disk, to read from and write to the card.  Current capacity is about 100 MB.  The smart card capabilities of this product enable the data on the card to be encrypted or otherwise protected using smart card technology.


Holographic Storage

Storing information in three dimensions is far from a new idea.   DVDs use multiple layers to store more data on the same size disc as a CD, and 3D memory chips are in the works.  One very promising technology for 3D storage systems being researched by such companies as IBM, Lucent Technologies, and Imation is holographic storage.

Holographic storage systems use multiple laser beams to store data in three dimensions, in order to store more data on the disc.  Data is stored in a "page" format, in which all data on each page is stored and retrieved together.  Because a million or more bits can be located on each page and thousands of pages can be stored in material no larger than a small coin, holographic systems offer the possibility of compact storage media holding many terabytes of information.  The additional advantages of no moving parts and simultaneous access of all information stored in a page give this technology the potential for very rapid access.   Some predictions include a data-throughput rate of at least 1 billion bps (bits per second).

Potential applications for holographic data storage systems include high-speed digital libraries and image processing for medical, video, and military purposes -- applications for which data needs to be stored or retrieved quickly in large quantities, but rarely changed.  Rewritable holographic storage is an expected improvement in the future.


COMPARING STORAGE ALTERNATIVES

Storage alternatives are often compared by weighing a number of product characteristics and cost factors.  Some of these product characteristics include speed, compatibility, storage capacity, convenience, and the portability of the media.  

Keep in mind that each storage alternative normally involves trade-offs.   For instance, most systems with removable media are slower than those with fixed media, and external drives are typically slower than internal ones.   Although cost is a factor when comparing similar devices, it is often not the most compelling reason to choose a particular technology.   For instance, although the flash memory drives are very expensive per GB, many users find them essential for transferring files between work and home or for taking presentations on the road.

For drives that use a USB interface, the type of USE port is also significant.   For instance, a typical flash memory drive designed for the original USE 1.1 port transfers data at up to 1.5 MB per second; USB 2.0 flash memory drives are about 40 times faster.

With so many different storage alternatives available, it's a good idea to research which devices and media are most appropriate for your personal situation.   In general, most users today need a hard drive (for storing programs and data), some type of CD or DVD drive (for installing programs, backing up files, and sharing files with others), and a floppy drive (for sharing small files with others).  Some users may choose to include an additional drive for a particular type of high-capacity removable media, such as Zip disks, if they only need to use the disks in their PC or a PC they know has a drive compatible with that medium.

Users who plan to transfer music, digital photos, and other multimedia data on a regular basis between several different devices -- such as a PC, digital camera, handheld PC, and printer -- will want to select and use the flash memory media that is most compatible with the devices they are using and obtain the necessary adapter for their PC.  Some of the most common types of portable storage media are compared in 
Figure 4-24 .


CHAPTER SUMMARY

Properties of Storage Systems

Storage systems make it possible to save programs, data, and processing results for later use.  They provide non-volatile storage, so when the power is shut off, the data stored on the storage medium remains intact.  This differs from memory, which is volatile.  The most common types of storage media are magnetic disks and optical discs, which are read by the appropriate type of drive.  Drives can be internal or external.

All storage systems involve two physical parts: A storage device and a storage medium.  In addition to being non-volatile, storage devices can record data either on removable media, which provide access only when inserted into the appropriate storage device, or fixed media, in which the media is permanently located inside the storage device.  Removable media provide the advantages of unlimited storage capacity, transportability, safer backup capability, and security.  Fixed media have the advantages of higher speed, lower cost, and greater reliability.

Two basic access methods characterize secondary storage systems: Sequential and random access.  Sequential access allows a computer system to retrieve the records in a file only in the same order in which they are physically stored.  Random access (or direct access) allows the system to retrieve records in any order.

Files (sometimes called documents) stored on a storage medium are given a filename and can be organized into folders.  This is referred to as logical file representation.  Physical file representation refers to how the files are physically stored on the storage medium by the computer.


Magnetic Disk Systems

Magnetic disk storage is most widely available in the form of hard disks and floppy disks.  Computer systems commonly include floppy disk storage because it provides a uniform removable storage system at a low cost.  Each side of a floppy disk holds data and programs in concentric tracks encoded with magnetized spots representing Os and 1 s.

Sector boundaries divide a floppy disk surface into pie-shaped pieces.  The part of a track crossed by a fixed number of contiguous sectors forms a cluster.  The disk's file directory or file allocation table (FAT), which the computer system maintains automatically, records where files stored on the disk are physically located.  To use a floppy disk, you insert it into a floppy disk drive.  

Today, floppy disks are facing challenges from other removable media with much higher storage capacities, such as Zip disks, SuperDisks, CDs, DVDs, and flash memory media.

Hard disk drives are the main storage medium for most PCs.  They offer faster access than floppy disks and much greater storage capacity.  A hard drive contains one or more hard disks permanently sealed inside along with an access mechanism.  A separate read/write head corresponds to each disk surface, and the access mechanism moves the heads in and out among the tracks to read and write data.  All tracks in the same position on all surfaces of all disks in a hard drive form a disk cylinder.  Hard drives can be divided into multiple partitions (logical drives) to reduce cluster size or to facilitate multiple users or operating systems.

Three events determine the time needed to read from or write to most disks: seek time, rotational delay, and data movement time.  The sum of these three time components is called disk access time.  A disk cache strategy, in which the computer fetches program or data contents in neighboring disk areas and transports them to RAM whenever disk content is retrieved, can speed up access time.

Three disk standards—EIDE, SCSI, and Fibre Channel—dominate the hard drive market, although some external drives can connect via a USB port.  If portability is required, portable hard drives, in which either the entire drive or a removable hard drive cartridge can be moved to another PC, are available.  Hard drives for notebook PCs can be internal, external, or in a PC card format.

Disk drives on larger computers implement many of the same standards as PC-based hard drives.  Instead of finding a single set of hard disks inside a hard drive permanently installed within a system unit, however, a storage system separate from the system unit often encloses several removable racks of hard disk drives, sometimes called an enterprise storage system.  Network attached storage (NAS) and storage area networks (SANs) are commonly used to provide storage for a business network.  RAID technology can be used on larger systems to increass fault tolerance and performance.



Optical Disc Systems

Optical discs store data optically using laser beams much more densely than magnetic disks.  They are divided into tracks and sectors like magnetic disks, but use a single grooved spiral track instead of concentric tracks.  Optical discs are available in a wide variety of CD and DVD formats and are read by CD or DVD drives.

CD-ROM discs come with data already stored on the disc.  Data is represented by pits and lands permanently burned into the surface of the disk.  CD-ROM discs cannot be erased or overwritten—they are read-only.  DVD-ROM discs are similar to CD-ROM discs, but they hold much more data (4.7 GB instead of 700 MB).  Recordable discs (CD-R, DVD-R, and DVD+R discs) and rewritable disks (CD-RW, DVD-RW, DVD+RW, DVD-RAM, and blue laser discs) can all be written to, but only recordable discs can be erased and rewritten to, similar to a floppy disk or hard drive.

Recordable CDs and DVDs store data by burning permanent marks onto the disc, similar to CD-ROM and DVD-ROM discs; rewritable discs typically use phase-change technology to change the reflectivity of the disc to represent Is and Os.  It is expected that, eventually, some form of DVD disc will eventually replace CDs as the optical disc standard.


Other Types of Storage Systems

Other types of storage systems include magneto-optical (MO) discs, which use a combination of magnetic and optical technology and magnetic tape, which stores data on plastic tape coated with a magnetizable substance.  Magnetic tapes are usually enclosed in cartridges and are inserted into a tape drive to be used.

Flash memory media are a rapidly growing new storage alternative.  They are used with digital cameras, portable PCs, and other portable devices, as well as with desktop PCs.  Plash memory can be in the form of flash memory sticks, flash memory cards, or flash memory drives.  Remote storage—using a storage device that is not directly a part of your PC system—typically involves using a network storage device or an online storage service.  

Online storage services enable users to share files with others over the Internet, access files while on the road, and backup documents.  Smart cards are credit-card-sized pieces of plastic that contain a chip or other circuitry usually used to store data or a monetary value.

Holographic storage, which uses multiple laser beams to store data in three dimensions, is a possibility for the future.


Comparing Storage Alternatives

Most PCs today include a hard drive, floppy disk drive, and some type of CD or DVD drive.  The type of optical drive and any additional storage devices are often determined by weighing a number of product characteristics and cost factors.  These characteristics include speed, compatibility, capacity, removability, and convenience.