Measured in MHz or GHz.

MHz = megahertz (millions of cycles per second)

GHz = Gigahertz (Billions of cycles per second)

Random Access Memory

184 pin module

A type of hard drive

They can be parallel or serial.

Sockets are always female.

Use in PC's is fading.

Universal Serial Bus connectors.

Hot-swappable and copliant with plug-and-play.

Buses also have a certain speed, measured in megahertz, which determines how fast the data can travel.

External Data Bus

This bus is typically rather quick and is independent of the rest of the computer’s operations.

The number of registers that a CPU has and the size of each (number of bits) help determine the power and speed of a CPU. For example a 32-bit CPU is one in which each register is 32 bits wide. Therefore, each CPU instruction can manipulate 32 bits of data.

Every computer contains an internal clock that regulates the rate at which instructions are executed and synchronizes all the various computer components.

The CPU requires a fixed number of clock ticks (or clock cycles) to execute each instruction. The faster the clock, the more instructions the CPU can execute per second.

a type of computer memory that can be accessed randomly; that is, any byte of memory can be accessed without touching the preceding bytes.

Short for binary digit, the smallest unit of information on a machine.

Abbreviation for binary term, a unit of storage capable of holding a single character. On almost all modern computers, a byte is equal to 8 bits.

the natural data size of a computer. The size of a word varies from one computer to another, depending on the CPU. For computers with a 16-bit CPU, a word is 16 bits (2 bytes). On large mainframes, a word can be as long as 64 bits (8 bytes).

The width of the address bus (that is, the number of wires) determines how many unique memory locations can be addressed.

Modern PCs and Macintoshes have as many as 36 address lines, which enables them theoretically to access 64 GB (gigabytes) of main memory. However, the actually amount of memory that can be accessed is usually much less than this theoretical limit due to chipset and motherboard limitations.

Northbridge is the chip or chips that connect a CPU to memory, the PCI bus, Level 2 cache and AGP activities.

The Northbridge can consist of more than one discrete chip.

the Southbridge is the chip that controls all of the computers I/O functions, such as USB, audio, serial, the system BIOS, the ISA bus, the interrupt controller and the IDE channels.

the Southbridge is typically only one discrete chip that rests on the Northbridge’s PCI bus.

Square in shape and has a removal arm.

• 5th Generation CPU
• External Speed Range = 50-66 MHz
• Internal Speed Range = 60-200 MHz
• Mulitplier range = x1-x3
• L1 cache? Yes, 16KB
• L2 cache? No
• L3 cache? No
• Package = PGA
• Sockets = 4 & 5

• 5th Generation CPU
• External Speed Range = 50-75 MHz
• Internal Speed Range = 60-150 MHz
• Mulitplier range = x1.5-x2
• L1 cache? Yes, 16KB
• L2 cache? No
• L3 cache? No
• Package = PGA
• Sockets = 7

• 6th Generation CPU
• External Speed Range = 60-66 MHz
• Internal Speed Range = 166-200 MHz
• Mulitplier range = x2.5-x3
• L1 cache? Yes, 16KB
• L2 cache? Yes, 256KB, 512KB, and 1MB
• L3 cache? No
• Package = PGA
• Sockets = 8

• 5th Generation CPU
• External Speed Range = 66-75 MHz
• Internal Speed Range = 166-200 MHz
• Mulitplier range = x2.5-x4.5
• L1 cache? Yes, 32KB
• L2 cache? No
• L3 cache? No
• Package = PGA
• Sockets = 7

• 6th Generation CPU
• External Speed Range = 66-100 MHz
• Internal Speed Range = 233-450 MHz
• Mulitplier range = x3.5-x4.5
• L1 cache? Yes, 32KB
• L2 cache? Yes, 512KB
• L3 cache? No
• Package = SEC
• Sockets = Slot 1

• 6th Generation CPU
• External Speed Range = 100-133 MHz
• Internal Speed Range = 450 MHz-1.26 GHz
• Mulitplier range = x4-x10
• L1 cache? Yes, 32KB
• L2 cache? Yes, 256KB or 512KB
• L3 cache? No
• Package = SEC-2, PGA
• Sockets = Slot 1, Socket 370

• 6th Generation CPU
• External Speed Range = 66-100 MHz
• Internal Speed Range = 533-700 MHz
• Mulitplier range = x8-x11.5
• L1 cache? Yes, 32KB
• L2 cache? Yes, 128KB
• L3 cache? No
• Package = PGA
• Sockets = 370

• 6th Generation CPU
• External Speed Range = 100 MHz (Doubled)
• Internal Speed Range = 500 Mhz-1 GHz
• Mulitplier range = x5-x10
• L1 cache? Yes, 128KB
• L2 cache? Yes, 512KB
• L3 cache? No
• Package = SEC
• Sockets = Slot A

• 6th Generation CPU
• External Speed Range = 100-133 MHz
• Internal Speed Range = 650 MHz-1.4 GHz
• Mulitplier range = x3.5-x14
• L1 cache? Yes, 128KB
• L2 cache? Yes, 256KB
• L3 cache? No
• Package = PGA
• Sockets = A

• 6th Generation CPU
• External Speed Range = 100 MHz
• Internal Speed Range = 600 Mhz-1.3 GHz
• Mulitplier range = x3.5-x14
• L1 cache? Yes, 128KB
• L2 cache? Yes, 64KB
• L3 cache? No
• Package = PGA
• Sockets = A

• 7th Generation CPU
• External Speed Range = 100-133-200 MHz (Quad speed)
• Internal Speed Range = 1.3-3.2 GHz
• Mulitplier range = x13-x23
• L1 cache? Yes, 128KB
• L2 cache? Yes, 256KB or 512KB
• L3 cache? No
• Package = 423-pin PGA, 478-pin PGA
• Sockets = 423, 478

• 7th Generation CPU
• External Speed Range = 133-166 MHz (Dual speed)
• Internal Speed Range = 1.3-2.16 GHz
• Mulitplier range = x13-x16.5
• L1 cache? Yes, 128KB
• L2 cache? Yes, 256KB or 512KB
• L3 cache? No
• Package = 462-pin PGA
• Sockets = A

• 6th Generation CPU
• External Speed Range = 100 MHz
• Internal Speed Range = 400-450 MHz
• Mulitplier range = x4-x4.5
• L1 cache? Yes, 32KB
• L2 cache? Yes, 512KB - 2MB
• L3 cache? No
• Package = SEC
• Sockets = Slot 2

• 6th Generation CPU
• External Speed Range = 100-133 MHz
• Internal Speed Range = 500 MHz-1 GHz
• Mulitplier range = x5-x7.5
• L1 cache? Yes, 32KB
• L2 cache? Yes, 256KB, 512KB, 1MB, 2MB
• L3 cache? No
• Package = SEC
• Sockets = Slot 2

• 7th Generation CPU
• External Speed Range = 100-133 MHz
• (Quad speed)
• Internal Speed Range = 1.4-3.06 GHz
• Mulitplier range = x14-x28
• L1 cache? Yes, 12KB
• L2 cache? Yes, 256KB, 512KB
• L3 cache? No
• Package = PGA
• Sockets = 603

• 7th Generation CPU
• External Speed Range = 100-133 MHz (Quad speed)
• Internal Speed Range = 1.4-2 GHz
• Mulitplier range = x14-x20
• L1 cache? Yes, 12KB
• L2 cache? Yes, 256KB, 512KB
• L3 cache? Yes, 512KB, 1MB, 2MB
• Package = PGA
• Sockets = 603

• 8th Generation CPU
• Physical Address = 44 bits
• Frontside Bus Width = 64 bit
• External Speed Range = 133 MHz (Quad speed)
• Internal Speed Range = 733-833 MHz
• Mulitplier range = x6-x8.5
• L1 cache? Yes, 32KB
• L2 cache? Yes, 96KB
• L3 cache? Yes, 2MB, 4MB
• Package = PAC
• Sockets = 418

• 8th Generation CPU
• Physical Address = 50 bits
• Frontside Bus Width = 128 bits
• External Speed Range = 100 MHz (Quad speed)
• Internal Speed Range = 900 MHz, 1GHz
• Mulitplier range = x9-x10
• L1 cache? Yes, 32KB
• L2 cache? Yes, 256KB
• L3 cache? Yes, 1.5MB, 3MB
• Package = OLGA
• Sockets = 611

• 8th Generation CPU
• Physical Address = 40 bits
• Frontside Bus Width = 128 bit
• External Speed Range = 6.4 GHz (Hypertransport)
• Internal Speed Range = 1.4-1.8 GHz
• Mulitplier range = x14-x20
• L1 cache? Yes, 128KB
• L2 cache? Yes, 1MB
• L3 cache? No
• Package = Micro-PGA
• Sockets = 940

• Number of Pins 273
• Pin Rows 4
• Voltage 5 volts
• Motherboard 1st Generation Pentium
• Supported Processors Pentium 60-66 MHz, Pentium OverDrive

Pentium MMX processors are not compatible with Socket 5. Consisting of 320 pins, this was the first socket to use a staggered pin grid array, or SPGA. This design allowed the chip's pins to be spaced closer together.

• Type: ZIF
• Contacts: 321
• FSB : 66 - 83 Mhz System Clock
• Voltage range : 2.5V - 3.5V
• Processors: 75-233 MHz Intel
Pentium, AMD K5 through K6, Cyrix 6x86 (and MX) P120 - P233

• Type PGA-ZIF
• Chip form factors Plastic pin grid array (PPGA) and Flip-chip pin grid array (FC-PGA and FC-PGA2)
• Contacts 370
• Bus Protocol GTL+
• FSB 66, 100 and 133 MHz
• Voltage range 1.05–2.1 V
• Processors Intel Celeron (PPGA, 300–533 MHz) Intel Celeron (FC-PGA, 533–1100 MHz) Intel Celeron (FC-PGA2, 900–1400 MHz) Intel Pentium III (FC-PGA, 500–1133 MHz) Intel Pentium III (FC-PGA2, 1000–1400 MHz) VIA Cyrix3/C3 (500–1200 MHz)

• Type PGA-ZIF
• Chip form factors Flip-chip pin grid array (FC-PGA2)
• Contacts 478
• Bus Protocol AGTL+
• FSB 400 MT/s 533 MT/s 800 MT/s
• Voltage range
• Processors Intel Pentium 4 (1.4 - 3.4 GHz) Intel Celeron (1.7 - 3.2 GHz) Celeron D (to 3.2 GHz) Intel Pentium 4 Extreme Edition (3.2, 3.4 GHz)

• Type PGA-ZIF
• Chip form factors Ceramic Pin Grid Array (CPGA) Organic Pin Grid Array (OPGA)
• Contacts 453
• Bus Protocol EV6
• FSB 100 MHz, 133 MHz, 166 MHz and 200 MHz equivalent to FSB200, FSB266, FSB333 and FSB400 (Double data rate Bus)
• Voltage range 1.1 - 2.05 V
• Processors AMD Athlon (650 MHz - 1400 MHz) AMD Athlon XP (1500+ - 3300+) AMD Duron (600 MHz - 1800 MHz) AMD Sempron (2000+ - 3300+) AMD Athlon MP (1000 MHz - 3000+)

Super Socket was used by AMD K6-2 and K6-III processors, and some of the final Cyrix M-II processors. It is backward compatible with Socket 7, meaning a Socket 7 CPU can be used with a Super Socket 7 motherboard, but a Super Socket 7 CPU cannot operate at full speed in a Socket 7 motherboard.

Socket 5 CPUs are pin-compatible with Super Socket 7, but no motherboard designed for Super Socket 7 ever supported the voltages needed for Socket 5 CPUs.

While AMD had previously always used Intel sockets for their processors, Socket 7 was the last one for which AMD retained legal rights. Intel had hoped by discontinuing Socket 7 development and moving to Slot 1 that AMD would be left with an outdated platform, making their processors non-competitive. By extending the FSB from 66 to 100 MHz, Super Socket 7 gave AMD the stopgap solution they needed while developing their own independent motherboard infrastructure, Slot A.

While the architecture was cheap, and served the intended purpose, many of the third party chipsets provided by VIA, SiS, and others, were of low quality, especially in regard to the AGP implementations. The reputation AMD gained for buggy incompatible motherboards lingered, and subsequently AMD put in place a quality assurance program for the Athlon processor.

Once the data is stored in the cache, future use can be made by accessing the cached copy rather than refetching or recomputing the original data, so that the average access time is lower.

When L2 cache is found on the processor, if cache is found on the motherboard, it is properly known as L3 cache.

First introduced with the Intel Pentium and Pentium Pro computers. In some cases this required that the user install or have installed optional L2 cache for L2 support.

L2 Cache on computers can range from 64KB to 2MB in size. However, early versions of the Intel Celeron had no L2 cache.

When referring to computer processors, L1 cache is cache built on to the processor and is commonly the fastest cache in the computer.

DRAM stores its information in a cell containing a capacitor and transistor; because of this design, these cells must be refreshed with new electricity every few milliseconds allowing the memory to keep its charge and hold the data as long as needed.

When using this type of memory, if the computer is powered off, the information within memory will be lost.

Memory refresh-rate is commonly displayed as xK, where x is a number such as 2 or 4. If the memory has a refresh-rate of 2k, this indicates how many rows it takes to complete the refresh.

In our example of 2k, this indicates there are 2 thousand rows. The refresh cycle indicates the time it requires to complete one row.

RAM is volatile, but ROM is not. It is because of RAM's volatile nature that users must frequently save their work to a permanent medium, such as a hard drive, to avoid losing data if the system's power is interrupted.

Wait states occur frequently when the computer's processor has a higher clock speed than other components in the computer.

One common wait state is unnoticeable; however, several can affect the overall performance of a computer.

Although quicker than DRAM, SRAM is much more expensive and requires more power; therefore, it is commonly only used in cacheive video card memory.

A flip-flop typically includes zero, one, or two input signals; a clock signal; and an output signal, though many commercial flip-flops additionally provide the complement of the output signal. Some flip-flops include a clear input signal, which resets the current output.

Because flip-flops are implemented as integrated circuit chips, they also require power and ground connections.

Branch predictors are crucial in today's modern, superscalar processors for achieving high performance. They allow processors to fetch and execute instructions without waiting for a branch to be resolved.

The elements of a pipeline are often executed in parallel or in time-sliced fashion; in that case, some amount of buffer storage is often inserted between elements.

Computer-related pipelines include:
Instruction pipelines, such as the classic RISC pipeline, which are used in processors to allow the parallel execution of two or more consecutive instructions from a nominally sequential stream; the processing elements are the logical circuits that implement the various stages of an instruction (address decoding and arithmetic, register fetching, cache lookup, etc.).

For every tick of the front side bus (FSB) clock, a frequency multiplier causes the CPU to perform x cycles, where x is the multiplier.

By using the CPUID opcode, software can determine processor type and the presence of features (like MMX/SSE).

The CPUID opcode is 0FA2h and the value in the EAX register specifies what information to return.

These regulators reduce the 5V signal to those voltages typically needed by processors: 3.3V or lower.

It can be soldered to the motherboard or be an installable device. It allows processors with different supply voltage to be mounted on the same motherboard.

Some voltage regulators provide a fixed supply voltage to the processor, but most of them sense the required supply voltage from the processor. In particular, VRMs that are soldered to the motherboard are supposed to do the sensing, according to the Intel specification.

The correct supply voltage is communicated by the microprocessor to the VRM at startup via a number of bits called VID (voltage identificator). In particular, the VRM initially provides a standard supply voltage to the VID logic, which is the part of the processor whose only aim is to then send the VID to the VRM.

When the VRM has received the VID identifying the required supply voltage, it starts acting as a voltage regulator, providing the required constant voltage supply to the processor.

• 1. Quad pipelining
• 2. Dynamic processing
• <3. On-chip L2 cache

Electrically Erasable Programmable Read-Only Memory.

A type of ROM technology which can be electrically reprogrammed while still in the PC.

A device or medium for temporary storage of programs and data during program execution.

A computer’s temporary storage capacity is measured in kilobytes (KB) or megabytes (MB) of RAM (random-access memory).

Long-term data storage on disks is also measured in kilobytes, megabytes, gigabytes, and terabytes.

Memory-Management Unit

Upper Memory Area

Upper Memory Block

In any IBM PC-compatible computer, the first 640K of the computer’s RAM.

This amount proved to be insufficient because of programs that demanded more memory and users who wanted to run more than one program at a time. Many DOS users equipped their systems with extended or expanded memory and the memory management programs needed to access that memory.

Extended Memory Services

Expanded Memory Specification

A method of memory management developed by Intel, Lotus, and Microsoft that enabled MS-DOS computers to use memory exceeding the 640K limit imposed by MS-DOS.

Memory that conforms to this standard is called expanded memory. Originally, this standard required that a special hardware device be added to the computer, but that was quickly supplanted by a less efficient but much cheaper software solution.

Expanded memory has been made almost obsolete by the Windows environment, which isn’t limited by the 640K barrier. EMS is now required almost solely by DOS-based games.

Pertaining to a device or facility that does not physically exist, yet behaves as if it does.

For example, a system with 4MB of virtual memory may have only 1MB of physical memory plus additional (slower and cheaper) auxiliary memory. Yet programs written as if 4MB of physical memory were available will run correctly.

Terminate and Stay Resident

A DOS program that closes immediately after starting up, but leaves a tiny piece of itself in memory.

TSRs are used to handle a broad cross-section of DOS-level system needs, such as running hardware (MOUSE.COM) or applying higher-level functionality to hardware already under the control of device drivers.

MSCDEX.EXE, for example, assigns a drive letter to a CD-ROM drive after the CD-ROM driver has loaded in CONFIG.SYS.

Cylinders, Heads, and Sectors per Track

The initials for the combination of the three critical geometries used to determine the size of a hard drive—cylinders, heads, and sectors per track.

The first sector on an IBM-PC hard drive or floppy disk, track 0.

The boot-up software in ROM tells the computer to load whatever program is found there. If a system disk is read, the program in the boot record directs the computer to the root directory to load the operating system.

A segment of one of the concentric tracks encoded on the disk during a lowlevel format.

Sectors hold 512 bytes of data.

The basic unit of storage on a floppy or hard disk.

Two or more sectors are contained in a cluster. When DOS/Windows 9x stores a file on disk, it writes those files into dozens or even hundreds of contiguous clusters.

If there aren’t enough contiguous open clusters available, the operating system finds the next open cluster and writes there, continuing this process until the entire file is saved.

The FAT tracks how the files are distributed among the clusters on the disk.

Also called Thinnet or Thin Ethernet

An Ethernet LAN designed to run on common coax RG-58 cabling, almost exactly like the coax for cable television.

It runs at 10 megabits per second and has a maximum segment length of 185 meters. It uses baseband signaling and BNC connectors.

No industry standard spelling exists, so sometimes written 10BASE-2 or 10Base2.

Also called Thicknet or Thick Ethernet

The original Ethernet LAN, designed to run on specialized coax cabling. It runs at 10 megabits per second and has a maximum segment length of 500 meters.

10Base5 uses baseband signaling. It uses DIX or AUI connectors and external transceivers.

No industry standard spelling exists, so sometimes written 10BASE-5 or 10Base5.

The maximum length for the cabling between the NIC and the hub (or switch, repeater, etc.) is 100 meters. It uses baseband signaling.

No industry standard spelling exists, so sometimes written 10BASE-T or 10BaseT.

Fiberoptic implementation of Ethernet that runs at 100 megabits per second.

It uses baseband signaling. Maximum length of the cable is 400 meters.

No industry standard spelling exists, so sometimes written 100BASE-FX or 100BaseFX.

A generic term for any Ethernet cabling system that is designed to run at 100 megabits per second on UTP cabling.

It uses baseband signaling. No industry standard spelling exists, so sometimes written 100BASE-T or 100BaseT.

An Ethernet LAN designed to run on UTP cabling.

It runs at 100 megabits per second and uses four pairs of wires on CAT3 or better cabling. No industry standard spelling exists, so sometimes written 100BASE-T4 or 100BaseT4.

486 CPUs that operate externally at one speed and internally at a speed which is two, three, or four times faster.

Although the internal speed can be more than two times as fast as the external speed, these CPUs are known collectively as “clock doublers.”

Advanced Configuration and Power Interface

A power management specification that far surpasses its predecessor, APM, by providing support for hot swappable devices and better control of power modes.

The wires leading from the CPU to the memory controller chip (chipset) that enable the CPU to address RAM.

Also used by the CPU for I/O addressing. An internal electronic channel from the microprocessor to random-access memory, along which the addresses of memory storage locations are transmitted.

Like a post office box, each memory location has a distinct number or address; the address bus provides the means by which the microprocessor can access every location in memory.

Asymmetric Digital Subscriber Line

Accelerated Graphics Port

Arithmetic Logic Unit

American National Standards Institute

Application Programming Interface

Advanced Power Management

An attribute of a file that shows whether the file has been backed up since the last change.

Each time a file is opened, changed, or saved, the archive bit is turned on. Some types of backups will turn off this archive bit to indicate that a good backup of the file exists on tape.

Address Resolution Protocol

American Standard Code for Information Interchange

Advanced SCSI Programming Interface

A program that converts symbolically coded programs into object-level machine code.

In an assembler program, unlike a compiler, there is a one-to-one correspondence between human-readable instructions and the machine-language code.

AT Attachment

A type of hard drive and controller.

ATA was designed to replace the earlier ST506 and ESDI drives without requiring replacement of the AT BIOS— hence, AT attachment. These drives are more popularly known as IDE drives. See also ST506, ESDI, and IDE.

ATA/33 A hard drive standard with drive transfer speeds up to 33MB/s. (See Ultra DMA.)

ATA/66 A hard drive standard with drive transfer speeds up to 66MB/s. (See Ultra DMA.)

ATA/100 A hard drive standard with drive transfer speeds up to 100MB/s. (See Ultra DMA.)

AT Attachment Packet Interface

AT eXtended

Attachment Unit Interface

A batch file that DOS executes when you start or restart the system.

AUTOEXEC.BAT is not necessary, but when you’re running a computer to which you’ve attached several devices and several different software applications, the file is essential for efficient operation.

AUTOEXEC.BAT files commonly include PATH statements that tell DOS where to find application programs, and commands to install a mouse or operate your printer.

The set of wires that connect the CPU to Level 2 cache.

First appearing in the Pentium Pro, most modern CPUs have a special backside bus. Some buses, such as that in the later Celeron processors (300A and beyond), run at the full speed of the CPU, whereas others run at a fraction.

Earlier Pentium IIs, for example, had backside buses running at half the speed of the processor. See also Frontside Bus and External Data Bus.

A piece of the spectrum occupied by some form of signal, such as television, voice, fax data, etc.

Signals require a certain size and location of bandwidth in order to be transmitted. The higher the bandwidth, the faster the signal transmission, allowing for a more complex signal such as audio or video.

Because bandwidth is a limited space, when one user is occupying it, others must wait their turn. Bandwidth is also the capacity of a network to transmit a given amount of data during a given period.

Beginner’s All-purpose Symbolic Instruction Code

A number system with a base of 2, unlike the number systems most of us use which have bases of 10 (decimal numbers), 12 (measurement in feet and inches), and 60 (time).

Binary numbers are preferred for computers for precision and economy. An electronic circuit that can detect the difference between two states (on-off, 0-1) is easier and more inexpensive to build than one that could detect the differences among ten states (0-9).

Basic Input/Output Services

Classically, the software routines burned onto the System ROM of a PC. More commonly seen as any software that directly controls a particular piece of hardware.

A set of programs encoded in Read-Only Memory (ROM) on computers. These programs handle startup operations and low-level control of hardware such as disk drives, the keyboard, and monitor.

Binary Digit

The term is derived from “pull yourself up by your bootstraps.” PCs must do that because RAM doesn’t retain program instructions when power is turned off.

A cold boot occurs when the PC is physically switched on. A warm boot loads a fresh OS without turning off the computer, lessening the strain on the electronic circuitry. To do a warm boot, press the CTRL-ALTDELETE keys at the same time twice in rapid succession (the three-fingered salute).

Bits Per Second

For example, an Ethernet bridge only looks at the Ethernet address. Bridges filter and forward packets based on MAC addresses and operate at Level 2 (Data Link layer) of the OSI seven-layer model.

Electronic storage, usually DRAM, that holds data moving between two devices.

Buffers are used in situations where one device may send or receive data faster or slower than the other device with which it is in communication.

For example, the BUFFERS statement in DOS is used to set aside RAM for communication with hard drives.

Compact Disk-Recordable

Compact Disk/Read Only Memory

Compact Disk-Rewritable

Complex-Instruction Set Computing

The pulses are used to synchronize the flow of information through the computer’s internal communication channels. Some computers also contain a circuit that tracks hours, minutes, and seconds.

The basic unit of storage on a floppy or hard disk.

Two or more sectors are contained in a cluster. When DOS/Windows 9x stores a file on disk, it writes those files into dozens or even hundreds of contiguous clusters. If there aren’t enough contiguous open clusters available, the operating system finds the next open cluster and writes there, continuing this process until the entire file is saved.

The FAT tracks how the files are distributed among the clusters on the disk.

Complementary Metal-Oxide Semiconductor

Originally, the type of nonvolatile RAM that held information about the most basic parts of your PC such as hard drives, floppies, and amount of DRAM.

Today, actual CMOS chips have been replaced by Flash-type non-volatile RAM. The information is the same, however, and is still called CMOS (pronounced see-moss) even though it is now almost always stored on Flash RAM.

Short for coaxial.

A set of Ethernet segments that receive all traffic generated by any node within those segments.

Repeaters, amplifiers, and hubs do not create separate collision domains, but bridges, routers, and switches do.

A system name that refers to the serial communications ports available on your computer.

When used as a program extension, .COM indicates an executable program file limited to 64K.

In DOS, a file that contains the command processor.

An ASCII text file in the root directory that contains configuration commands.

CONFIG.SYS enables the system to be set up to configure high, expanded, and extended memories by the loading of HIMEM.SYS and EMM386.EXE drivers, as well as drivers for non-standard peripheral components.

In any IBM PC-compatible computer, the first 640K of the computer’s RAM.

This amount proved to be insufficient because of programs that demanded more memory and users who wanted to run more than one program at a time.

Many DOS users equipped their systems with extended or expanded memory and the memory management programs needed to access that memory.

Central Processing Unit

Cyclical Redundancy Check

A very accurate mathematical method that is used to check for errors in long streams of transmitted data.

Before data is sent, the main computer uses the data to calculate a CRC value from the data’s contents. If the receiver calculates a different CRC value from the received data, the data was corrupted during transmission and is resent. Ethernet packets have a CRC code.

Continuity RIMM

Carrier Sense Multiple Access with Collision Detection

In a TCP/IP network, the nearest router to a particular host.

Dynamic Host Configuration Protocol

Dual Inline Memory Module

Provides fault tolerance by writing data across multiple drives and then including an additional drive, called a parity drive, that stores information to rebuild the data contained on the other drives.

Disk striping with parity requires at least three physical disks: two for the data and a third for the parity drive. It provides data redundancy at RAID levels 3–5 with different options.

Digital, Intel, Xerox

Direct Memory Access

Domain Name System

Term used to describe groupings of users, computers, or networks.

In Microsoft networking, a domain is a group of computers and users that share a common account database, called a SAM, and a common security policy.

On the Internet, a domain is a group of computers that share a common element in their hierarchical name.

Other types of domains exist—e.g., collision domain, etc.

Dynamic Random-Access Memory

The memory used to store data in most personal computers. DRAM stores each bit in a “cell” composed of a transistor and a capacitor.

Because the capacitor in a DRAM cell can only hold a charge for a few milliseconds, DRAM must be continually refreshed, or rewritten, to retain its data.

Digital Subscriber Line

A high-speed Internet connection technology that uses a regular telephone line for connectivity.

DSL comes in several varieties, including Asynchronous (ADSL) and Synchronous (SDSL), and many speeds. Typical home-user DSL connections are ADSL with a download speed of up to 1.5 Mbps and an upload speed of 384 Kbps.

Digital Versatile Disk

External Data Bus

Enhanced IDE

Enhanced ISA

Electro-Magnetic Interference

EMI is electrical interference from one device to another, resulting in poor performance of the device being interfered with.

An example is having static on your TV while running a blow dryer, or placing two monitors too close together and getting a “shaky” screen.

Expanded Memory Specification

A method of memory management developed by Intel, Lotus, and Microsoft that enabled MS-DOS computers to use memory exceeding the 640K limit imposed by MS-DOS.

Memory that conforms to this standard is called expanded memory. Originally, this standard required that a special hardware device be added to the computer, but that was quickly supplanted by a less efficient but much cheaper software solution.

Expanded memory has been made almost obsolete by the Windows environment, which isn’t limited by the 640K barrier. EMS is now required almost solely by DOS-based games.

Refers here to the process of putting the packets from one protocol inside the packets of another protocol.

An example of this is TCP/IP encapsulation in NetWare servers, where IPX/SPX packets are placed inside TCP/IP packets, enabling Novell NetWare to use TCP/IP for transport while still allowing the network operating system to get the data it needs from IPX/SPX.

Enhanced Small Device Interface

File Allocation Table

A hidden table of every cluster on a hard disk. The FAT records how files are stored in distinct clusters. The address of the first cluster of the file is stored in the directory file.

In the FAT entry for the first cluster is the address of the second cluster used to store that file.

In the entry for the second cluster for that file is the address for the third cluster, and so on until the final cluster, which gets a special ‘end of file’ code.

This table is the only way DOS knows where to access files. There are two FATs, mirror images of each other, in case one is destroyed or damaged.

File allocation table that uses 32 bits for addressing clusters. Commonly used with Windows 98 and Windows ME systems.

Some Windows 2000 Professional systems also use FAT32, although most use the more robust NTFS.

Fiber Distributed Data Interface

An IEEE 1394 standard to send wide-band signals over a thin connector system that plugs into TVs, VCRs, TV cameras, PCs, etc.

This serial bus, developed by Apple and Texas Instruments, enables connection of 60 devices at speeds ranging from 100 to 400 megabits per second.

A type of ROM technology which can be electrically reprogrammed while still in the PC.

Flash is the overwhelmingly most common storage medium of BIOS in PCs today, as it can be upgraded without even having to open the computer on most systems.

Fast Page Mode

Basic component of communication over a network.

A group of bits of fixed maximum size and well-defined format that is switched and transmitted as a complete whole through a network. It contains source and destination address, data, and control information. See also Packet.

Name for the wires that connect the CPU to the main system RAM.

Generally running at speeds of 66-133 MHz. Distinct from the Expansion Bus and the Backside Bus, though it shares wires with the former.

File Transfer Protocol

The technical meaning is a hardware or software setup that translates between two dissimilar protocols.

For example, Prodigy has a gateway that translates between its internal, proprietary e-mail format and Internet e-mail format.

Another, less technical meaning of gateway is any mechanism for providing access to another system, e.g., AOL might be called a gateway to the Internet. See also Default Gateway

A base-16 numbering system using 10 digits (0 through 9) and six letters (A through F).

Used in the computer world as a shorthand way to write binary numbers, by substituting one hex digit for a four-digit binary number (for example, hex 9  binary 1001).

High Memory Area)

The first 64K–16 bytes of memory above 1 megabyte is known as the HMA. Programs that conform to XMS can use HMA as a direct extension of conventional memory.

Most of the portions of DOS that must be loaded into conventional memory can be loaded into the HMA.

A single device (usually a computer) on a TCP/IP network that has an IP address—any device that can be the source or destination of a data packet.

A single device (usually a computer) on a TCP/IP network that has an IP address—any device that can be the source or destination of a data packet.

HyperText Markup Language)

HyperText Transfer Protocol)

An electronic device that sits at the center of a star topology network, providing a common point for the connection of network devices.

In a 10BaseT Ethernet network, the hub contains the electronic equivalent of a properly terminated bus cable; in a Token Ring network, the hub contains the electronic equivalent of a ring.

Intelligent or Integrated Drive Electronics)

A PC specification for small- to medium-sized hard drives in which the controlling electronics for the drive are part of the drive itself, speeding up transfer rates and leaving only a simple adapter (or “paddle”).

IDE only supported two drives per system of no more than 504 megabytes each, and has been completely supplanted by Enhanced IDE.

EIDE supports four drives of over 8 gigabytes each and more than doubles the transfer rate.

The more common name for ATA drives. (See also ATA.)

(Institute of Electronic and Electrical Engineers)

Input/Output

A general term for reading and writing data to a computer.

The term “input” includes data from a keyboard, pointing device (such as a mouse), or loading a file from a disk.

“Output” includes writing information to a disk, viewing it on a CRT, or printing it to a printer.

Also called Internet Address

The numeric address of a computer connected to the Internet. The IP address is made up of octets of 8-bit binary numbers that are translated into their shorthand numeric values.

The IP address can be broken down into a network ID and a host ID.

Internet Protocol

The Internet standard protocol that provides a common layer over dissimilar networks used to move packets among host computers and through gateways if necessary.

Part of the TCP/IP protocol suite.

Interrupt Request,/p>

A signal from a hardware device, such as a modem or a mouse, indicating that it needs the CPU’s attention.

In PCs, IRQs are sent along specific IRQ channels associated with a particular device. It is therefore important to ensure that two devices do not share a common IRQ channel.

Industry Standard Architecture

The Industry Standard Architecture design is found in the original IBM PC for the slots on the motherboard that allowed additional hardware to be connected to the computer’s motherboard.

An 8-bit, 8.33 MHz expansion bus was designed by IBM for its AT computer and released to the public domain.

An improved 16-bit bus was also released to the public domain. Various other designs such as IBM’s MicroChannel and EISA bus tried to improve on the design without much popularity.

ISA only supports 8- and 16-bit data paths, so 32-bit alternatives such as PCI and AGP have become popular.

Although ISA slots linger on most motherboards, they are on the way out, replaced by the newer 32-bit slots.

Integrated Services Digital Network

The CCITT (Comité Consutatif Internationale de Télégraphie et Téléphonie) standard that defines a digital method for communications to replace the current analog telephone system.

ISDN is superior to telephone lines because it supports up to 128 Kbps transfer rate for sending information from computer to computer.

It also allows data and voice to share a common phone line.

Joint Photographic Experts Group

A method of formatting images for efficient storage and transfer across phone lines; JPEG files are often a factor of 10 or more times smaller than non-compressed files.

JPEG is a lossy format, meaning that some data is lost when an image is converted.

Most JPEG conversion software enables the user to decide between more or less compression, more coming at the cost of image quality.

JPEG supports 24-bit images (up to 16.7 million colors). Because computers running MS-DOS are limited in their filenames, this format is also referred to as JPG.

Local Area Network

In the communications field, a grouping of related tasks involving the transfer of information.

Also, a level of the OSI reference model used for networking computers.

In graphics work, images can be created in layers, which can be manipulated separately and then flattened into a single image.

Logical Block Addressing

A translation (algorithm) of IDE drives promoted by Western Digital as a standardized method for breaking the 504MB limit in IDE drives.

Subsequently universally adopted by the PC industry, LBA is standard on all EIDE drives.

A networking term referring to a twisted-pair Ethernet segment, formally defined as a point-to-point full duplex medium that connects two and only two MDIs (attachment mechanisms).

Functionally that means an Ethernet cable connected at each end to a computer or another network device.

Media Access Control

Unique 48-bit address assigned to each network card. IEEE assigns blocks of possible addresses to various NIC manufacturers to help ensure that the address is always unique.

The Data Link layer of the OSI model uses MAC addresses for locating machines.

Also called Math Unit or Floating-Point Unit (FPU)

A secondary microprocessor whose function is the handling of floating-point arithmetic. Although originally a physically separate chip, math coprocessors are now built into today’s CPUs.

Multistation Access Unit

Micro Channel Architecture

Memory Controller Chip

Megahertz

Also called CPU

Musical Instrument Digital Interface

MIDI is a standard that describes the interface between a computer and a device for simulating musical instruments.

Rather than sending large sound samples, a computer can simply send “instructions” to the instrument describing pitch, tone, and duration of a sound.

MIDI files are therefore very efficient. Because a MIDI file is made up of a set of instructions rather than a copy of the sound, it is easy to modify each component of the file.

Additionally, it is possible to program many channels, or “voices,” of music to be played simultaneously, creating symphonic sound.

Memory-Management Unit

MOdulator/DEModulator

A device that converts a digital bit stream into an analog signal (modulation) and converts incoming analog signals back into digital signals (demodulation).

The analog communications channel is typically a telephone line and the analog signals are typically sounds.

Motion Picture Experts Group

A sophisticated video standard that enables digital video to be compressed using a form of JPEG image compression and a technique called differencing, in which only the differences between frames are recorded, rather than the frame itself.

MPEG-1 Audio Layer 3

An audio compression scheme used extensively on the Internet.

Multistation Access Unit—Also called MAU

NetBIOS Extended User Interface

Network Basic Input/Output System

A collection of two or more computers interconnected by telephone lines, coaxial cables, satellite links, radio, and/or some other communication technique.

A computer “network” is a group of computers that are connected together and that communicate with one another for a common purpose.

Computer networks support “people and organization” networks, users who also share a common purpose for communicating.

Network Interface Card

Network Operating System

An NOS is an operating system that provides basic file and supervisory services over a network.

While each computer attached to the network will have its own OS, the NOS describes which actions are allowed by each user and coordinates distribution of networked files to the user who requests them.

NT File System

Open Systems Interconnect

An architecture model based on the OSI protocol suite that defines and standardizes the flow of data between computers. The seven layers are:

• Layer 1 The Physical layer defines hardware connections and turns binary into physical pulses (electrical or light). Repeaters and hubs operate at the Physical layer.
• Layer 2 The Data Link layer identifies devices on the Physical layer. MAC addresses are part of the Data Link layer. Bridges operate at the Data Link layer.
• Layer 3 The Network layer moves packets between computers on different networks. Routers operate at the Network layer. IP and IPX operate at the Network layer.
• Layer 4 The Transport layer breaks data down into manageable chunks. TCP, UDP, SPX, and NetBEUI operate at the Transport layer.
• Layer 5 The Session layer manages connections between machines. NetBIOS and Sockets operate at the Session layer.
• Layer 6 The Presentation layer, which can also manage data encryption, hides the differences between various types of computer systems.
• Layer 7 The Application layer provides tools for programs to use to access the network (and the lower layers). HTTP, FTP, SMTP, and POP3 are all examples of protocols that operate at the Application layer.

The Physical layer

The Data Link layer

The Network layer

The Transport layer

The Session layer

The Presentation layer

The Application layer