ToggIT Technopedia
 

Toggit's Encyclopedia of Networking Terms and Computer Dictionary.

º»   To search for a term, type in a word, phrase or acronym. You do not need to use periods or punctuations. If you are unsure of the spelling, just type in the first few letters of the word. You can also use the links below to search by letter: A B C D E F G H I J K L M N O P Q R S T U V W X Y Z   Earn your ASSOCIATE DEGREE through Penn Foster College Great programs for IT Technology! Save big on your AA on Applied Computer Science, Multimedia & Design, Web Programming and PC Maintenance Technology.

Search for: p

P See peta-.

Search Alphabetical Listings for the letter 'P' P See peta-. PABX See private automatic branch exchange. PACE See Priority Access Control Enabled. Packard Bell NEC Major manufacturer of PCs and servers; manages all NEC Corporation’s computer manufacturing outside Japan. Packard Bell has recently established a configure-it-yourself direct sales promotion known as NEC Now. For more information on Packard-Bell, see www.packardbell.com. See also NEC Corporation. packet Any block of data sent over a network or communications link. Each packet may contain sender, receiver, and error-control information, in addition to the actual message, which may be data, connection management controls, or a request for a service. Packets may be fixed- or variable-length, and they will be reassembled if necessary when they reach their destination. The actual format of a packet depends on the protocol that creates the packet; some protocols use special packets to control communications functions in addition to data packets. See also datagram; frame; packet switching. packet assembler/disassembler Abbreviated PAD. A device that is connected to a packet-switched network and that converts a serial data stream from a characteroriented device, such as a bridge or a router, into packets suitable for transmission. It also disassembles packets back into characters for transmission to the character device. PADs are often used to connect a terminal or computer to an X.25 packetswitched network. Packet Burst Protocol Abbreviated PBP. A Novell NetWare protocol built on top of IPX that speeds up the transfer of multipacket NetWare Core Protocol (NCP) data transfers between a workstation and a server by removing the need to sequence and acknowledge every packet. Using PBP, a workstation or server can transmit a burst of packets before requiring an acknowledgment, thus reducing network traffic. See also Internetwork Packet eXchange. packet filtering A process used by bridges to limit protocol-specific traffic to one segment of the network, to isolate e-mail domains, and to perform other traffic-control functions. The network administrator sets the packet-filtering specifications for each bridge. If a packet matches the specifications, the bridge can either accept it or reject it. See also application-level filter; firewall.   packet-level filter packet-level filter A category of firewall that provides a high degree of convenience, but a relatively low level of security. A packet-level filter blocks or forwards a packet solely on its merits, without taking into account past history; the filter may also look at the source and destination address information contained in the packet. This kind of filter is easy to implement, has little effect on network operations, but can be bypassed by encapsulating a blocked protocol within an allowed protocol. See also application-level filter; firewall. Packet-level Procedure Abbreviated PAP. An X.25 full-duplex protocol for the transfer of packets between a computer and a modem. packet signature See NCP packet signature. packet sniffer A program used by an intruder to monitor a data stream for a pattern such as a password or credit card number. Packet sniffers also have a more salutary purpose when used for network analysis and troubleshooting by the system administrator. See also network analyzer; protocol analyzer. packet-switched network A network that consists of a series of interconnected circuits that route individual packets of data over one of several routes, offering flexibility and high reliability. A packet-switched network may also be called connectionless because it contains many different possible connections and routes that a packet might take to reach its destination. The term often refers to networks using the international standard X.25. See also packet switching. packet switching A data-transmission method that simultaneously routes and transmits data packets from many customers over a communications channel or telephone line, thus optimizing use of the line. An addressed packet is routed from node to node until it reaches its destination, although related packets may not all follow the same route to that destination. Because long messages may be divided into several packets, packet sequence numbers are used to reassemble the original message at the destination node. The standard for packet-switching networks is defined in CCITT recommendation X.25. The Internet is an example of a packet-switching network. See also Consultative Committee for International Telephony and Telegraphy; packet- switched network. PAD See packet assembler/disassembler. page A single document available on the World Wide Web or on a corporate intranet. A page can contain any combination of text, graphics, animated graphics, audio, and video and can be static or dynamic. See also HyperText Markup Language. page fault In Microsoft Windows, the fault that occurs when Windows attempts   parallel processing to read to or write from a virtual memory location that is designated not present. In Microsoft Windows Task Manager, the number of times information is read from disk because it was not present in memory. page-mode RAM A memory-management technique used to speed up the performance of dynamic RAM (DRAM). In a page-mode memory system, the memory is divided into pages by specialized DRAM chips. Consecutive accesses to memory addresses in the same page result in a page-mode cycle that takes about half the time of a regular DRAM cycle. paged memory management unit Abbreviated PMMU. A specialized chip designed to manage virtual memory. High-end processors, such as the Motorola 68040 and the Intel Pentium, have all the functions of a PMMU built into the chip itself. See also virtual memory. paging file In Microsoft Windows 2000, a hidden file used to hold programs and data files that will not currently fit into memory. The paging file, along with an area of physical memory, make up the virtual memory system. In most other operating systems, the paging file is known as the swap file. PalmPilot The hand-held computer from 3Com Corporation, which has proved to be extremely popular, with more than 1 million units sold to date. See also 3Com Corporation. PAP See Packet-level Procedure; Password Authentication Protocol; Printer Access Protocol. parallel communications The transmission of information from computer to computer or from computer to peripheral device, in which all the bits that make up the character are transmitted at the same time over a multiline cable. See also serial communications. parallel port An input/output (I/O) port that manages information eight bits at a time; often used to connect a parallel printer. See also parallel communications; RS-232- C; serial communications; serial port. parallel processing A computing method that can be performed by systems containing two or more processors operating simultaneously. In parallel processing, all processors work on different aspects of the same program at the same time, in order to share the computational load. Parallel-processing computers can achieve incredible speeds. The Cray X-MP48 peaks at 1000 million floating-point operations per second (1000 MFLOP) using four extremely powerful processors, and parallelhypercube systems, first marketed by Intel, can exceed 65,536 processors with speeds of up to 262 billion floating-point operations per second (262 GFLOP). In all but the most trivial parallel-processing applications, the programmer or the operating system must assign approximate processor loads; otherwise, it is possible for nonoptimized systems to fail to take advantage of the power available and, in the worst case,   parent directory run more slowly than on single-processor systems. All this speed is used for applications such as weather forecasting, in which the predictive programs can take as long to run as the weather actually takes to arrive, 3-D seismic modeling, groundwater and toxic flow studies, and the modeling of full-motion dinosaur images used in movies. See also asymmetrical multiprocessing; symmetrical multiprocessing. parent directory In a hierarchical directory system, such as that used by MS-DOS, OS/2, Windows, and Unix, the directory immediately above the current directory. The special symbol .. is shorthand for the name of the parent directory. See also period and double-period directories. parent domain In Microsoft Windows 2000 Server, any domain that has another domain subordinate to it. See also domain; first-layer domain. parent object In Novell Directory Services (NDS), an object that contains another object. This term is relative, because any particular parent object also has parent objects of its own and can be considered a child object from a certain perspective. parity In communications, a simple form of error checking that uses an extra or redundant bit after the data bits but before the stop bit or bits. Parity may be set as follows: _ Odd Indicates that the sum of all the 1 bits in the byte plus the parity bit must be odd. If the total is already odd, the parity bit is set to zero; if it is even, the parity bit is set to one. _ Even If the sum of all the 1 bits is even, the parity bit must be set to zero; if it is odd, the parity bit must be set to one. _ Mark The parity bit is always set to one and is used as the eighth bit. _ Space The parity bit is set to zero and is used as the eighth bit. _ None If parity is set to none, there is no parity bit, and no parity checking is performed. The parity settings used by both communicating computers must match. Most online services, such as CompuServe or America Online, use no parity and an 8-bit data word. See also asynchronous transmission; parity checking; parity error. parity bit An extra or redundant bit used to detect data transmission errors. See also parity. parity checking A check mechanism applied to a character or a series of characters that uses the addition of extra or redundant parity bits. Parity checking is useful for a variety of purposes, including asynchronous communications and computer memory coordination. See also parity. parity error A mismatch in parity bits that indicates an error in transmitted data. See also parity. partition 1. A portion of a hard disk that the operating system treats as if it were a   password separate drive. In Windows, a hard disk can be divided into several partitions. A primary partition, generally assigned the drive letter C, might contain files that start the computer. You could also create a non- Windows partition and use it for a different operating system. In Novell NetWare, a server must have a NetWare partition defined in order to function as a part of the NetWare file system. Other partitions can manage file systems used by other operating systems. 2. In Novell NetWare, a grouping or collection of objects in the Novell Directory Services (NDS) database. Each partition consists of a container object, all the objects in it, and data about all those objects. Partitions do not include any information about the file system or the directories or files contained there. The data in a partition is also known as a replica. See also disk mirroring; leaf object; replica; replication; Root object; volume. partition table 1. An area of storage on a hard disk that contains information about the partitions the disk contains. This information is usually recorded during the initial preparation of the hard disk before it is formatted. 2. In Novell Directory Services (NDS), a list on each server containing the NDS replicas. For each replica on the server, the partition table contains the partition name, type, time stamp, and partition state. passive hub A device used in some networks to split a transmission signal, allowing additional hubs to be added to the network, sometimes at the expense of distance. See also active hub. passive termination A method used to terminate a Small Computer System Interface (SCSI) chain of devices. Passive termination is a simple termination method that works best with four or fewer devices on a SCSI daisy chain. See also active termination; forced perfect termination; Small Computer System Interface. pass-through authentication In Microsoft Windows NT, an authentication method used when the user account must be authenticated, but the computer used for the logon is neither the domain controller for the account nor the computer on which the user account is defined. In such a case, the computer used for the logon passes the logon information through to the domain controller where the user account is defined. See also domain; domain controller; user account. password A security method that identifies a specific, authorized user of a computer system, a network, or a resource by a unique string of characters. In general, passwords should be a mixture of upper- and lowercase letters and numbers and should be more than six characters. Here are some general guidelines: _ Passwords should be kept secret and changed frequently. The worst passwords are the obvious ones: people’s names or initials, place names, phone numbers,   Password Authentication Protocol birth dates, and anything to do with computers or Star Trek. There are a limited number of words in the English language, and it is easy for a computer to try them all relatively quickly. _ Change all passwords every 90 days, and change those associated with high-security privileges every month. Some network operating systems require that passwords expire even more frequently. For example, in NetWare 5, passwords expire after 40 days by default. _ Some systems provide default passwords, such as MANAGER, SERVICE, or GUEST, as part of the installation process. These default passwords should be changed immediately. _ Limit concurrent sessions to one per system. _ Do not allow more than two or three invalid password attempts before disconnecting. _ Do not allow generic accounts. _ Promptly remove the accounts of transferred or terminated people, as well as all unused accounts. _ Review the security log files periodically. See also authentication; Challenge- Handshake Authentication Protocol; Password Authentication Protocol. Password Authentication Protocol Abbreviated PAP. A security protocol that requires a user to enter a user name and password before gaining access to a secure server. See also Challenge-Handshake Authentication Protocol. password encryption In certain operating systems, the password you enter to gain access to the system is not stored as ordinary text, but is encrypted, and this encrypted form is compared against the encrypted password stored on the server. If the two match, the logon continues; if not, the logon attempt is rejected. See also Challenge-Handshake Authentication Protocol; Password Authentication Protocol. password protection The use of one or more passwords to prevent unauthorized access to computer systems. patch panel A panel, usually located in a wiring closet, that contains rows of telephone- type modular jacks. A patch panel allows the network administrator to connect, disconnect, move, and test network devices by changing these connections. path The complete location of a directory or a file in the file system. Also called pathname or directory path. See also Universal Naming Convention. pathname See path. Payload Type Identifier Abbreviated PTI. In an Asynchronous Transfer Mode (ATM) cell, a field contained in the 5-byte cell header that defines the type of information in the payload area, including user, network, and management information. See also Cell Loss Priority; Header Error Control; Virtual Channel Identifier; Virtual Path Identifier. PB See petabyte.   PC-DOS PBP See Packet Burst Protocol. PBX See private branch exchange. PC98 A personal computer design guide for 1998–99 from Intel, Microsoft, and others, covering the PC system, bus, and devices, including guidelines for various types of mobile PCs. PC98 also describes requirements for manageability, remote boot support, and specifications for 1394. The basic PC98 should have no ISA (Industry Standard Architecture) devices; PC99 is expected to do away with ISA slots as well. See also Advanced Configuration and Power Interface; Web-Based Enterprise Management; Wired for Management; Zero Administration for Windows. PC-based router A router, such as the NetWare Multiprotocol Router or Multiprotocol Router Plus, that operates on a standard Intel-based personal computer. PC Card A term that describes plug-in cards that conform to the PCMCIA (Personal Computer Memory Card International Association) standard. A PC Card is about the size of a credit card and uses a 68- pin connector with longer power and ground pins that will always engage before the signal pins engage. Several versions of the standard have been approved by PCMCIA: _ Type I The thinnest PC Card, only 3.3 millimeters (0.13 inch) thick, used for memory enhancements, including dynamic RAM, static RAM, and flash memory. _ Type II A card used for modems or network interface cards, 5 millimeters (0.2 inch) thick; may also hold a Type I card. _ Type III A 10.5 millimeter (0.4 inch) card, used for mini-hard disks and other devices that need more space, including wireless network interface cards; may also hold two Type I or Type II cards. In theory, each PC Card adapter can support 16 PC Card sockets (if there is enough space), and as many as 255 adapters can be installed in a PC that follows the PCMCIA standard; in other words, PCMCIA allows a maximum of 4080 PC Cards on one computer. Most PC Card devices are modems, Ethernet and token-ring network adapters, dynamic RAM, and flash memory cards, although mini-hard disks, wireless LAN adapters, and SCSI adapters are also available. See also PC Memory Card International Association. PC Card slot An opening in the case of a portable computer, intended to receive a PC Card; also known as a PCMCIA slot. PC Connection, Inc. A direct marketer of PCs, peripherals, accessories, and networking products to the home, government, business, and educational markets. PC Connection was one of the first companies to provide overnight delivery of products and toll-free technical support. For more information on PC Connection, see www.pcconnection.com. PC-DOS See Personal Computer Disk Operating System.   PCI local bus PCI local bus See Peripheral Component Interconnect local bus. PCI-X A revision to the PCI standard proposed by IBM, Hewlett-Packard, and Compaq that increases the bus width to 64 bits, the bus speed to 133MHz, and the maximum throughput to 1GB per second. This revision is aimed at future workstation and server design. Several vendors also offer hot-plug PCI slots that allow you to replace a failed component without a system reboot. See also 1394; local bus; Peripheral Component Interconnect local bus; Plug and Play; Universal Serial Bus. PCMCIA See PC Memory Card International Association. PCMCIA slot See PC Card slot. PC Memory Card International Association Abbreviated PCMCIA. A nonprofit association, formed in 1989, with more than 320 members in the computer and electronics industries, that developed a standard for credit-card-sized, plug-in adapters designed for portable computers. PCONSOLE A Novell NetWare workstation utility used to set up and manage print queues and print servers on the network. See also NetWare Administrator. PCR See Peak Cell Rate. PCS See Personal Communications Services. PC Service and Support Certified Professional A certification from Learning Tree designed for the technician. Courses and exams cover basic PC assembly and troubleshooting, installation and configuration of operating systems and peripherals, and optimization of networks. PDA See personal digital assistant. PDF See Portable Document Format. PDN See private data network; public data network. Peak Cell Rate Abbreviated PCR. In Asynchronous Transfer Mode (ATM), the maximum cell rate that a specific data source on the network can sustain. See also Asynchronous Transfer Mode; Available Cell Rate; Minimum Cell Rate; Sustainable Cell Rate. peer-to-peer architecture A network architecture in which two or more nodes can communicate with each other directly, without the need for any intermediary devices. In a peer-to-peer system, a node can be both a client and a server. See also peer-to-peer network. peer-to-peer network A LAN in which drives, files, and printers on each PC can be available to every other PC on the network, eliminating the need for a dedicated file server. Each PC can still run local applications. Peer-to-peer networks introduce their own system management problems, including administration and responsibility for system backup, reliability, and security. Peerto- peer systems are often used in relatively   Pentium Pro small networks, with two to ten users, and can be based on MS-DOS, Windows, or Unix. Performance is not usually as good on a peer-to-peer network as under the client/ server model, especially under heavy traffic loads. See also client/server architecture. PEM See Privacy Enhanced Mail. Pentium A family of microprocessors introduced by Intel in 1993. The Pentium represents the continuing evolution of the 80486 family of microprocessors and adds several notable features, including instruction code and data caches and a built-in floating-point processor and memory management unit. It also has a superscalar design and dual pipelining, which allow the Pentium to execute more than one instruction per clock cycle, a 32-bit address bus, and a 64-bit data bus. Available in a range of clock speeds, from 60MHz all the way up to 233MHz, the Pentium is equivalent to 3.1 million transistors, more than twice that of the 80486. See also Intel Corporation; Pentium II; Pentium III; Pentium Pro. Pentium II A family of microprocessors from Intel. The Pentium II represents the continuing evolution of the Pentium family of microprocessors and adds several notable features, including integrated L1/L2 caches of up to 2MB that can be accessed at the full clock speed and a built-in floatingpoint processor and memory management unit. It also has a superscalar design and dual pipelining, which allow the Pentium II to execute more than one instruction per clock cycle. Available in a range of clock speeds, from 233MHz all the way up to 450MHz, the Pentium II can use a 100MHz system bus and is equivalent to 7.5 million transistors, more than twice that of the Pentium. See also Intel Corporation; Pentium; Pentium Pro. Pentium III A family of microprocessors from Intel. The Pentium III represents the continuing evolution of the Pentium family of microprocessors and adds several notable features, including 50 new floatingpoint instructions and 8 new registers to speed up floating-point calculations in scientific and engineering calculations, along with 12 new multimedia instructions to increase MPEG-2 performance and speech recognition. The most controversial new feature is the processor serial number, designed to increase network and online shopping security, but feared by many as a threat to privacy. Available in a whole range of clock speeds, initially from 450MHz to 500MHz versions, the Pentium III can use the Pentium II 100MHz system bus and is equivalent to 9.5 million transistors. See also Intel Corporation; Pentium; Pentium II; Streaming SIMD Extensions. Pentium Pro A family of microprocessors introduced by Intel in 1995. The Pentium Pro is optimized for the execution of 32-bit software and is available with clock speeds from 150 to 200MHz. With a 32-bit data bus running at 60 or 66MHz, it supports superscalar   PeopleSoft, Inc. architecture and pipelines and contains the equivalent of 5.5 million transistors. Dynamic execution (a combination of branch prediction and speculative execution) allows the processor to anticipate and schedule the next instructions for execution. Pentium Pro offers up to 1MB of Level 2 cache that runs at the same speed as the processor. See also Intel Corporation; Pentium; Pentium II. PeopleSoft, Inc. A major developer of large enterprise resource planning (ERP) applications for finance, materials and inventory management, distribution, human resources, and manufacturing, all within a single computing environment. For more information on PeopleSoft, see www.peoplesoft.com. Performance Monitor In Microsoft Windows NT, a network and server monitoring tool that displays resource use for selected system-level components; you can also use Performance Monitor to troubleshoot performance problems and assess hardware upgrade requirements. Information can be logged to a file for later analysis. period The . character; pronounced “dot.” Used to indicate the name of the current directory in a pathname and to separate the elements in a domain name, as in www.sybex.com . period and double-period directories In a hierarchical directory system, a shorthand way of referring to directories. The period (.) represents the current directory, and the double period (..) represents the directory immediately above the current directory. See also parent directory; root directory. Peripheral Component Interconnect local bus Abbreviated PCI local bus. A specification introduced by Intel in 1992 for a local bus that allows up to 10 PCIcompliant expansion cards to be plugged into the computer. One of these expansion cards must be the PCI controller card, but the others can include a video card, network interface card, SCSI interface, or any other basic function. The PCI controller exchanges information with the computer’s processor, either 32 or 64 bits at a time, and allows intelligent PCI adapters to perform certain tasks concurrently with the main processor by using bus-mastering techniques. PCI is compatible with ISA (Industry Standard Architecture), EISA (Extended Industry Standard Architecture), and MCA (Microchannel Architecture) expansion buses for backward compatibility with older technologies. PCI can operate at a bus speed of 32MHz and can manage a maximum throughput of 132MBps with a 32-bit data path or a rate of 264MBps with a 64-bit data path. See also 1394; local bus; PCI-X; Plug and Play; Universal Serial Bus. Perl Acronym formed from Practical Extraction and Report Language. Perl is an interpreted programming language developed by Larry Wall, used to manipulate text, files, and processes and to print reports on the extracted information.   Personal Computer Disk Operating System Perl is rapidly becoming the system administrator’s answer to all those problems that a C program does not seem to fit. It does not have the arbitrary limitations of other languages, so lines can be of any length, arrays can be of any size, variable names can be as long as you care to make them, and binary data does not cause problems. permanent swap file A swap file that, once created, is used over and over again. This file is used in virtual memory operations, in which hard-disk space is used in place of RAM. See also temporary swap file. permanent virtual circuit Abbreviated PVC. A permanent communications circuit, created and maintained even when no data is being transmitted. A PVC has no setup overhead and gives improved performance for periodic transmissions that require an immediate connection. Packets are transferred in order over a specific path and arrive at their destination in the same order. See also switched virtual circuit. permissions In a network or multiuser environment, the ability of a user to access certain system resources, including files and directories. Permissions are based on the rights given to user accounts by the system administrator. See also rights. permuted index A special kind of index used in several of the Unix system manuals. Many of the Unix manuals treat each command on a separate page, and these pages are not numbered continuously; they are numbered only within each command. This makes it easy to add or remove pages as requirements change, but it can make it difficult to find specific information. The permuted index is the solution. The permuted index has three columns. The central column, where you start your search, is in alphabetic order. The column to the right lists the command that performs the function and the section number in the man pages where you will find a detailed description, and the column on the left contains additional keywords to help confirm that you have found the correct entry. See also man pages. Personal Communications Services Abbreviated PCS. A digital wireless communications technology that includes voice, data, and video. PCS competes with the traditional analog cellular phone system, but PCS’s digital technology can provide clearer voice quality, better security through encryption, and lower costs, as well as additional services such as messaging, voice mail, and caller ID. Personal Computer Disk Operating System Abbreviated PC-DOS. The version of the DOS operating system supplied with PCs made by IBM. PC-DOS and MS-DOS began as virtually identical operating systems, with only a few minor differences in device driver names and file sizes, but after the release of DOS 6 (MS-DOS 6.2 and PC-DOS 6.1), the two grew much further apart. See also Microsoft Disk Operating System.   personal digital assistant personal digital assistant Abbreviated PDA. A tiny, pen-based, battery-powered computer that combines personal organization software with fax and e-mail facilities into a unit that fits into your pocket. PDAs are available from several manufacturers, including Apple’s Newton, and others from 3Com, Casio, Tandy, Toshiba, Motorola, Sharp, Sony, GRiD, and AT&T. Personal Information Manager Abbreviated PIM. A multipurpose software package that combines a word processor, database, and other accessory modules to allow the user to manipulate data in a less structured way than required by conventional programs. A PIM can store notes, memos, names and addresses, appointments, and to-do lists, and it may be part of the software used in a PDA (personal digital assistant). Personal NetWare Novell’s peer-topeer network replacement for NetWare Lite, released in 1994, that provides MS-DOS and Microsoft Windows users with the ability to share files, printers, CD-ROMs, and other resources, as well as run standard network applications. Other features include simplified network administration, increased security, and a single login so that users can view or access all network resources at once. Personal NetWare can manage a maximum of 50 workstations per server and a maximum of 50 servers on each network, giving a maximum of 2500 nodes per network. personalized menus In Microsoft Windows 2000, a feature that hides infrequently used menu selections. See also cascading menus. peta- Abbreviated P. A metric system prefix for one quadrillion, or 10 15 . In computing, based on the binary system, peta has the value of 1,125,899,906,842,624, or the power of 2 (2 50 ) closest to 1 quadrillion. petabyte Abbreviated PB. Usually 1,125,899,906,842,624 bytes (2 50 ), but may also refer to 1 quadrillion bytes (10 15 ). PGP See Pretty Good Privacy. physical address See hardware address. physical device An item of hardware, such as a disk drive or a tape drive, that is physically separate from other devices. physical drive A real drive in the computer that you can see or touch, as opposed to a conceptual or logical drive. One physical drive may be divided into several logical drives, which are parts of the hard disk that function as if they were separate disk drives. See also partition; volume. physical layer The first and lowest of the seven layers in the OSI Reference Model for computer-to-computer communications. The physical layer defines the physical, electrical, mechanical, and functional procedures used to connect the equipment. See also OSI Reference Model.   pipeline physical unit Abbreviated PU. The name used in IBM’s Systems Network Architecture (SNA) to indicate a physical device and its associated resources within the network. See also logical unit. PIM See Personal Information Manager. pin-compatible A description of a chip or other electronic component with connecting pins exactly equivalent to the connecting pins used by a different device. With a pin-compatible chip, you can easily upgrade a system by replacing the older chip with the newer version. See also plug-compatible. Ping Acronym formed from packet internet groper. A TCP/IP command used to test for network connectivity by transmitting a special ICMP (Internet Control Message Protocol) diagnostic packet to a specific node on the network, forcing the node to acknowledge that the packet reached the correct destination. If the node responds, the link is operational; if not, something is wrong. The word ping is often used as a verb, as in “ping that workstation to see if it’s alive.” Ping is designed for network testing, troubleshooting, and measurement, and because of the large load it can impose on a busy, working network, it should not be used during normal operations, unless the system administrator is tracing a specific problem on the network. ping of death A very large, specially constructed ICMP packet that violates the rules for packet size and content, designed to crash the receiving computer. See also brute-force attack; denial of service attack; dictionary attack. pinouts The configuration and purpose of each pin in a multipin connector. pipe A section of memory that can be used by a program or a command to pass information to a second command for processing. The information is stored in a firstin first-out basis and is not altered during transmission. A pipe is opened like a file and is read from or written to in the same way; pipes are also unidirectional in that one pipe is used to read data and another is used to write data. A special form of pipe, known as a named pipe, originated in the Unix operating system. A named pipe allows two processes to exchange information. This concept has been extended in several network operating systems as a method of interprocess communication, allowing data to be exchanged between applications running on networked computers. See also mailslots ; named pipe; semaphore; shared memory; socket. pipeline A mechanism used in microprocessors that speeds up the processing of instructions. The Intel Pentium processor features two pipelines, one for data and one for instructions, and can process two instructions per clock cycle. A processor with two or more pipelines is said to be superscalar. See also superscalar.   pipeline burst cache pipeline burst cache A secondary or L2 cache associated with a microprocessor that allows fast data-transfer rates. Pipeline burst cache requires RAM chips that can synchronize with the microprocessor’s clock. pipeline stall A microprocessor design error that leads to delays in the processing of an instruction. pipelining 1. In processor architecture, a method of fetching and decoding instructions that ensures that the processor never needs to wait; as soon as one instruction is executed, the next one is ready. 2. In parallel processing, the method used to pass instructions from one processing unit to another. See also parallel processing. PKI See Public Key Infrastructure. PKUNZIP A very popular file decompression utility available as shareware. PKUNZIP uncompresses files or archives created by PKZIP; both programs are usually available together. See also PKZIP; WinZip. PKZIP A very popular file compression utility available as shareware. PKZIP not only compresses files to save disk space or cut modem transmission times, but also combines compressed files to create compressed archives. See also PKUNZIP; WinZip. plaintext Text that has not been encrypted in any way and that can be intercepted and read easily while in transit; usually applied to an unencrypted password. See also cleartext. platform 1. An operating system environment, such as a NetWare platform or a Unix platform. 2. A computer system based on a specific microprocessor, such as an Intel-based platform or a PowerPC-based platform. platform-specific routers Routers based on a specific and proprietary hardware architecture, which is usually vendorspecific. player A small program launched or used by a Web browser to process a specific type of file that the browser cannot handle. A player is a program that deals with sound files. See also helper; plug-in; viewer. plenum cable Cable with a special Teflon coating designed for use in suspended ceilings, in inside walls, or between floors. The Teflon coating provides low flamespread and low, nontoxic smoke in the case of an accident. Plenum cables should meet the CMR (Communications Riser Cable) or CMP (Communications Plenum Cable) specifications of the National Electric Code and are often used for cable runs in airreturn areas. See also riser cable. Plug and Play Abbreviated PnP. A standard from Compaq, Microsoft, Intel, and Phoenix that defines techniques designed to make PC configuration simple and   Point-to-Point Protocol automatic. A user can plug in a new device, and the operating system will recognize it and configure it automatically when the system is next started. PnP adapters contain configuration information stored in nonvolatile memory, which includes vendor information, serial number, and checksum information. The PnP chipset allows each adapter to be isolated, one at a time, until all cards have been properly identified by the operating system. The PnP-compatible BIOS isolates and identifies PnP cards at boot time, and when you insert a new card, the BIOS performs an auto-configuration sequence enabling the new card with appropriate settings. New PCs with flash BIOS will be easy to upgrade so that they can take advantage of PnP; older systems with ROM-based BIOS will need a hardware change before they can take full advantage of PnP. See also Peripheral Component Interconnect local bus; Plug and Pray. Plug and Pray What most of us do when our Plug-and-Play systems do not work automatically. plug-compatible Any hardware device designed to work in exactly the same way as a device manufactured by a different company. For example, all external serial devices are plug-compatible, because you can replace one with another without changing the cabling or connector. See also pin-compatible. plug-in A small program you can link in to your Web browser to add a special capability not originally present or to recognize new file types or content. Plug-ins are available from a huge number of companies and are usually free. See also helper. PMMU See paged memory management unit. PNNI See Private Network-to-Network Interface. PnP See Plug and Play. POH See power-on hours. PointCast, Inc. The largest privately held media company on the Internet, providing online news to corporations. The PointCast broadcast receives more than 120 million hits a day and offers access to a collection of more than 600 leading business sources. Unlike the World Wide Web and other Internet applications, PointCast uses server push technology, in which the server automatically sends new data to a client without a specific request from that client. For more information on PointCast, see www.pointcast.com. point-to-point link A direct connection between two, and only two, locations or nodes. Point-to-Point Protocol Abbreviated PPP. A TCP/IP protocol used to transmit IP datagrams over serial lines and dial-up telephone point-to-point connections. PPP allows a PC to establish a temporary direct connection to the Internet via modem and appear to the host system as if it were an Ethernet port on the host’s network.   Point-to-Point Tunneling Protocol PPP provides router-to-router, host-torouter, and host-to-host connections and also provides an automatic method of assigning an IP address so that mobile users can connect to the network at any point. See also PPP Multilink; Point-to-Point Tunneling Protocol; Serial Line Internet Protocol. Point-to-Point Tunneling Protocol Abbreviated PPTP. A proprietary networking protocol proposed by Microsoft that supports virtual private networks, allowing remote users to access Windows NT Server systems across the Internet without compromising security. PPTP allows corporations to use public networks rather than leasing its own lines for wide area communications. See also encapsulation; Layer 2 Tunneling Protocol; PPP Multilink; tunneling. point of presence Abbreviated POP. A connection to the telephone company or to long-distance carrier services. polling A method of controlling the transmission sequence of devices on a shared circuit or multi-drop line by sending an inquiry to each device asking if it wants to transmit. If a device has data to send, it sends back an acknowledgment, and the transmission begins. Three methods are in common use: _ Roll-call A master station uses a polling list to locate the next node to poll. _ Hub A node polls the next node in sequence. _ Token-passing A token is passed to the next node in sequence. This node can transmit or pass the token to the next device. polymorphic virus A form of virus that can change its appearance to avoid detection. The virus encrypts itself using a special formula each time an infection occurs. Virus-detecting software uses special scanning techniques to find and remove polymorphic viruses. See also boot sector virus; file-infecting virus; macro virus; multipart virus; stealth virus, vaccine; virus. POP See point of presence; Post Office Protocol. port 1. To move a program or an operating system from one hardware platform to another. For example, Windows NT portability refers to the fact that the same operating system can run on both Intel and reduced instruction set computing (RISC) architectures. 2. The point at which a communications circuit terminates at a network, serial, or parallel interface card, usually identified by a specific port number or name. 3. A number used to identify a connection point to a specific Internet protocol. See also portable; port number. portability The ability to transfer an application or operating system from one vendor’s hardware to another, quickly and easily, without rewriting the software and without affecting its performance. This can be achieved in several ways: _ Write the program in a portable language, such as C, C++, or Java.   port number _ Use only standard programming language features. _ Use only standard libraries. _ Don’t make assumptions about word size or byte ordering. _ Use layers of software to distance the application from operating system or hardware dependencies. portable Describes the degree to which a program can be moved easily to different computing environments with a minimum number of changes. Applications written for the Unix operating system are often described as portable applications, as are Java applets. See also Java. portable computer Any computer light and small enough to be carried easily. There are two types: _ Laptop computers , which are small enough to be used in an airplane seat and powerful enough to run major operating systems and popular business applications. Extended battery life is making the laptop a serious alternative to the desktop system. _ Notebook computers , which are smaller than laptops and about the size of a textbook or student notebook, but still capable of running major applications. A notebook computer will easily fit into a briefcase. Major advances in battery life and the use of flash memory are part of the continuing development of portable computers. See also docking station; port replicator. Portable Document Format Abbreviated PDF. A file format standard developed by Adobe Systems and others for use in electronic documents. A file in this format usually has the filename extension of .PDF. Portable Operating System Interface Abbreviated POSIX. A collection of IEEE standards that defines a complete set of portable operating system services. POSIX is based on Unix services, but it can be implemented by many other operating systems. Each of the standards defines a specific aspect of an operating system, including such areas as system administration, system security, networking, and the user interface. When program or operating system service meets the appropriate POSIX standard, it is said to be POSIX-compliant. See also IEEE standards. port multiplier A concentrator that provides multiple connections to the network. port number The default identifier for a TCP/IP (Transmission Control Protocol/ Internet Protocol) or Internet process. For example, ftp (File Transport Protocol), HTML (HyperText Markup Language), and Telnet are all available at preassigned unique port numbers so that the computer knows how to respond when it is contacted on a specific port; Web servers use port 80, and SMTP (Simple Mail Transfer Protocol) e-mail is always delivered to port 25. You can override these defaults by specifying different values in a URL, but whether they will work depends on the configuration on the target system.   port replicator A total of 65,535 port numbers are available for use with TCP, and the same number are available for UDP (User Datagram Protocol). See also port. port replicator A device containing standard computer ports used to avoid constantly connecting and disconnecting peripherals from a portable computer. A port replicator duplicates all your computer’s ports and may even add a Small Computer System Interface (SCSI) port or a second Universal Serial Bus port. The external monitor, full-sized keyboard, and mouse you use in the office are connected to the port replicator; when it is time to take the portable computer on the road, you simply unplug the port replicator, leaving everything attached to the replicator for your return. See also docking station. portal A large Web site that acts as a gateway to the Internet and may also offer search facilities, free e-mail, online chat, instant messaging, as well as other services, including hard news, sports, and personal finance. Portals make money by selling advertising space. POSIX See Portable Operating System Interface. POST See power-on self test. post An individual article or e-mail message sent to a Usenet newsgroup or mailing list, rather than a message sent to an individual. See also posting; Usenet. Postal Telephone and Telegraph Abbreviated PTT. The official government body that administers and manages the telecommunications systems in many European countries. posting Sending an article or an e-mail message to a Usenet newsgroup. See also post; Usenet. Post Office Protocol Abbreviated POP. An Internet mail server protocol that also provides an incoming mail storage mechanism. POP works with Simple Mail Transfer Protocol (SMTP), which actually moves the e-mail from one system to another, and the latest version of the standard is POP3. When a client connects to a POP3 server, all the messages addressed to that client are downloaded; there is no ability to download messages selectively. Once the messages are downloaded, the user can delete or modify messages without further interaction with the server. In some locations, POP3 is being replaced by another standard, Internet Mail Access Protocol (IMAP) version 4. See also Internet Mail Access Protocol; Simple Mail Transfer Protocol. power conditioning The use of protective and conditioning devices to filter out   power supply power surges and spikes and ensure clean power. There are three main types of power-conditioning devices: _ Suppression , which protects against sudden destructive transient voltages. _ Regulation, which modifies the power waveform back to a clean sine wave. A UPS (uninterruptible power supply) is a common form of voltage regulator. It may be online, actively modifying the power, or offline and available only after the line voltage drops below a certain level. _ Isolation, which protects against noise. These types of devices are often expensive. Because power conditioning is expensive, usually only the servers or hosts in a network are protected. Surge suppressors may be used with workstations or other important network nodes, such as bridges or routers. See also blackout; brownout; power surge; spike; surge. Power Mac A series of computers from Apple Computer, Inc., based on the Power- PC chip. Although Power Macs run on the PowerPC chip rather than on the traditional Motorola chips, they run a version of the Macintosh operating system and look and feel just like 680x0-based computers. They can also run MS-DOS and Windows software under emulation. power-on hours Abbreviated POH. A cumulative count of the hours since the last time the system was started. See also mean time between failures. power-on self test Abbreviated POST. A set of diagnostic programs loaded from ROM before any attempt is made to load the operating system; designed to ensure that the major system components are present and operating. If a problem is found, the POST firmware displays an error message on the screen, sometimes with a diagnostic code number indicating the type of fault. PowerPC A family of RISC-based, superscalar microprocessors jointly developed by Apple, Motorola, and IBM, with a 32-bit address bus and a 64-bit data bus. _ The 601 houses 2.8 million transistors, runs at 110MHz, and is designed for use in high-performance, low-cost PCs. _ The 66MHz 602 is targeted at the consumer electronics and entry-level computer markets. _ The low-wattage 603e is aimed at batterypowered computers. _ The 604 is for high-end PCs and workstations. _ The 64-bit 620 is available in a 133MHz version capable of executing four instruction per clock cycle and is designed for servers and high-performance applications. _ The 750 (also known as the G3) is available in a range of processors running from 333 to 400MHz with an integrated L2 cache of 1MB and is equivalent to 6.35 million transistors. power supply A part of the computer that converts the power from a wall outlet into the lower voltages, typically 5 to 12 volts DC (direct current), required internally in the computer. PC power supplies are usually rated in watts, ranging from 90 to 300 watts. If the power supply in a computer fails, nothing works—not even the fan.   power surge power surge A sudden, brief, and often destructive increase in line voltage. A power surge may be caused by an electrical appliance, such as a photocopier or elevator, or by power being reapplied after an outage. See also power conditioning; surge; surge suppressor. PPP See Point-to-Point Protocol. PPP Multilink An extension to the Pointto- Point Protocol that can provide bandwidth on demand by combining multiple links between two systems; a process also known as bonding . PPP Multilink provides the negotiation features and protocols that allow systems to indicate that they can bond. The links can be of different types and different speeds. See also inverse multiplexing; Point-to- Point Protocol. PPTP See Point-to-Point Tunneling Protocol. preemptive multitasking A form of multitasking in which the operating system executes an application for a specific period of time, according to its assigned priority. At that time, it is preempted, and another task is given access to the CPU for its allocated time. Although an application can give up control before its time is up, such as during input/output waits, no task is ever allowed to execute for longer than its allotted time period. See also cooperative multitasking; timeslice multitasking. presentation layer The sixth of seven layers of the OSI Reference Model for computer-to-computer communications. The presentation layer defines the way in which data is formatted, presented, converted, and encoded. See also OSI Reference Model. Pretty Good Privacy Abbreviated PGP. A popular public-key encryption and digital certificate program, originally written by Phil Zimmermann, available at no charge from certain Internet sites. PGP uses Diffie-Hellman public-key algorithms, is available for Microsoft Windows and Macintosh platforms, and works with most popular messaging applications such as Microsoft Exchange, Eudora, and Claris Emailer for the Macintosh. See also Privacy Enhanced Mail; RSA Data Security. PRI See Primary Rate ISDN. primary domain controller In a Microsoft Windows NT domain, a computer running Windows NT Server that authenticates domain logons and manages the directory database for the domain. All changes to all accounts in the domain are automatically tracked and sent to the primary domain controller. There can be only one primary domain controller in any domain. See also backup domain controller; domain; domain controller. primary key See key. primary member One of two members of a mirror set. The primary member contains the original data; the shadow member contains the copy.   Priority Access Control Enabled See also disk mirroring; shadow member. Primary Rate ISDN Abbreviated PRI. An ISDN (Integrated Services Digital Network) service that provides 23 B (bearer) channels, capable of speeds of 64Kbps, and one D (data) channel, also capable of 64Kbps. The combined capacity of 1.544Mbps is equivalent to one T1 channel. See also Basic Rate ISDN. primary time server In Novell Net- Ware, a server that provides time information to secondary time servers and to workstations. A primary time server must synchronize time information with at least one other primary or reference time server. See also reference time server; secondary time server; single reference time server. PRINTCON A Novell NetWare workstation utility used to create, view, or modify print-job configurations on the network. Configuration options include the printer to be used, the print queue to process the job through, the print-device mode, the printer form number, and the number of copies. PRINTDEF A Novell NetWare workstation utility used to create, view, and modify printer definitions on the network. Printer Access Protocol Abbreviated PAP. The protocol used in AppleTalk networks to manage communications between computers and printers. See also AppleTalk. Printer Agent In Novell Distributed Print Services, a printer object that replaces the Print Queue, Printer, and Printer Server objects used in other Novell printing environments. printer emulation The ability of a printer to change modes so that it behaves like a printer from another manufacturer. For example, many dot-matrix printers offer an Epson printer emulation in addition to their own native mode. Most laser printers offer a Hewlett-Packard LaserJet emulation. See also emulator; terminal emulation. print queue A collection of documents waiting to be printed on a particular network printer. See also Novell Distributed Print Services. print server A server that handles printing for all users on the network. A print server collects print jobs sent by applications running on other networked PCs, places them in a print queue on the hard disk, and routes them to one or more printers attached to the print server. See also local printer; Novell Distributed Print Services. print spooler In an operating system or network operating system, the software that coordinates print jobs sent to a shared printer when that printer is busy. Each print job is stored in a separate file and is printed in turn when the printer becomes free. Priority Access Control Enabled Abbreviated PACE. A technology from 3Com Corporation designed to deliver on-time multimedia over switched Ethernet networks   Privacy Enhanced Mail with insufficient bandwidth and without the ability to prioritize traffic. This is accomplished by the use of the PACE-enabled switches that allow the switch port and end stations to take turns when transmitting. Privacy Enhanced Mail Abbreviated PEM. An e-mail standard that uses a patented RSA encryption scheme to provide a confidential method of authentication. PEM is little used due to the proprietary nature of the encryption scheme. See also Secure MIME; RSA. private automatic branch exchange Abbreviated PABX. An automatic telephone system that serves a particular location, such as an office, providing connections from one extension to another, as well as a set of connections to the external telephone network. Many PABXs handle computer data and may include X.25 connections to a packetswitched network. See also private branch exchange. private branch exchange Abbreviated PBX. A telephone system, usually owned by the customer, that serves a particular location, such as an office, providing connections from one extension to another, as well as a set of connections to the external telephone network. See also private automatic branch exchange. private data network Abbreviated PDN. A highly secure and very expensive network of leased lines built for a single user, usually a corporation. PDNs are used to transmit highly sensitive data such as banking and other financial information. The service provider guarantees a certain bandwidth will always be available, although some of that bandwidth may go unused during periods of light traffic. See also Virtual Private Network. private key One of two keys used in public key encryption. The user keeps the private key secret and uses it to encrypt digital signatures on outgoing messages and to decrypt incoming messages. See also public key encryption. private leased circuit A leased communications circuit, available 24 hours a day, 7 days a week, that connects a company’s premises with a remote site. Private Network-to-Network Interface Abbreviated PNNI. A dynamic linkstate routing protocol for Asynchronous Transfer Mode (ATM)-based networks. Any given ATM network may include ATM switches from several vendors; PNNI provides a routing protocol to communicate configuration information about the network to these groups of switches. See also Integrated-Private Network-to- Network Interface. privileged mode An operating mode supported in protected mode in Intel processors that allows the operating system and certain classes of device drivers to manipulate parts of the system, including memory and input/output ports. Applications cannot be executed in privileged mode.   programming language See also protected mode; privilege level; real mode. privilege level 1. Those rights granted to a user or a group of users by the network administrator that determine the functions the user can execute. Rights form an important component of network security and can include supervisor rights and read, write, erase, and modify rights, along with several others. 2. A form of protection built into Intel microprocessors. The Intel microprocessor architecture provides two broad classes of protection. One is the ability to separate tasks by giving each task a separate address space. The other mechanism operates within a task to protect the operating system and special processor registers from access by applications. Within a task, four privilege levels are defined. The innermost ring is assigned privilege level 0 (the highest, or most trusted, level), and the outermost ring is privilege level 3 (the lowest, or least privileged, level). Rings 1 and 2 are reserved for the operating system and operating system extensions; level 3 is available to applications. This protection is maintained by complex circuitry in the processor’s memory management unit. PRN In MS-DOS, Windows, and OS/2, the logical device name for a printer, usually the first parallel port, which is also known as LPT1. See also parallel port. process In a multitasking operating system, a program or a part of a program. All EXE and COM files execute as processes, and one process can run one or more other processes. See also session; thread. Professional Server Expert Abbreviated PSE. A certification from IBM that offers specialization in a specific network operating system, including Novell NetWare, OS/2 Warp Server, or Windows NT Server. See also Professional Server Specialist. Professional Server Specialist Abbreviated PSS. An introductory hardware certification from IBM designed to assess knowledge of IBM Netfinity and PC server architecture, installation, configuration, and management. See also Professional Server Expert. Profile object In Novell NetWare, a special Novell Directory Services (NDS) object used to assign the same login script to a group of users. A Profile login script is executed after the container login script has executed, but before the user login script. programming language A language used to write a program that the computer can execute. Almost 200 programming languages exist. An example is the popular C language, which is well suited to a variety of computing tasks. With C, programmers can write anything from a device driver, to an application, to an operating system. Certain kinds of tasks, particularly those involving artificial intelligence (LISP or Prolog), process control (Forth), or highly mathematical applications (Fortran and APL), can benefit from a more specific language.   Project Athena Programming languages are also divided into low-level languages, such as assembly language, and high-level languages, such as C, C++, and Java. See also assembly language; compiler; interpreter; machine language. Project Athena A Massachusetts Institute of Technology project that ran from 1983 to 1991, sponsored by MIT, DEC, and IBM, and developed the X Window system and Kerberos authentication, as well as several other important relational database and network-related systems. See also Kerberos; X Window. promiscuous mode A mode in which a network device or interface card captures all the packets on the network, not just those addressed to it specifically. Network analyzers work in promiscuous mode to monitor network traffic and to perform statistical analyses of the traffic. See also network analyzer; sniffer. propagation delay In communications, any delay between the time a signal enters the transmission channel and the time it is received. This delay is relatively small across a LAN, but can become considerable in satellite communications, in which the signal must travel from one earth station to the satellite and back to earth again. Unusually long delays may require the use of specialized hardware to ensure that the link is not broken prematurely. property 1. In Novell NetWare, a characteristic of an object in Novell Directory Services (NDS); also known as an attribute. User object properties include name, login name, password restrictions, e-mail address, and other related information. 2. In Microsoft Windows, a characteristic of an object or device, accessed via that object’s Properties dialog box. property rights In Novell NetWare, characteristics of an object in Novell Directory Services (NDS). Property rights are Add or Delete Self, Compare, Read, Supervisor, and Write. proprietary software Software developed in-house by a particular business or government agency and never made available commercially to the outside world. The operating systems used in certain portable computers and PDAs (personal digital assistants) may also be considered proprietary, because they are specific to one system and are not generally available anywhere else. protected mode In Intel processors, an operating state that supports advanced features. Protected mode in these processors provides hardware support for multitasking and virtual memory management, and it prevents programs from accessing blocks of memory that belong to other executing programs. In 16-bit protected mode, supported on 80286 and higher processors, the CPU can address a total of 16MB of memory directly; in 32-bit protected mode, supported on 80386 and higher processors, the CPU can address up to 4GB of memory.   PU Microsoft Windows NT, OS/2, and most versions on Unix running on Intel processors run in protected mode. See also real mode. protocol In networking and communications, the formal specification that defines the procedures to follow when transmitting and receiving data. Protocols define the format, timing, sequence, and error checking used on the network. See also communications protocol; OSI Reference Model; protocol stack. protocol analyzer A hardware or combined hardware and software product used to analyze the performance data of the network and to find and troubleshoot network problems. Protocol analyzers vary greatly in complexity. Some use dedicated hardware and can decode as many as 150 protocols; others convert an existing networked PC into a network-specific analyzer. See also network analyzer; sniffer. protocol converter A combined hardware and software product that converts from one network protocol to another; used when two dissimilar networks are connected. See also gateway. protocol stack The several layers of software that define the computer-to-computer or computer-to-network protocol. Several companies have developed important proprietary protocol stacks, including Novell NetWare’s IPX/SPX, but the trend these days is moving toward more open systems such as TCP/IP (Transmission Control Protocol/Internet Protocol). See also OSI Reference Model. protocol suite See protocol stack. protocol tunneling See tunneling. proxy server A software package running on a server positioned between an internal network and the Internet. The proxy server filters all outgoing connections so that they appear to be coming from the same machine, in an attempt to conceal the underlying internal network structure from any intruders. By disguising the real structure of the network, the proxy server makes it much more difficult for an intruder to mount a successful attack. A proxy server will also forward your requests to the Internet, intercept the response, and then forward the response to you at your network node. A system administrator can also regulate the external sites to which users can connect. See also dual-homed host; firewall. PSE See Professional Server Expert. PSS See Professional Server Specialist. PSTN See Public Switched Telephone Network. PTI See Payload Type Identifier. PTT See Postal Telephone and Telegraph. PU See physical unit.   public data network public data network Abbreviated PDN. Any government-owned or governmentcontrolled commercial packet-switched network, offering wide-area services to dataprocessing users. public key encryption An encryption scheme that uses two keys. In an e-mail transaction, the public key encrypts the data, and a corresponding private key decrypts the data. Because the private key is never transmitted or publicized, the encryption scheme is extremely secure. For digital signatures, the process is reversed; the sender uses the private key to create the digital signature, which can then be read by anyone who has access to the corresponding public key. This confirms that the message really is from the apparent sender. See also digital signature; private key. Public Key Infrastructure A proposal to provide a structure for verifying and authenticating users involved in transactions on the Internet or on corporate intranets and extranets. The proposal involves a set of trusted certificate authorities (CAs) who would publish a person’s public key and vouch for the authenticity of the data using a digital signature or certificate. Other CAs would then vouch for those CAs, and they in turn would be vouched for by other CAs. See also certificate authority; digital signature. public network Normal voice telephone systems; also called the direct distance dial (DDD) network. Public Switched Telephone Network A designation used by the ITU to describe the local telephone company. Public trustee In Novell NetWare, a special trustee, used only for trustee assignments. The Public trustee allows objects in Novell Directory Services (NDS) that do not have any other rights to have the rights granted to the Public trustee. This is similar to the way the user GUEST or the group EVERYONE worked in earlier versions of Novell NetWare. punch-down block A connecting device used for telephone lines; also known as a quick-connect block. The wires are pushed into metal teeth that strip the insulation away and make a good connection. push See server push. PVC See permanent virtual circuit.   quarter-inch cartridge Q Port Number Protocol 21 File Transfer Protocol 23 Telnet 25 Simple Mail Transfer Protocol 70 Gopher 79 Finger 80 Hypertext Transfer Protocol 119 Network News Transfer Protocol   TDM