|
S
Satellite Communications
Satellite
communications systems transmit signals from earth stations to satellites
located in space. Antennas located on the earth are pointed at the
geostationary satellite (also called geosynchronous) and microwave signals
are sent up to the satellite (uplink) where the satellite repeats the
signal and sends it down (downlink) to a remote receiving antenna. There
are three frequency ranges that satellite systems use, C band (4Ghz
downlink, 6Ghz uplink), Ku band (11Ghz downlink, 14Ghz uplink) and Ka band
(20Ghz downlink, 30Ghz uplink).
SCSI
SCSI stands for
Small Computer System Interface. It is a computer interface which
essentially expands the 8-bit PC bus so that you can connect peripherals
such as printers, hard drives and floppy drives and exchange data at a rate
of 5Mbps without stealing CPU cycles from the PC's main processor. Up to
seven devices can be daisy chained on a SCSI port. SCSI-2 is a 16-bit
implementation of the SCSI bus.
SDLC
SDLC stands for
Synchronous Data Link Control. SDLC is a subset of the HDLC (High-Level
Data Link Control) standard. It is the Link Layer protocol most often found
in SNA networks.
Secondary Station
In a
data communications network there are three types of stations. These terms
are normally used in reference to devices running HDLC or a subset of HDLC.
- Primary station. Sends command frames and
accepts responses (it would be considered a "master" node).
- Secondary station. Accepts commands and
sends responses (it would be considered a "slave" node).
- Combined station. Can send or accept
commands and respond appropriately.
Section Layer
The
section layer is the SONET layer responsible for reliable delivery of
frames between regenerators.
Segment Switching
A
switch used in computer networks may switch individual stations
(port-switching) or entire segments (Ethernet). Segment switching refers to
a switch that can switch entire Ethernet segments.
Self-Healing Operation
Self-healing
operation refers to the ability of a counter rotating ring technology such
as FDDI or DQDB to recover when a portion of the link fails. A major
advantage of counter rotation is recoverability. According to the standard,
every node has the ability to take over the function of a head or tail node
when a link fails. This allows the MAN to recover from failures in a manner
somewhat similar to FDDI, as shown in the DQDB Fault diagram. When an FDDI
link fails, shunts within the nodes to either side of the failure can
"shunt" the dual rings into a single larger ring. With DQDB, if
the link between node 5 and node 4 fails, node 4 becomes the head node for
bus A and the tail node for bus b, while node 3 assumes the opposite roles.
Semaphore
Semaphores are
bits used to synchronize the access to information. They are especially
used in interprocess communication.
Serial Communication
Serial
communication refers to the transmission of bits that occur serially in
time. Serial communication also implies that only a single communication
channel is used. The asynchronous transmission of characters using the
ASCII code set and a parity bit is a common method of serial communication.
Serial Interface
The
serial interface is that point on a computer which transforms a serial bit
stream into the parallel internal bit stream needed for a computer bus. The
serial port of a computer is the PC card (serial interface card) in the
computer that contains the electronics necessary to do this. Printers and
modems attach to the serial interface of a computer.
Service Access Point
The
Logical Link Control Protocol (LLC) contains service access points (SAPs)
which identify logical network connections being managed by the LLC data
link sublayer. Listed below are examples of SAP values and associated
protocols:
- 04 IBM SNA (station)
- 05 IBM SNA (group)
- 06 IP
- 80 XNS
- E0 Novell NetWare
- F0 NetBIOS
Session
A session is
analogous to a conversation between humans. Certain conventions in
conversation allow for the orderly and complete transfer of information
between the parties:
- They first agree to talk to one another.
- They (usually) don't talk at the same time.
- They divide the conversation into parts.
("Let me describe it to you, and you can tell me what you
think.")
- They end the conversation in an orderly
fashion. ("I'll talk to you later." "Okay. Bye.")
Similarly, the Session Layer provides the higher layers
with services that can be invoked to conduct sessions, including:
- establishing a session (separately from a
connection)
- the ability to conduct dialogs (prevent both
parties from transmitting data at the same time)
- the ability to manage activities (divide the
session into parts)
- the ability to end the session gracefully
(both ends agree to stop)
Shielded Twisted Pair
A
twisted pair line, as the name implies, consists of a pair of wires twisted
together. Because the wires are twisted together, electrical interference
tends to affect both wires equally, so it does not affect the difference in
the potential between the two wires. This makes twisted pair cable less
susceptible to signal loss than if it weren't twisted. Twisted pair lines
are suitable for bit rates up to 1Mbps over short distances (less than
100m) and lower bit rates over longer distances. More sophisticated driver
and receiver circuits enable similar or even higher rates to be achieved
over much longer distances. These are referred to as UTP or unshielded
twisted pair. UTP is used extensively in telephone networks and in many
data communications applications.
Shielded
twisted pair, like coaxial cable, is able to transfer data faster and over
greater distances than unshielded twisted pair because of the additional
protection from interference.
Signal Attenuation
All
transmission media have some resistance to the flow of electromagnetic
force. Over a great enough distance, the signal will be attenuated
(reduced) to the point that an unacceptable number of errors is introduced. Fiber optic transmission is far less
susceptible to attenuation than electrical transmission. Broadband is less
susceptible than baseband.
Signal Propagation Delay
Signal
propagation delay is the time it takes a signal to get from sending to
receiving device. This signal may be an electrical signal, a light signal
or a radio wave. Even at 186,000 miles per second, the time that it
takes for a signal to propagate from one end of the network to the other is
an important limiting factor.
Simple Mail Transfers Protocol
Unlike
other communications protocols, which use binary codes in structured
fields, SMTP uses plain English headers. SMTP defines a protocol and a set
of processes that use the protocol to transfer e-mail messages between
users' mailboxes. It does not define the programs used to store and
retrieve mail messages. In fact, although a basic mail reader program is
included with TCP/IP on virtually every operating system,
many different mail readers have been developed.
Simple Network Management Protocol
The
Simple Network Management Protocol is most often referred to as SNMP. SNMP
is designed to be simple. It does this three ways.
First, by reducing the development cost of the agent software, SNMP has
decreased the burden on vendors who wish to support the protocol. This
increases the acceptance of the SNMP. Second, SNMP is extensible because it
allows vendors to add their own network management functions. Third, it
separates the management architecture from the architecture of network
devices such as workstations and routers. This further widens the
multivendor acceptance and support for this protocol.
Simplex
Simplex mode or
simplex operation refers to a data transmission mode where bits are
transmitted in a single direction only. A simplex channel refers to a
channel that is capable of transmitting data in one direction only.
SIP
The internal SMDS
protocols are called SMDS Interface Protocol-1, -2 and -3 (SIP-1 through
-3). They are a subset of IEEE 802.6. SMDS uses cell relay at layer 1,
employing a cell with the same format as described for ATM; however, the
format of the 5-byte header is different. The phone companies defined the
SIP protocols for use.
Slave
In a
communications system that is hierarchical in arrangement, processes are
often spoken of as being either masters or slaves. The issue is one of
control rather than service. Master/slave relations often occur in cases
where one node has greater "intelligence" than the other, that
is, greater computing capacity. Master/slave relations occur in the IBM SNA
environment because of its hierarchical nature. The protocol used in
master/slave relations is generally the same for communication in either
direction, but the master node is in complete control. An example is the
relationship between a terminal and a 3270 cluster controller.
Sliding Window
Windowing
is also referred to as a sliding window. Windowing allows multiple PDUs to
be sent at one time. If a receiving station allocates seven input buffers
it can accept seven PDUs. To keep track of the acknowledged PDUs each is
labeled with a sequence number from 0 to 7. The receiving station
acknowledges receipt of a PDU by sending the number of the next PDU
expected. If two stations are sending and receiving data, two windows must
be maintained, one for transmit and one for receive. Acknowledgments and
data can be sent together; this is known as "piggybacking."
SLIP
Serial
communication over the telephone network has become much more important,
and with this change has come the need to extend TCP/IP into the serial
world of communications. Two protocols address this need. They are the
Serial Line Interface Protocol (SLIP) and the Point-to-Point Protocol
(PPP). The SLIP protocol consists of two special characters, ESC (escape)
and END. The END character (hex C0) marks the beginning and end of a SLIP
frame. The ESC character (hex DB) is used to indicate where the data
contains the ESC or END character so that the receiver will not interpret
those occurrences as delimiters.
SMTP
SMTP stands for
Simple Mail Transfer Protocol. Unlike other communications protocols, which
use binary codes in structured fields, SMTP uses plain English headers. SMTP
defines a protocol and a set of processes that use the protocol to transfer
e-mail messages between users' mailboxes. It does not define the programs
used to store and retrieve mail messages. In fact, although a basic mail
"reader" program is included with TCP/IP on virtually every
operating system, many different mail readers have
been developed. These include programs based on graphical user interfaces
such as Motif and OPEN LOOK. But they all use the SMTP service routines to
send and receive.
SNA
SNA stands for
Systems Network Architecture. It is IBM's network architecture for computer
networking. SNA was designed for transaction processing in mission-critical
applications, often involving services provided to customers. SNA networks
usually involve a large number of terminals communicating with a mainframe.
Typical transactions perform inquiries and update information in a
database. For example, a commercial bank might have a number of 3270-type
display units and printers in each of hundreds of branch offices which are
used to access a central database in the home office.
SNAP
SNAP stands for
Sub-Network Access Protocol. It is an LLC header extension. It consists of
five bytes of data, the first three bytes are referred to as the protocol
ID and the last two bytes are referred to as the Ethertype.
SNMP
SNMP stands for
Simple Network Management Protocol. It is based on the manager/agent model.
Its primary purpose is to allow the manager and the agents to communicate.
This protocol provides the structure for commands from the manager,
notifies the manager of significant events from the agent, and responds to
either the manager or agent. The original version of SNMP was derived from
the Simple Gateway Monitoring Protocol (SGMP) and was published in 1988.
Socket
A socket is an
API (Application Programming Interface) for interprocess communication
between a UNIX based system and a TCP/IP network. A socket consists of
either a TCP port and an IP address or a UDP port and an IP address.
Application programs use sockets to communicate across a network to a peer
process (TCP port).
SONET
SONET stands for
Synchronous Optical NETwork. SONET standardizes optical transmission. The
SONET standard defines a signal hierarchy. The basic building block is the
STS-1 51.84Mbps signal, chosen to accommodate a DS3 signal. The STS
designation refers to the interface for electrical signals. The optical
signal standards are correspondingly designated OC-1 (Optical Carrier-1),
OC-2, etc.
Source Address
The
source address is one of at least three addresses of the sending device.
The three addresses are the link address, the network address and the
process address. An example of a Link Layer address would be the NIC source
address. The packet or datagram address is found at the Network Layer. An
Internet address would be an example of a Network Layer address. Finally,
the process address is the address of the application sending the
information. FTP port 21 would be an example of a process address.
SPX
SPX stands for
Sequenced Packet eXchange. It provides Transport Layer functionality for
the Novell NetWare System.
SSAP
SSAP stands
Source Service Address Point. The SSAP is associated with the Logical Link
Control layer. It indicates the Network Layer process that the Link Layer
information is from.
Stackable Switch
A
stackable switch is a switch that can stack on top of another switch. The
connection between switch is normally found in the back of the switch and
the network connections in the front.
Star/Ring
A star/ring is a
topology found in Token Ring networks. In a star/ring topology, the network
is physically configured in a star configuration, with network device
connecting to a central hub (MAU). However, data in the network travels
logically in a ring, going from device to device.
Start Bit
Start
and a stop bits are used in asynchronous communications to delimit each
character transmitted from a terminal or a host.
Statistical Time Division
Multiplexing
Statistical
Time Division Multiplexer (STDM) is a type of multiplexer allows more
terminals to be attached to a circuit than the capacity of the circuit can
handle. It does this by taking into account that all devices do not
transmit constantly at their maximum rates. A STDM can be used instead of
an ordinary multiplexer when voice channels are not required.
Stop Bit
Start
and a stop bits are used in asynchronous communications to delimit each
character transmitted from a terminal or a host.
STP
STP stands for
shielded twisted pair. A twisted pair line, as the name implies, consists
of a pair of wires twisted together. Because the wires are twisted
together, electrical interference tends to affect both wires equally, so it
does not affect the difference in the potential between the two wires. This
makes twisted pair cable less susceptible to signal loss than if it weren't
twisted. Shielded twisted pair, like coaxial cable, is able to transfer
data faster and over greater distances than unshielded twisted pair because
of the additional protection from interference.
Stream/streams
Generically, a
stream of data is simply of series of data bits going across a physical
medium. STREAMS is an operating environment which
allows multiple communications protocols on a network. Applications running
on top of STREAMs can access the available protocols for communicating data
across a network.
Sub-Network Address
The
Sub-Network Access Protocol, or SNAP, is an LLC header extension. It
consists of five bytes of data, the first three bytes are referred to as
the protocol ID and the last two bytes are referred to as the Ethertype.
Subnet ID
An IP
network can have a subnetwork address that is a number, unique to the
network, assigned by the network administrator.
Subnet Mask
Internet
(IP) addresses must all be unique. The IP address contains a network
portion and a host portion. The host ID portion of the address must be
unique for the particular network. Every node on the network must know how
to tell which bits in the Internet address correspond to their physical
network, or subnet. This is accomplished through a "subnet mask"
which is set through the software in each node. If the user or system
administrator incorrectly sets the subnet mask, the node will not be able
to recognize its address in messages on the LAN and will not be able to
communicate. The subnet mask, which must be consistent throughout the
network, is a 32-bit hexadecimal word which "masks out" the node
address.
Subnetwork
Networks can be
classified according to the area over which they extend. A LAN can consist
of a few nodes up to several hundred but will typically be confined to a
few buildings within a few thousand meters of one another. It can consist
of subnetworks linked together in certain ways to form the larger, but
still local, network.
Subrate Facilities
Any
facility that operates at a data rate less than DS0 is called a subrate
facility. AT&T's Dataphone Digital Service (DDS) and British Telcom's
KiloStream are examples of subrate facilities. The first digital
telecommunications facilities made available to subscribers were based on
DSO in North America and equivalent
services in other parts of the world and were therefore subrate facilities.
We will discuss DDS and DDS II only. Other subrate facilities are very
similar.
Switched-56
Switched-56
(SW56) is another service provided by the telephone company. They are
digital and require a CSU/DSU combination to attach a router to the phone
line. The interface between the router and the CSU/DSU is typically a V.35
serial cable. Once configured, the switched-56 lines operate as any other
dial-out lines by dialing a distance switch-56 station. Switched-56 is
often used as a backup line for higher speed connections such as T1.
Switched Line
A
switched line is another term for the analog telephone network. It is
termed switched because each telephone call is switched through the network
to get to the final destination. Switched lines are used in data
communications with modems.
Synchronous Transmission
Synchronous
transmission is the opposite of asynchronous transmission. The modems must
first closely synchronize their internal timing circuits (usually by
transmitting a burst of bits of a feed length when the connection is
established). To transmit data, the sending modem puts a one or a zero on
the line every so often. The receiving modem samples the line on the same
timetable and transmits the condition of the line (one or zero) to the DTE.
They must stay in synchronization in order to communicate.
Sync Bits
Sync bits refer to the bits used in asynchronous or
synchronous transmission which are sent to acquire and maintain
synchronization between the sending and receiving devices. An example would
be the preamble of an Ethernet MAC frame.
|
