Ethernet practical guide
|
 |
On the face of it, networking can be somewhat intimidating. But it
doesn't have to be that way. This Practical Guide to Plug-and-Play Ethernet
Networks should answer any questions you have about how to get started
building your own local area network.
Frequently Asked Networking Questions
Why should I invest in an Ethernet networking solution?
Networking of any kind - Ethernet or not - is a good idea.
The low-tech alternative, commonly referred to as "sneakernet"
- walking around with floppy disks to share, print or transfer files
- is not only antiquated, it's very inefficient. Linking your company's
computers, printers, servers and other peripheral equipment on a common
system lets you and your co-workers share application programs, exchange
files and swap data quickly, easily and reliably. Networks make electronic
mail possible, they support server-based data back-ups, they provide
access to the internet - in short, networks represent a highly productive
way to do business that saves you considerable time and money. Plus,
you always have immediate access to mission-critical information, which
leads to better management and more efficient decision making - a distinct
competitive advantage.
What if I already use AppleTalk or Arcnet -
why should I switch to Ethernet?
Access methods such as 10 megabit-per-second (Mbps) Ethernet
- as well as 16 Mbps Token Ring and 100 Mbps FDDI - are much faster
than these proprietary solutions, letting you take full advantage of
the performance benefits available with your computing resources. Not
only that, these industry-standard protocols are widely used and supported,
opening up a whole new set of options for future growth and expansion.
What is Ethernet?
Ethernet, as defined by the Institute of Electrical and Electronic
Engineer's (IEEE) 802.3 standard, is an access method that determines
how data, in the form of electrical signals, is transmitted across a
physical cabling medium that connects computing devices. Ethernet conforms
to the carrier sense multiple access with collision detection (CSMA/CD)
network access protocol, which essentially allows any station on a network
to transmit data at any time. If the network is busy - that is, if another
station is already broadcasting - the transmitting station backs off
and waits a random length of time (measured in milliseconds) before
trying again. Once the network is clear, the station is free to send.
The IEEE 802.3 standard features several sub-specifications which define
the specific cabling used by that particular Ethernet application. For
instance, 10BASE-5 specifies the use of thick coaxial cabling; 10BASE-2
specifies thin coaxial cabling (like cable TV wire); and 10BASE-F, 10BASE-FB,
10BASE-FL and FOIRL all cover fiber optic cabling. Finally, the most
popular method - IEEE 802.3I 10BASE-T - specifies the use of unshielded
twisted pair data cabling with RJ45 data jacks for easy connections.
What do I need to build an Ethernet network?
At the very least, you need a network operating system that
facilitates resource sharing. The network operating system defines the
"language" used by the network to control input and output
operations. This capability can either be built into the operating system
(such as on Apple- and UNIX- based computers), or it can be an addition
to the primary operating system (such as NetWare is added to DOS or
MS Windows). In addition to the network operating system, you also need
network interface cards (NICs) for all networked devices (PC's, servers,
and others) to provide the appropriate connection to the LAN.
However, if you're just starting out, probably the most important piece
of equipment you'll need is a "hub" - a device that, as the
name implies, resides at the center of a network, where it provides
a common link between users. Hubs come in a variety of sizes for supporting
everything from small, low-density workgroups to large corporate-wide
networks. You simply buy a hub that has enough connections, or "ports,"
to suit your current needs. As your requirements grow, you can easily
add more hubs or other components such as bridges, routers or network
management to keep data moving and help you maintain control over the
system.
What's the benefit of using a hub?
Hubs offer a number of advantages. First, they implement networks
in a "physical star" topology - just like the telephone system
- in which all connections radiate out from a single location. This
arrangement provides a central point for network installation, maintenance
and management, allowing moves, additions and changes to be accomplished
quickly and easily. The physical star topology also ensures that a problem
on one device will not affect other users on the network. LED indicators
on the hub alert you to these faults, helping you immediately identify
and resolve problems on one link while the rest of the system continues
running. Finally, hubs operate over a building's structured cabling
system - in most cases, the unshielded twisted pair data cabling already
installed in your office.
What exactly is a "plug-and-play" solution?
"Plug-and-play" means just what it says: a hub that
requires little or no set up, preparation or configuration. Unshielded
twisted pair 10BASE-T hubs are a good example; they use simple RJ45
connectors. You just install the hub in the wiring closet or on a desktop,
plug in the connections (both at the hub and at the PC, server, printer
or other piece of equipment) and turn it on. That's it - you're up and
running. Networking doesn't get any easier.
What if my needs grow after I make my initial
investment?
Then you simply expand the network by adding new hubs to support
the additional load. Most hubs are designed to be "stackable"
or "expandable" or "modular" - all just different
ways to say that they can work together, seamlessly and transparently,
to support your growing needs.
If the network grows too big and performance suffers noticeably, you
can add other devices such as routers or bridges to divide, or segment,
the system to give it a performance boost. If only very specific areas
need the additional speed or bandwidth, you could upgrade to a new solution
featuring high-speed 100 Mbps switched Ethernet, extending these capabilities
to the overburdened workgroups. And if the network expands to the point
that monitoring performance becomes too difficult, you can add network
management to assist in the process. The key to this type of interoperability
between devices is standards compatibility - adherence to industry standards
that ensures everything speaks the same "language," or protocol.
By using standards-based solutions, you can grow your network indefinitely
- without sacrificing your original investments.
What is network management? And when would I
need it?
Network management is an all-encompassing term that describes
the ability to monitor network performance and activity, via software,
to detect, locate and fix problems that adversely affect the network.
A number of different management applications are available, offering
capabilities ranging from simple to extremely complex. At the very least,
a network management system should help you rapidly detect and isolate
network errors such as excessive collisions, bad packets and the like
- events that could indicate or lead to potentially larger problems.
Determining whether you need network management depends on your requirements.
How can I begin planning a solution that's
right for my company?
A variety of solutions are available for every size company.
A reputable reseller can come to your place of business, evaluate your
needs and recommend a solution that not only satisfies your existing
requirements, but can also grow to support your expanding business.
Glossary of Terms
Asynchronous Transfer Mode (ATM) - A type of switching
technology in which the switches are small, fixed-length cells containing
data.
Backbone - A segment of network that links several
individual workgroup or department LANs together in a single building.
It is also used to link several building LANs together in a campus environment.
Bridges - Bridges filter packets between LANs by making
a simple forward/don't forward decision on each packet they receive
from any of the networks to which they are connected.
Carrier Sense Multiple Access with Collision Detection
(CSMA/CD) - An element defined by the 802.3 specification.
It is an access method which is used by stations connected to an Ethernet
LAN. In this method each station contends for access to the shared medium.
Collision - When two stations try to send packets
at the same time. In Ethernet networks, collisions are considered normal
events and the CSMA/CD access method is designed to quickly restore
the network to normal activity after a collision occurs.
Ethernet - The most popular LAN technology in use
today.
Fiber Distributed Data Interface (FDDI) - LAN technology
that runs at 100 Mbps over fibre, a much higher data rate than Ethernet
or Token Ring. Originally FDDI networks required fiber optic cable,
but today they can be run on UTP as well.
IEEE 802.3 - An Ethernet specification commonly defined
by the Institute of Electrical and Electronic Engineers (IEEE). The
802.3 specification covers rules for configuring Ethernet LANs, the
types of media that can be used, and how the elements of the network
should interact.
Intelligent Hubs - Intelligent hubs are wiring concentrators
which can be monitored and managed by network operators.
LAN Internetwork - Connecting disparate and geographically
dispersed local area networks together to form an enterprise system.
Local Area Network (LAN) - A LAN is a high-speed communications
system designed to link computers and other data processing devices
together within a small geographic area such as a workgroup, department,
or a single floor of a multi-story building.
Manageable Hubs -Another definition for intelligent
hubs. Each of the ports on the managed hub can be configured, monitored,
and enabled or disabled by a network operator from a hub management
console.
Modular hubs -A modular hub starts with a chassis,
or card cage, with multiple card slots, each of which can accept a communications
card, or module. Each module acts like a stand-alone hub; when the communications
modules are placed in the card slots in the chassis, they connect to
a high-speed communications backplane that links them together so that
a station connected to a port on one module can easily communicate with
a station on another module.
Network Interface Card (NIC) - The physical connection
from the computer to the network is made by putting a network interface
card (NIC) inside the computer and connecting it to the shared cable.
Open Systems Interconnect Reference Model (OSI) -
A communications model developed by the International Standards Organization
(ISO) to define all of the services a LAN should provide. This model
defines seven layers, each of which provides a subset of all of the
LAN services. This layered approach allows small groups of related services
to be implemented in a modular fashion that makes designing network
software much more flexible.
Packet - In a shared media network, when one stations
wishes to send a message to another station it uses the network software
to put the message in an envelope. This envelope is called a packet.
Routers - Routers are more complex internetworking
devices and are also typically more expensive than bridges. They use
Network Layer Protocol Information within each packet to route it from
one LAN to another.
Shared Media - Shared media technology means that
all of the devices attached to the LAN share a single communications
medium, usually a coaxial, twisted pair or fiber optic cable.
Shielded twisted pair (STP)- Cable which comes with
a shielding around the cable to provide more protection against electromagnetic
interference (EMI).
Stackable hubs - Stackable hubs look and act like
stand-alone hubs except that several of them can be "stacked"
or connected together, usually by short lengths of cable. When they
are linked, they act like a modular hub in that they can be managed
as a single unit.
Stand-alone hubs - Single box-level products with
a number of ports. Stand-alone hubs usually include some method of linking
them to other stand-alone hubs - either by connecting them together
with a length of 10BASE-5 coaxial cable or cascading them using twisted
pair between individual ports on each hub.
10BASE-T - The specification for running Ethernet
on UTP. This stands for 10 Mbps, baseband signaling (the signaling method
used by Ethernet networks), over twisted pair cable.
10BASE-5 - An Ethernet specifications which uses a
thick coaxial cable. 10BASE-5 is seldom installed in new Ethernet networks
today.
10BASE-2 - An Ethernet specification that uses a thin
coaxial cable medium. 10BASE-2 is only used in very small office networks.
Token Ring - A major LAN technology in use today.
Token Ring rules are defined in the IEEE 802.5 specification. Like Ethernet,
the Token Ring protocol provides services at the Physical and Data Link
Layers of the OSI model. Token Ring networks can be run at two different
data rates, 4 Mbps or 16 Mbps.
Unshielded twisted pair (UTP)- UTP cable is similar
to telephone cable, but has somewhat more stringent specifications regarding
its susceptibility to outside EMI than common telephone wire. UTP is
used much more often than STP.
|
