Bridges(Networking)

A bridge is a device that connects two or more local area networks, or two or more segments of
the same network. For example, suppose that your network includes both 10BaseT Ethernet and localTalk connections. You can use a bridge to connect these two networks so that they can
share information with each other.
In addition to connecting networks, bridges perform an additional, important function. They filter information so that network traffic intended for one portion of the network does not congest the rest of the network.

Bridges operate at the data link layer (Layer 2) of the OSI model. Bridges inspect incoming traffic and decide whether to forward or discard it. An Ethernet bridge, for example, inspects each incoming Ethernet frame - including the source and destination MAC addresses, and sometimes the frame size - in making individual forwarding decisions. Bridges serve a similar function as switches, that also operate at Layer 2. Traditional bridges, though, support one network boundary, whereas switches usually offer four or more hardware ports. Switches are sometimes called "multi-port bridges" for this reason.

When bridges were introduced in the 1980’s, they typically joined two homogeneous networks
(for example, two kinds of Ethernet networks). More recently it has become possible for bridges
to connect networks with different physical and data link level protocols. For example, you can
use a bridge to connect a LocalTalk network to an Ethernet network, or an Ethernet network to a TokenRing network.
Like switches, bridges learn the MAC addresses of all connected clients, servers, and peripherals,
and associate each address with a bridge port (network connection). When a bridge (or switch)
receives an incoming frame, it opens and reads its destination MAC address. If the port that will
receive the frame is different from the port connected to the sender, then the bridge forwards the frame to the destination port. If the port that will receive the frame is the same as the port
connected to the sender, the bridge drops the frame. (Since the bridge is by definition at the end
of the network segment, the receiving computer presumably intercepted a copy of the frame on its way to the bridge.) If the bridge cannot determine which port is associated with a destination
address, it passes the frame along to all ports.
Traditional bridges connect a single workgroup to another workgroup. More recently, however, manufacturers have produced multiport bridges. Multiport bridges allow network managers to connect more than two network segments to each other. Additionally, you can reconfigure or expand networks because simply by replacing one network interface card inside the multiport bridge with another (for example, adding a LocalTalk interface to a multiport Ethernet bridge).
Bridges generally inspect data link level information within a network signal—information like
the Ethernet or LocalTalk (MAC) destination address. They do not attend to network routing or
transport protocol information such as that carried within the TCP/IP, IPX/SPX, or AppleTalk
portions of the signal. However, bridges can be fitted with custom filters that enable them to read this information—including network routing or transport source address, packet size, or type of protocol—and reject or forward information based on it. Custom filters enable network managers to isolate particular areas of the network and control which protocols enter or leave each area.
For example, custom filters might allow requests from the Internet (outside the school district) not to enter certain areas of the network.
Bridges are relatively simple and efficient traffic regulators. However, in some networks they
have been replaced by their more powerful cousins—hubs, switches, and routers. Each of these
traffic regulators brings a unique set of strengths and weaknesses to its work:
• Hubs, switches, bridges, and routers can interconnect two different kinds of networks such as
10BaseT Ethernet and 100BaseTX.
• Hubs (unlike switches, bridges, and routers) do not filter traffic between the two networks.
• Switches have the unique capability to enable communicating devices momentarily to utilize
the full bandwidth (data carrying capacity) of the network.
• However, switches (and hubs) cannot accommodate the variety of protocols and cabling types
that bridges can.
• Routers are much more expensive and much more difficult to install and manage than hubs,
switches, or bridges, but they can filter and route information much more precisely. (We
discuss routers in more detail later in this chapter.)
When you purchase equipment, make sure you understand how each of these details affects your
network. Then work with your technical staff or network integrator to choose the best equipment
for each situation.
Because bridges (like switches) generally depend upon MAC addresses, we say in the parlance of
the OSI model that bridges are level 2 devices. You must purchase a bridge that is compatible
with your physical network and your data link protocols.

Purchase Considerations
When you consider purchase of a bridge, you should follow
these guidelines:

• Before you decide on your purchase, take a moment to clarify what
you wish to achieve (connecting a Macintosh LocalTalk lab to
Ethernet? connecting two Ethernet segments?). Then work with your
technical staff, or with manufacturers and consultants, to determine
your options. You can often use a hub, switch, or router in the same
places that you can use a bridge. Each device brings its unique set of
strengths and weaknesses to the job.
• Make sure that the bridge is compatible with your physical and data
link protocols.
• Purchase bridges from a known manufacturer whose support you
trust. Make sure the manufacturer provides a competitive warranty.
• Install your bridges in a room that is cool and free of dust, if
possible. Additionally, plug your bridges into an uninterruptible
power supply (UPS) to ensure that they receive clean power

Advantages of network bridges:

• Self-configuring
• Simple bridges are inexpensive
• Isolate collision domain
• Reduce the size of collision domain by microsegmentation in non-switched networks
• Transparent to protocols above the MAC layer
• Allows the introduction of management/performance information and access control
• LANs interconnected are separate, and physical constraints such as number of stations, repeaters and segment length don't apply
• Helps minimize bandwidth usage.

Disadvantages of network bridges:

• Does not limit the scope of broadcasts
• Does not scale to extremely large networks
• Buffering and processing introduces delays
• Bridges are more expensive than repeaters or hubs
• A complex network topology can pose a problem for transparent bridges. For example, multiple paths between transparent bridges and LANs can result in bridge loops. The spanning tree protocol helps to reduce problems with complex topologies.

Three types of bridges are used in networks:

• Transparent bridge Derives its name from the fact that the devices on the network are unaware of its existence. A transparent bridge does nothing except block or forward data based on the MAC address.

• Source route bridge Used in Token Ring networks. The source route bridge derives its name from the fact that the entire path that the packet is to take through the network is embedded within the packet.

• Translational bridge Used to convert one networking data format to another; for example, from Token Ring to Ethernet and vice versa.