The bridges

Contents

THE BRIDGES

The bridge is a network machine that has some intelligence, since it must store and forward the frames that arrive through its ports based on their content.

Therefore, they are small microcomputers that perform a series of basic operations on the network.

FEATURES

  • They allow to isolate traffic between network segments.        
  • They operate transparently at the network level and above.         
  • There is no conceptual limitation to the number of bridges in a network.         
  • They process the frames, which increases the delay.         
  • They use routing algorithms, which generate additional traffic on the network.         
  • They filter frames by physical address and by protocol.        
  • They are used in local area networks.

LEVEL OF OPERATION

Bridges operate at OSI level 2, that is, their basic unit of operation is the network frame (Figure 1). When a bridge must pass a frame from one segment of the network to another, it normally executes the following phases:

bridge scheme

Figure 1: Diagram of the operation of a bridge.

PHASES

  • Stores in memory the frame received by any port for later analysis.        
  • It checks the error control field of the frame in order to ensure the integrity of the frame. If it finds an error, it will remove the frame from the network, so that incomplete or erroneous frames will not cross the border of the network segment where the failure occurred.         
  • Some bridges are able to easily tweak the frame format (add or remove fields), in order to adapt it to the format of the target segment of the frame.         
  • The bridge forwards the frame if it determines that the recipient is on a network segment accessible through one of its ports.

TYPES OF BRIDGES

Bridges have traditionally been classified as transparent and non-transparent.

  • A transparent or spanning tree bridge is a bridge that does not require any configuration for its operation. It determines the forwarding of frames based on the events it observes for each of its ports.        
  • A non-transparent bridge requires the frame to carry information about how it should be forwarded. This type of bridge is more efficient in terms of performance, however, its compatibility in network connection is much lower, so, except in very specific applications, it is rarely used.

 A second classification for bridges addresses whether or not the two networks to be connected are in close proximity. According to this the bridges can be:

  • Local. A local bridge connects all network segments to the same machine (Figure 2).        
  • Remotes. A remote bridge is divided into two parts. Each one of them connects a network segment and the two parts are normally interconnected through the line of a WAN network; for example, a telephone or ISDN line (Figure 2).
Connection scheme of two networks

Figure 2: Diagram of connection of two local area networks with a local bridge or a remote bridge.

OPERATION OF A TRANSPARENT BRIDGE

Network interconnection example

Figure 3: Example of network interconnection using transparent bridges.

HUBS

  • A hub or concentrator is a device where the wiring connections of a network are concentrated, acting as a multiport repeater and concentrator.        
  • A hub essentially functions as a "bus". Just like a bus network, each connector located outside the hub is an output to that bus. When the hub listens to one of these "connectors", what it does is repeat that packet of bits in the others.

TYPES OF HUB

  • Passive hubs only repeat the signal on the network.        
  • The assets regenerate and amplify the signal. Hubs that regenerate the signal are better.
  • Smart hubs do what assets do but can also be managed. A network administrator can monitor each port and even get statistical information about it, they have better addressing features. All current hubs are smart.
  • There are two terminologies. In Ethernet it is known as a hub, something that in Token Ring is called MAU (Multistation Access Unit). They basically do the same thing but internally they work differently.
  • Solos are simply a box with connections, they usually stick to a wall from where they work, they are normal in small office and home connections where no expansion is needed, where the average number of users is 12.
  • The stackable ones can be mounted one on top of the other, and are connected to one another by means of a cable. By stacking one on top of the other they are almost modular and save companies from investing in the chassis involved in a modular hub.
  • The modular hubs consist of a series of cards that are connected to a chassis, from there they are interconnected and form part of the network. These are the highest point of connection management and capacity, so you only see them in truly industrial connections or telephone exchanges.

THE INTELLIGENT SWITCH OR HUB

The switch is a switch that has OSI layer 2 functions and therefore resembles a bridge in its operation.

Example of network interconnection through transparent bridging

Figure 3: Example of network interconnection using transparent bridges.

FEATURES

  • The switch is always local        
  • It connects network segments rather than networks, although at these lower levels it is not easy to tell one case from another.
  • The operating speed of the switch is higher than that of the bridge, which introduces longer delay times.
  • In a switch, the bandwidth of the network can be distributed in an appropriate way in each network segment or in each node, in a transparent way to the users.
  • Much of the commercial switch models are stackable, and therefore easily scalable, which gives them flexibility similar to repeaters, but with the functionality of bridges in terms of network traffic management.
  • Some very high performance switches are modularly connected to a very high speed bus (backplane) over which they switch.

IMPLEMENTATION TECHNOLOGY

  • On the fly or cut through technology. It does not wait to start forwarding the frame for it to arrive in its entirety. It is enough that the destination address field arrives for it to be retransmitted immediately. Therefore, the delay it produces is very small.        
  • Store-and-forward technology. It retains the entire frame in a buffer before forwarding it, thus, it is able to analyze the frame error information, in order to detect them. This supposes a greater delay in the retransmission, but gains in control capacity, providing the most complete and most adaptable solution to emerging technologies.
  • Fast routing bridge technology. It is traditional bridging technology, not a true switch. Its main drawback is that, since it has to analyze the network protocols, it generates very significant delays, one hundred times greater than in other technologies.
How a bridge can pass packets between two networks

How a bridge can pass packets between two networks.

ROUTERS

  • A Router is a device that allows information to be moved from a source to a destination, mainly in large networks.        
  • The difference between a router and a bridge is that they both work at different layers. Although they seem to do the same thing, first note that the router works at the network layer while the bridge works at the link layer. This causes the two devices to see different aspects of the information.
  • A router must do two things, determine the optimal route and transport the information in groups or packets as they are called, this process is called switching in routers but it tends to be more complex than this activity.

CHARACTERISTICS OF THE ROUTERS

  • They act at the network level.        
  • You can select one of several paths based on parameters such as transmission delay, congestion, etc.
  • These devices depend on the protocol used.
  • At the network level, the lifetime of a packet is controlled, the time required for a packet to go from one point to another in the internet of networks) will make the maximum size of this larger or smaller.