Join Greg Sowell for an in-depth discussion in this video Hubs, bridges, and switches, part of Foundations of Networking: Network Media (WANs).
- Hubs, bridges and switches are the lower-level devices in a network, operating in layer one and two of the OSI model. Hubs are the grandfather of networking devices. Nothing more than a basic electrical repeater, hubs receive a signal from one port and forward that traffic at all other ports. In this fashion, they operate at layer one. One of the major weaknesses of a hub network, is that they utilize a single collision domain, running at half duplex. Only a single device can transmit at a time on the network.
If more than one transmission is sent, they collide. Hence the name "collision domain." Just like a group of people with walkie talkies. Only one person can talk at a time. This severely limits the size a network can grow to. Hubs are also a huge security risk, since they send traffic at all ports, any other host can intercept a private session. Bridges were introduced to fix these problems. The first bridge had two ports, and would be inserted between hubs. A bridge makes every port its own collision domain.
Meaning each port can send and receive simultaneously. Bridges were the first devices to use MAC address tables. Instead of blindly forwarding everything, bridges track the layer two MAC addresses of clients, and associate them with the ports they can be reached on. If a bridge receives a frame destined for a MAC address that is not in its table, the bridge will flood that frame at every other port, save for the source port. When the end device is found, the bridge adds that new MAC address to the table. With the MAC address logged, the bridge can quickly forward future frames to the correct port.
This greatly improved layer two network efficiency. Bridges live at layer two. Bridges are still used in modern networks, albeit few and far between. Software bridging is also used for spanning wireless links and using routed ports as layer two forwarding ports. Bridges used reduced instruction set computing, or RISC, and complex instruction set computing, or CISC processors. Evenually, switches were built, using ASICs, or application-specific integrated circuits, designed to forward frames.
Bridges gave way to switches. In effect, switches are hardware-based bridges. Switches have more ports and are much more efficient. As ASICs got cheaper, switches quickly increased port count and dropped in price. Today, a simple five-port switch can be had for $8. They also quickly began expanding on their capabilities. Switches are what make up the axis level of a network. They're responsible for aggregating user and server traffic. While hubs, bridges and switches are typically utilized on a LAN, understanding this hardware is critical to maintaining a robust connection between end users and the WAN.
He discusses different WAN technologies and features such as speeds, spans, and price points—including inexpensive options such as VPN. He then covers switches (the devices that connect computers in your building) and routers (devices that control the transmission of network data). Along the way, Greg shows how to build private connections, implement free networking over the Internet, build switch networks, and overlay-routed networks. He'll also introduce different routing protocols, such as OPSF link-state routing and distance-vector routing with RIPv2, EIGRP, and BGP.
Note that this course maps to domains 1 and 2 of the Microsoft Technology Associate (MTA) Networking Fundamentals certification exam (98-366).
- Understanding the technology: from dial-up to VPNs
- Working with hubs, bridges, and switches
- Ensuring hardware redundancy
- Using switching types and MAC tables
- Preventing bridge loops with STP
- Routing with routing tables
- Using NAT
- Securing your switches and routers
- Setting up firewalls
- Working with different routing protocols: RIPv2, OSPF, EIGRP, and more