Many applications are very demanding on network links and as a result data is often inefficiently transmitted. To combat this, switches provide clear paths to and from each node whilst maintaining multiple, simultaneous data transmissions, unlike bridges and routers that process packets on an individual 'first come first served' basis.

Switching technology has therefore superseded bridging as the process of forwarding data from one segment of the LAN to another using layers 2, 3 and 4 of the OSI model. Switches are hardware based and provide enhanced throughput performance, higher port density, lower per-port cost and greater flexibility.

A switch examines the data and uses the addressing information to determine its destination. It then creates a virtual link to the destination and transmits the packet there.

  • Layer 2 Switches (The Data-Link Layer) - Individual devices are identified using physical network addresses. As they are just sorting physical addresses layer 2 switches are very fast. However, they aren't very intelligent as the packet is not examined any further than to see its destination.
  • Layer 3 Switches (The Network Layer) - IP addresses are used to identify locations as well as identifying the actual device. Layer 3 switches are therefore more intelligent than layer 2 switches and incorporate routing functions.
  • Layer 4 Switches (The Transport Layer) - Layer 4 switches identify application protocols and manage communications based on the MAC address, IP address and the application to which a packet belongs. Priorities can be applied to mission-critical applications when switching at this level and security policies are enforced through management of protocols at port and user level.

Benefits of a managed switch environment

  • A network can be segmented into VLANs - Virtual Private Networks (possibly by departmental or hierarchical groups) providing improved manageability and security.
  • Troubleshooting is aided through port monitoring, packet error detection, and fault isolation.
  • Greater network insight and manageability allows for a pro-active, rather than a reactive approach to network management, reducing downtime, and assisting with future infrastructure planning.
  • Tracking and isolation of issues acts as a back-up should any problems occur.
  • Workload and trend analysis ensures that the network is running at it's optimum performance and also highlights the impact of any changes made.
  • Increased functionality allows for greater flexibility, scalability and the provision of capacity reservation.
  • Class of Service enables traffic prioritisation through packet classification at the network edge, utilising Layer 3 and 4 information, and provides bandwidth control.