![]() | Technology Reference Guide 7. Wide-Area Networks E. Frame Relay |
| What is it? Frame Relay is a relatively new wide-area network technology. It offers higher speeds than an X.25 network, and was designed to support the needs of LAN-to-LAN connections. The initial standards for Frame Relay networks were defined in 1988, but there are still some aspects of the technology that have not been fully standardized.
What does it do? Frame relay currently supports only Permanent Virtual Circuits (PVCs). This means that a pre-established path through the network is established and can only be changed manually. As standards mature for switched circuits, this capability is expected to become available. Frame relay interfaces with higher level protocols in much the same way as Token-Ring or Ethernet. Data from these upper level protocols is segmented into packets called frames. An address conversion between the upper-layer network address and the frame relay address format is performed, and the addressed frames are transmitted across the virtual circuit. Frame relay services are usually identified by a committed information rate (CIR), which specify the amount of bandwidth the customer expects to use. Users are allowed to exceed this rate on an occasional basis, when they have short-term needs to transfer large quantities of data. this is termed bursting. This amounts to the customer being able to temporarily increase their bandwidth, which is called dynamic bandwidth allocation. With a standard T-1 line, a customer was required to pay for the full 1.5 Mb/sec bandwidth all the time, whether they were using it or not. The data rates for Frame Relay extend from 56 Kb/sec to 1.3 Mb/sec. How is it used? Where do I get more information? |