Comparative performance of interleaved and non-interleaved pipelining in LAPD terminal adaptors

A comparison is made of the end-to-end delay performance of two service disciplines that a terminal adaptor (TA) can use to multiplex the pipelined synchronous native protocol frames arriving over the low-speed lines. In one scheme, the TA pipelines one synchronous frame at a time, waiting, if it needs to, for the successive link-access protocol D-channel (LAPD) packets from this one synchronous frame to accumulate. In the other scheme, the TA interleaves the LAPD packets from various synchronous frames. Using an analytical model for the first scheme, it is concluded that if the trunk speed (TA-to-network switch) to line speed ratio is greater than or equal to four, then over the range of useful loading, interleaved pipelining yields smaller mean delays than noninterleaved pipelining. If the ratio is less than four, interleaving yields larger mean delays for negative protocols which require a buildout delay at the egress TA