An effective scheme for pre-emptive priorities in dual bus Metropolitan Area Networks

Abstract

The IEEE 802.6 standard for Metropolitan Area Networks does not provide multiple priority tra c for connectionless data services. A priority m e c hanism that was considered in earlier versions of the standard showed to be not e ective. As of now, there exists no protocol for multiple access dual bus networks that is able to implement pre-emptive priorities and, at the same time, can satisfy minimal fairness requirements for transmissions at the highest priority level. In this study, a protocol with strictly pre-emptive priorities, i.e., a protocol that does not admit low priority tra c if the load from high priority tra c exceeds the capacity of the transmission channel, is presented. In addition, the protocol guarantees fairness for transmissions at the highest priority l e v el. By introducing a general characterization of bandwidth allocation schemes for dual bus networks, existing priority m e c hanisms can be categorized according to the provided quality of service. The unique existence of a bandwidth allocation scheme for multiple priority tra c is shown with a full utilization of the channel capacity, with a fair distribution of bandwidth respective to tra c from a particular priority l e v el and with pre-emptive priorities. The performance of the presented protocol is compared to existing proposals for multiple priority mechanisms. It is shown that adopting the new protocol results in shorter access delays for high priority transmissions. The protocol allows the stations of the network to react quickly to load changes. It is shown that the e ectiveness of the priority s c heme, compared to priority schemes using the bandwidth balancing mechanism, is less dependent on increasing the transmission speed of the network.

A dual bus approach for LAN interworking with ATM networks

“…Recently, many fair distributed queuing methods were proposed to overcome the drawbacks of the original DQDB standard [13], [8]. For our flow control, the scheme called Slot Interleaved Multiple Access (SIMA) which we have proposed in [9] is used.…”