A case study of a system level approach to exploration of queuing management schemes for input queue packet switches

In this paper a new system-level exploration method called architectural probing is proposed. It embodies the widely accepted concepts of behavior/architecture codesign, and allows to selectively gather functional information from a functional diagram and feed it into the architectural diagram allowing to reduce the amount of details that are simulated at the architectural level, while still maintaining a high degree of accuracy. The effectiveness of this novel method is demonstrated by a case study of exploring data queuing management schemes for packet switches with input queues. As shown in the experimental results, the method enables us to evaluate the impact of different data management algorithms and different system parameters such as DMA size and packet payload size, in terms of power and performance tradeoff by estimating the average energy per bit transferred, on each architectural component, such as memories, buses and bus arbiters, involved in the implementation of the queuing management system.