Applying network calculus for worst-case delay bound analysis in on-chip networks

In network-on-chip, computing worst-case delay bounds for packet delivery is crucial for designing predictable systems but yet an intractable problem due to complicated resource contention scenarios. In this paper, based on network calculus, we propose a technique for analyzing the communication delay bound for individual flows. The fundamental elements with the technique include three network calculus models that describe the traffic behaviors when flows are multiplexed, split, or controlled by feedback credits, respectively. Based on the basic models, we can compute the equivalent system service curve for individual flows and then calculate their packet delay bound.