vix.ing · top · new · best · stats · spec

Graph-based Approach for Buffer-aware Timing Analysis of Heterogeneous\n Wormhole NoCs under Bursty Traffic

2019/11/06 by Frédéric Giroudot, Giroudot, Frederic, Ahlem Mifdaoui +1
Computer Science · Materials Science · #FOS: Computer and information sciences #Interconnection Networks and Systems #Performance (cs.PF) #Software-Defined Networks and 5G #Supercapacitor Materials and Fabrication

paper · pdf · doi:10.48550/arxiv.1911.02430

openalex publication_date 2019/11/06 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

This paper addresses the problem of worst-case timing analysis of\nheterogeneous wormhole NoCs, i.e., routers with different buffer sizes and\ntransmission speeds, when consecutive packet queuing (CPQ) occurs. The latter\nmeans that there are several consecutive packets of one flow queuing in the\nnetwork. This scenario happens in the case of bursty traffic but also for\nnon-schedulable traffic. Conducting such an analysis is known to be a\nchallenging issue due to the sophisticated congestion patterns when enabling\nbackpressure mechanisms. We tackle this problem through extending the\napplicability domain of our previous work for computing maximum delay bounds\nusing Network Calculus, called Buffer-aware worst-case Timing Analysis (BATA).\nWe propose a new Graph-based approach to improve the analysis of indirect\nblocking due to backpressure, while capturing the CPQ effect and keeping the\ninformation about dependencies between flows. Furthermore, the introduced\napproach improves the computation of indirect-blocking delay bounds in terms of\ncomplexity and ensures the safety of these bounds even for non-schedulable\ntraffic. We provide further insights into the tightness and complexity issues\nof worst-case delay bounds yielded by the extended BATA with the Graph-based\napproach, denoted G-BATA. Our assessments show that the complexity has\ndecreased by up to 100 times while offering an average tightness ratio of 71%,\nwith reference to the basic BATA. Finally, we evaluate the yielded improvements\nwith G-BATA for a realistic use case against a recent state-of-the-art\napproach. This evaluation shows the applicability of G-BATA under more general\nassumptions and the impact of such a feature on the tightness and computation\ntime.\n

Related