No abstract

No abstract

This paper proposes a technique that mixes simulation and an analytical method to evaluate the characteristics of Networks-onChips (NoCs). The advantage of this technique is to reduce the simulation time by reducing the complexity of the NoC model while still obtaining accurate results for latency and throughput. The basis of this technique is: (i) to send the whole payload data at once in the packet header; (ii) to reduce the NoC simulation complexity by omitting the flit by flit payload forwarding; (iii) to use an algorithm for controlling the release of the packet trailer in order to close the connection at the right time. For the evaluation of this technique, an actor-oriented model of a NoC, JOSELITO, was created. Simulation results show that JOSELITO is in average 2.3 times faster in 88% of the executed case studies than the implementation without using the proposed technique. The worst case simulation results for latency and throughput have, respectively, 5.26% and 0.1% error compared to the corresponding Register Transfer Level (RTL) model.

No abstract

The evaluation of communication latency in multiprocessor platforms requires the profiling of the application, the description of the architecture of the platform and of the mapping of application tasks onto processing cores. In this paper, we describe a layered approach that allows application developers to obtain accurate figures for communication latency from an abstract model of the application functionality. A complete separation of concerns is a critical part of this approach, so the major contribution here is the definition of interfaces between different layers: an abstract application model, its executable counterpart, the mapping heuristic and the multiprocessor platform model. Case studies with a realistic application and a NoC-based multiprocessor platform show the potential of the proposed approach using two different system evaluation techniques: simulation and static analysis.