PROJECT TITLE :

A Flexible Energy- and Reliability-Aware Application Mapping for NoC-Based Reconfigurable Architectures

ABSTRACT:

This paper proposes a versatile energy- and reliability-aware application mapping approach for network-on-chip (NoC)-based reconfigurable architecture. A parameterized price model is initial developed by combining energy and reliability with a weight parameter that defines the optimization priority. Using this model, the overall mapping price might be evaluated. Subsequently, a mapping technique using branch and bound with a partial value ratio is utilized to seek out the simplest mapping by enumerating all the doable patterns organized during a search tree. To enhance the search efficiency, nonoptimal mappings are discarded at early stages using the partial cost ratio. Using the proposed approach, applications can be mapped onto most NoC topologies and running with numerous routing algorithms when considering both energy and reliability. Other state-of-the-art works have additionally done substantial analysis for the identical topic however solely restricted to a specific topology or routing algorithm. Even for the identical topology and routing algorithm, the proposed approach still shows considerable benefits in many aspects. Experiments show that this approach gains not only important reduction in energy but conjointly improvement in reliability. It also outperforms different approaches in throughput and latency with competitive run time.