LET Paradigm Scheduling Algorithm Considering Parallel Processing on Clustered Many-core Processor

Embedded systems such as self-driving require computing platforms with high power and low consumption. Multi/many-core meet these requirements effectively. However, for hard real-time applications, multiple demands on shared resources can impede performance, in particular, memory is a resource that restrict the desired performance significantly. Therefore, logical execution time (LET) paradigm has gained attention to make timing of access deterministic. This paper proposes theoretical scheduling method model applying LET directed acyclic graph (DAG) nodes multi/many-core platform. lacks scalability owing overhead caused because set longer than actual task. The proposed performs parallel calculation tasks utilizing many cores deal by adopting paradigm. evaluation shows benefits from adoption end-to-end latency variation smaller existing methods. Furthermore, maintain task deadline miss low, despite