A task-based multi-core allocation mechanism for packet acceleration

Abstract

Task allocation is a traditional and complicated problem for a system with many peripherals. Normal allocation methodology focuses on fuzzy data transfer but is unable to analyze some specific experimental statistics and does not improve system performance. Data Director (DD) is an innovative packet-based manager, is comprised of Data Index (DI) and Packet Direct Memory Access (PDMA) and realizes the enhanced data transfer in a complex system. This paper presents an improved task allocation model to enhance data transfer in a task-based model with three different allocation methodologies. The experimental results for the optimization allocations show that the proposed methodologies reduce the delay of transfer differently due to the schedule costs in these cases.