A Real-Time Programming Model for Heterogeneous MPSoCs

Authors:
Torsten Limberg;Bastian Ristau;Gerhard Fettweis
Affiliations:
Technische Universität Dresden Vodafone Chair Mobile Communications Systems, Dresden, Germany 01062;Technische Universität Dresden Vodafone Chair Mobile Communications Systems, Dresden, Germany 01062;Technische Universität Dresden Vodafone Chair Mobile Communications Systems, Dresden, Germany 01062
Venue:
SAMOS '08 Proceedings of the 8th international workshop on Embedded Computer Systems: Architectures, Modeling, and Simulation
Year:
2008

Citing 8
Cited 1

Cilk: an efficient multithreaded runtime system

PPOPP '95 Proceedings of the fifth ACM SIGPLAN symposium on Principles and practice of parallel programming
Advanced compiler design and implementation

Advanced compiler design and implementation
OpenMP: An Industry-Standard API for Shared-Memory Programming

IEEE Computational Science & Engineering
The Problem with Threads

Computer
Sequoia: programming the memory hierarchy

Proceedings of the 2006 ACM/IEEE conference on Supercomputing
CellSs: a programming model for the cell BE architecture

Proceedings of the 2006 ACM/IEEE conference on Supercomputing
Sequoia: programming the memory hierarchy

Proceedings of the 2006 ACM/IEEE conference on Supercomputing
Observations on power-efficiency trends in mobile communication devices

EURASIP Journal on Embedded Systems

Dimensioning heterogeneous MPSoCs via parallelism analysis

Proceedings of the Conference on Design, Automation and Test in Europe

Quantified Score

Hi-index	0.00

Visualization

Abstract

Modern multi-core processors suffer from the lack of a programming model which allows efficient utilization of the available hardware. Massive software overhead is required to handle task scheduling and synchronization, resulting in power inefficiencies. In this paper we present a C++ based, real-time enabled task level programming model, which allows efficient hardware utilization. Task scheduling and synchronization is performed by a hardware unit at run-time. The automated scheduler unit is guided by offline information extracted from source code by a specialized compiler