Performance Analysis of Parallel Computing Framework on Embedded Multi-core Trustworthy Systems

Due to the environmental constraints, the use of high-performance embedded multi-core processor to solve the problem of scientific computing is the only way in a specified environment. This paper discusses the parallel computing framework and method in embedded multi-core system that used in safety critical system, and analyzes the performance of computation with three methods: the single thread, multi-thread and OpenMP in loongson processor, Intel Q8300 processor and I7 860 processor. The research shows that the parallel computing framework for multi-core processors can significantly improve the efficiency of program execution when the complexity of the case is not too high. The simulation test expressed that the parallel speedup is better in the loongson platform than Q8300 although the real performance of the processor is more faster than loongson.

[1]  Bai Jun-feng.  Application Development Methods Based on Multi-core Systems , 2012, 2012 International Conference on Industrial Control and Electronics Engineering.

[2]  G. Amdhal,et al.  Validity of the single processor approach to achieving large scale computing capabilities , 1967, AFIPS '67 (Spring).

[3]  Konstantinos G. Margaritis,et al.  Computational Comparison of Some Multi-core Programming Tools for Basic Matrix Computations , 2012, 2012 IEEE 14th International Conference on High Performance Computing and Communication & 2012 IEEE 9th International Conference on Embedded Software and Systems.

[4]  James Reinders,et al.  Intel® threading building blocks , 2008 .

[5]  Konstantinos G. Margaritis,et al.  Scientific computations on multi-core systems using different programming frameworks , 2016 .

[6]  Bradford Nichols,et al.  Pthreads programming - a POSIX standard for better multiprocessing , 1996 .

[7]  Lionel M. Ni,et al.  Scalable Problems and Memory-Bounded Speedup , 1993, J. Parallel Distributed Comput..

[8]  David A. Bader,et al.  SWARM: A Parallel Programming Framework for Multicore Processors , 2007, 2007 IEEE International Parallel and Distributed Processing Symposium.

[9]  Fatos Xhafa,et al.  Programming multi-core and many-core computing systems , 2014 .

[10]  John L. Gustafson,et al.  Reevaluating Amdahl's law , 1988, CACM.