Partitioning Algorithms Analysis for Heterogeneous NoC Based MPSoC

Several new applications have high complexity degree, requiring high processing rate and memory usage. Multiprocessor System-on-Chip (MPSoC) is a promising architecture to fulfill these requirements, due to its high parallelism that enables several tasks been executed at the same time. One problem i...

Full description

Saved in:
Bibliographic Details
Published in2012 Brazilian Symposium on Computing System Engineering pp. 178 - 183
Main Authors Pinotti, I. K., Webber, T., Ribeiro, N., Fraga, C. N., Fagundes, R. D. R., Marcon, C.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2012
Subjects
Algorithm design and analysis
Computer architecture
Cost function
Energy consumption
Mapping
MPSoC
NoC
Partitioning
Partitioning algorithms
Program processors
Tiles
Online AccessGet full text

Cover

More Information
Summary:Several new applications have high complexity degree, requiring high processing rate and memory usage. Multiprocessor System-on-Chip (MPSoC) is a promising architecture to fulfill these requirements, due to its high parallelism that enables several tasks been executed at the same time. One problem in current heterogeneous MPSoC design is application's tasks partitioning aiming energy consumption minimization and load balance. In order to optimize partition problems, many algorithms have been applied to generate quality solutions. This work aims to analyze and compare stochastic and heuristic partitioning algorithms for obtaining low energy consumption and load balance when applied to tasks partitioning onto heterogeneous MPSoC.
ISBN:1467357472
9781467357470
ISSN:2324-7886
2324-7894
DOI:10.1109/SBESC.2012.42