Title :
A two-phase floorplanning approach for Application-specific Network-on-Chip
Author :
Shuang Yu ; Fen Ge ; Gui Feng ; Ning Wu
Author_Institution :
Coll. of Electron. & Inf. Eng., Nanjing Univ. of Aeronaut. & Astronaut., Nanjing, China
Abstract :
In this paper, we present a two-phase floorplanning approach based on genetic algorithm to arrange each IP core reasonably into a fixed-outline rectangle for Application-specific Network-on-Chip (NoC) design. Our algorithm consists of core clustering and cluster floorplanning. These steps are done to arrange IP cores with optimizing communication power consumption and chip area. Experimental results on several multimedia benchmarks show that our proposed algorithm saves about 30% of power consumption and 5% of chip area compared to random floorplanning on average.
Keywords :
VLSI; application specific integrated circuits; genetic algorithms; integrated circuit layout; integrated logic circuits; low-power electronics; microprocessor chips; network-on-chip; IP core; application-specific network-on-chip design; chip area; cluster floorplanning; communication power consumption; core clustering; genetic algorithm; two-phase floorplanning approach; Algorithm design and analysis; Biological cells; Clustering algorithms; IP networks; Network topology; Power demand; Topology;
Conference_Titel :
ASIC (ASICON), 2013 IEEE 10th International Conference on
Conference_Location :
Shenzhen
Print_ISBN :
978-1-4673-6415-7
DOI :
10.1109/ASICON.2013.6811835