Title :
Global Variable Partition with Virtually Shared Scratch Pad Memory to Minimize Schedule Length
Author :
Qiu, Meikang ; Zhang, Lei ; Guo, Minyi ; Hu, Fei ; Liu, Shaobo ; Sha, Edwin H -M
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Kentucky, Lexington, KY, USA
Abstract :
On-chip memory is a critical component of MPSoC architecture. Virtually shared scratch pad memory (VSSPM) architecture is a novel approach to close the processor-memory speed gap since it can take the advantages of both shared and private SPM. In this paper, we propose an efficient variable partitioning algorithms, global view prediction (GVP), to improve the overall throughput. The experimental results demonstrate the effectiveness of our approach.
Keywords :
memory architecture; shared memory systems; system-on-chip; MPSoC architecture; global variable partition; global view prediction; on-chip memory; processor-memory speed gap; schedule length minimisation; throughput; variable partitioning algorithms; virtually shared scratch pad memory architecture; Clocks; Computer architecture; Computer science; Delay; Memory architecture; Optimal scheduling; Parallel processing; Partitioning algorithms; Prediction algorithms; Scanning probe microscopy;
Conference_Titel :
Parallel Processing Workshops, 2009. ICPPW '09. International Conference on
Conference_Location :
Vienna
Print_ISBN :
978-1-4244-4923-1
Electronic_ISBN :
1530-2016
DOI :
10.1109/ICPPW.2009.22
