Data Allocation for Hybrid Memory With Genetic Algorithm

Meikang Qiu

doi:10.1109/TETC.2015.2398824

Data Allocation for Hybrid Memory With Genetic Algorithm

2015-12-01

70

PRE

AI

DOI：10.1109/TETC.2015.2398824

原文链接

原文求助

摘要

En 中文

The gradually widening speed disparity between CPU and memory has become an overwhelming bottleneck for the development of chip multiprocessor systems. In addition, increasing penalties caused by frequent on-chip memory accesses have raised critical challenges in delivering high memory access performance with tight power and latency budgets. To overcome the daunting memory wall and energy wall issues, this paper focuses on proposing a new heterogeneous scratchpad memory architecture, which is configured from SRAM, MRAM, and Z-RAM. Based on this architecture, we propose a genetic algorithm to perform data allocation to different memory units, therefore, reducing memory access cost in terms of power consumption and latency. Extensive and experiments are performed to show the merits of the heterogeneous scratchpad architecture over the traditional pure memory system and the effectiveness of the proposed algorithms.

Keyword:

Hybrid memory

SPM

chip multiprocessor

MRAM

Z-RAM

data allocation

genetic algorithm

Data Allocation for Hybrid Memory With Genetic Algorithm

摘要

AI总结

期刊

机构