CLC number:
On-line Access: 2023-02-02
Received: 2022-10-11
Revision Accepted: 2023-01-04
Crosschecked: 0000-00-00
Cited: 0
Clicked: 236
Juan FANG, Sheng LIN, Huijing YANG, Yixiang XU, Xing SU. A perceptual and predictive batch-processing memory-scheduling strategy for a CPU-GPU heterogeneous system[J]. Frontiers of Information Technology & Electronic Engineering, 1998, -1(-1): .
@article{title="A perceptual and predictive batch-processing memory-scheduling strategy for a CPU-GPU heterogeneous system",
author="Juan FANG, Sheng LIN, Huijing YANG, Yixiang XU, Xing SU",
journal="Frontiers of Information Technology & Electronic Engineering",
volume="-1",
number="-1",
pages="",
year="1998",
publisher="Zhejiang University Press & Springer",
doi="10.1631/FITEE.2200449"
}
%0 Journal Article
%T A perceptual and predictive batch-processing memory-scheduling strategy for a CPU-GPU heterogeneous system
%A Juan FANG
%A Sheng LIN
%A Huijing YANG
%A Yixiang XU
%A Xing SU
%J Journal of Zhejiang University SCIENCE C
%V -1
%N -1
%P
%@ 2095-9184
%D 1998
%I Zhejiang University Press & Springer
%DOI 10.1631/FITEE.2200449
TY - JOUR
T1 - A perceptual and predictive batch-processing memory-scheduling strategy for a CPU-GPU heterogeneous system
A1 - Juan FANG
A1 - Sheng LIN
A1 - Huijing YANG
A1 - Yixiang XU
A1 - Xing SU
J0 - Journal of Zhejiang University Science C
VL - -1
IS - -1
SP -
EP -
%@ 2095-9184
Y1 - 1998
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/FITEE.2200449
Abstract: When multiple processor (CPU) cores and integrated GPUs share off-chip main memory, CPU and GPU applications compete for the critical memory resource. This causes serious resource competition and has a negative impact on the overall performance of the system. This paper describes the competition for shared memory resources in CPU-GPU Heterogeneous multi-core architecture and a shared memory request scheduling strategy based on the perceptual and predictive batch-processing is proposed. By sensing the CPU and GPU memory request conditions in the request buffer, the proposed scheduling strategy estimates the GPU latency tolerance and reduces mutual interference between the CPU and GPU by processing CPU or GPU memory requests in batches. According to the experimental results, the scheduling strategy improves CPU performance by 8.53% and reduces mutual interference by 10.38% with low hardware complexity.
Open peer comments: Debate/Discuss/Question/Opinion
<1>