RT-DBR: Dynamic Bandwidth Reservation for Real-Time Application on CPU-GPU Heterogeneous SoC

When heterogeneous cores concurrently access shared memory, severe bandwidth contention limits data transmission, increases memory access latency for tasks, disrupts the predictability of real-time application execution, and affects system responsiveness. To address these issues, this paper proposes a dynamic bandwidth regulation mechanism based on core memory access patterns (RT-DBR). This mechanism enables fine-grained adjustment of core memory access request scheduling rate through the memory controller. Furthermore, it operates transparently to the application programs on the cores, effectively reducing system interruptions caused by bandwidth regulation. Experimental results demonstrate that the RT-DBR can effectively alleviate unnecessary bandwidth contention caused by bursty memory accesses and ensure real-time performance of programs.

Authors

Institutions

Publication Details

Published
2024-12-06
DOI
https://doi.org/10.1145/3709026.3709098
Primary Topic
Parallel Computing and Optimization Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

RT-DBR: Dynamic Bandwidth Reservation for Real-Time Application on CPU-GPU Heterogeneous SoC

Yiming Yan, Juan Fang, Sheng Lin
Parallel Computing and Optimization Techniques
article

RT-DBR: Dynamic Bandwidth Reservation for Real-Time Application on CPU-GPU Heterogeneous SoC

Yiming Yan, Juan Fang, Sheng Lin
article en

Abstract

When heterogeneous cores concurrently access shared memory, severe bandwidth contention limits data transmission, increases memory access latency for tasks, disrupts the predictability of real-time application execution, and affects system responsiveness. To address these issues, this paper proposes a dynamic bandwidth regulation mechanism based on core memory access patterns (RT-DBR). This mechanism enables fine-grained adjustment of core memory access request scheduling rate through the memory controller. Furthermore, it operates transparently to the application programs on the cores, effectively reducing system interruptions caused by bandwidth regulation. Experimental results demonstrate that the RT-DBR can effectively alleviate unnecessary bandwidth contention caused by bursty memory accesses and ensure real-time performance of programs.

Beijing University of Technology (CN)
Affordable and clean energy
Openalex Percentile: Top 70%
Parallel Computing and Optimization Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

RT-DBR: Dynamic Bandwidth Reservation for Real-Time Application on CPU-GPU Heterogeneous SoC — Yiming Yan, Juan Fang, et al. · (2024) | TGRS Research Map | TGRS