2020 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS) 2020
DOI: 10.1109/rtas48715.2020.00-17
|View full text |Cite
|
Sign up to set email alerts
|

Real-Time Scheduling upon a Host-Centric Acceleration Architecture with Data Offloading

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
4
3

Relationship

1
6

Authors

Journals

citations
Cited by 8 publications
(2 citation statements)
references
References 36 publications
0
2
0
Order By: Relevance
“…So far, most researchers have focused on improving the accuracy of perception and prediction methods and models. In contrast, there have been fewer studies on identifying the relationship and trade-off between runtime-accuracy and the end-to-end delay of autonomous driving operation [ 10 , 11 ]. Since autonomous driving operation consists of several different stages of process, as illustrated in Figure 1 , it is very important to complete the flow of these processes within some time constraints.…”
Section: Introductionmentioning
confidence: 99%
“…So far, most researchers have focused on improving the accuracy of perception and prediction methods and models. In contrast, there have been fewer studies on identifying the relationship and trade-off between runtime-accuracy and the end-to-end delay of autonomous driving operation [ 10 , 11 ]. Since autonomous driving operation consists of several different stages of process, as illustrated in Figure 1 , it is very important to complete the flow of these processes within some time constraints.…”
Section: Introductionmentioning
confidence: 99%
“…The demand to achieve high accuracy for increasingly computational tasks leads to ever-growing model sizes of modern DNNs [8], [9]. Processing such huge DNNs [5], [6], [7], [8], [9] in systems with conventional Von Neumann architectures [10], [11], [12] incurs enormous energy consumption and significant execution latency due to the vast data movement between the separate memory and computing elements. To tackle this challenge, processingin-memory (PIM) is introduced as a promising paradigm by co-locating compute and memory.…”
Section: Introductionmentioning
confidence: 99%