AdaMD: Adaptive Mapping and DVFS for Energy-efficient Heterogeneous Multi-cores
- Submitting institution
-
The University of Essex
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1378
- Type
- D - Journal article
- DOI
-
10.1109/tcad.2019.2935065
- Title of journal
- IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Article number
- -
- First page
- 2206
- Volume
- 39
- Issue
- 10
- ISSN
- 0278-0070
- Open access status
- Compliant
- Month of publication
- August
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
D - Robotics and Embedded Systems (RES)
- Citation count
- 3
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work extended our previous highly-cited paper by adapting to different number of cores over execution, further enhancing performance. The development, started in EPSRC's PRiME (EP/K034448/1) and Graceful (EP/L000563/1) projects along with industrial partner ARM-Holdings was the first paper to demonstrate this approach. This work underpinned discussions with numerous well-known chip design and usage companies, leading to several invited talks, NDAs and proposals for further work. Our new resource-allocation approach enables the latest performance-constrained applications to be executed on multi-core chip based systems, allocating each application on more than one types of cores with online adaptation to achieve energy savings.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -