Celebrating Diversity: A Mixture of Experts Approach for Runtime Mapping in Dynamic Environments
- Submitting institution
-
University of Edinburgh
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 58556211
- Type
- E - Conference contribution
- DOI
-
10.1145/2737924.2737999
- Title of conference / published proceedings
- Proceedings of the 36th ACM SIGPLAN Conference on Programming Language Design and Implementation
- First page
- 499
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- June
- Year of publication
- 2015
- 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
-
1
- Research group(s)
-
A - Computer Systems
- Citation count
- 10
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper is the first to use the idea of competing predictors to manage scheduling of multiple workloads in a dynamic environment. This paper was published at PLDI - the world's premier conference for programming language implementation, outperforming all existing approaches. It has been widely downloaded and applied to different scenarios including: cloud scheduling, embedded deep neural network selection and edge task allocation. It has also been incorporated into HPC scheduling at Lawrence Livermore National Labs. This work led to invitation and participation in the successful EU Bonseyes project, developing new technologies for constrained devices.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -