Discovery: Pattern Discovery and Program Shaping for Heterogeneous Manycore Systems
Lead Research Organisation:
University of Edinburgh
Department Name: Sch of Informatics
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Organisations
Publications
Castañeda Lozano R
(2020)
Parallelizing Parallel Programs
Brown C
(2020)
A Hybrid Approach to Parallel Pattern Discovery in C++
Ginsbach P
(2018)
Automatic Matching of Legacy Code to Heterogeneous APIs
Von Koch T
(2018)
Towards a compiler analysis for parallel algorithmic skeletons
Vasiladiotis C
(2021)
Loop Parallelization using Dynamic Commutativity Analysis
Lozano R
(2021)
Modernizing parallel code with pattern analysis
Description | We invented and demonstrated a new technique to understand the behaviour of legacy parallel programs. Legacy programs are those written previously, which may now require significant adaptation to make them perform well on new hardware. Our approach offers a new insight into this challenge, which offers developers and maintainers of parallel software a faster route to ongoing performance portability, as hardware evolves. |
Exploitation Route | Our approach could be developed to induustrial strenth robustness, and expanded to cover new patterns of parallelism. |
Sectors | Digital/Communication/Information Technologies (including Software) |