Back to Search Start Over

HPVM: Hardware-Agnostic Programming for Heterogeneous Parallel Systems.

Authors :
Ejjeh, Adel
Councilman, Aaron
Kothari, Akash
Kotsifakou, Maria
Medvinsky, Leon
Noor, Abdul Rafae
Sharif, Hashim
Zhao, Yifan
Adve, Sarita
Misailovic, Sasa
Adve, Vikram
Source :
IEEE Micro. Sep/Oct2022, Vol. 42 Issue 5, p108-117. 10p.
Publication Year :
2022

Abstract

We present Heterogeneous Parallel Virtual Machine (HPVM), a compiler framework for hardware-agnostic programming on heterogeneous compute platforms. HPVM introduces a hardware-agnostic parallel intermediate representation with constructs for the hierarchical task, data, and pipeline parallelism, including dataflow parallelism, and supports multiple front-end languages. In addition, HPVM provides optimization passes that navigate performance, energy, and accuracy tradeoffs, and includes retargetable back ends for a wide range of diverse hardware targets, including central processing units, graphics processing units, domain-specific accelerators, and field-programmable gate arrays. Across diverse hardware platforms, HPVM optimizations provide significant performance and energy improvements, while preserving object-code portability. With these capabilities, HPVM facilitates developers, domain experts, and hardware vendors in programming modern heterogeneous systems. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
02721732
Volume :
42
Issue :
5
Database :
Academic Search Index
Journal :
IEEE Micro
Publication Type :
Academic Journal
Accession number :
159210773
Full Text :
https://doi.org/10.1109/MM.2022.3186547