HiCR, an Abstract Model for Distributed Heterogeneous Programming
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914015869927424 |
|---|---|
| author | Martin, Sergio Miguel Terracciano, Luca Dichev, Kiril Baumann, Noah Lin, Jiashu Yzelman, Albert-Jan |
| author_facet | Martin, Sergio Miguel Terracciano, Luca Dichev, Kiril Baumann, Noah Lin, Jiashu Yzelman, Albert-Jan |
| contents | We present HiCR, a model to represent the semantics of distributed heterogeneous applications and runtime systems. The model describes a minimal set of abstract operations to enable hardware topology discovery, kernel execution, memory management, communication, and instance management, without prescribing any implementation decisions. The goal of the model is to enable execution in current and future systems without the need for significant refactoring, while also being able to serve any governing parallel programming paradigm. In terms of software abstraction, HiCR is naturally located between distributed heterogeneous systems and runtime systems. We coin the phrase \emph{Runtime Support Layer} for this level of abstraction. We explain how the model's components and operations are realized by a plugin-based approach that takes care of device-specific implementation details, and present examples of HiCR-based applications that operate equally on a diversity of platforms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_01425 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HiCR, an Abstract Model for Distributed Heterogeneous Programming Martin, Sergio Miguel Terracciano, Luca Dichev, Kiril Baumann, Noah Lin, Jiashu Yzelman, Albert-Jan Distributed, Parallel, and Cluster Computing We present HiCR, a model to represent the semantics of distributed heterogeneous applications and runtime systems. The model describes a minimal set of abstract operations to enable hardware topology discovery, kernel execution, memory management, communication, and instance management, without prescribing any implementation decisions. The goal of the model is to enable execution in current and future systems without the need for significant refactoring, while also being able to serve any governing parallel programming paradigm. In terms of software abstraction, HiCR is naturally located between distributed heterogeneous systems and runtime systems. We coin the phrase \emph{Runtime Support Layer} for this level of abstraction. We explain how the model's components and operations are realized by a plugin-based approach that takes care of device-specific implementation details, and present examples of HiCR-based applications that operate equally on a diversity of platforms. |
| title | HiCR, an Abstract Model for Distributed Heterogeneous Programming |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2509.01425 |