Dynamic Memory Management on GPUs with SYCL

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Standish, Russell K.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912577324318720
author Standish, Russell K.
author_facet Standish, Russell K.
contents Dynamic memory allocation is not traditionally available in kernels running on GPUs. This work aims to build on Ouroboros, an efficient dynamic memory management library for CUDA applications, by porting the code to SYCL, a cross-platform accelerator API. Since SYCL can be compiled to a CUDA backend, it is possible to compare the performance of the SYCL implementation with that of the original CUDA implementation, as well as test it on non-CUDA platforms such as Intel's Xe graphics.
format Preprint
id arxiv_https___arxiv_org_abs_2504_18211
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dynamic Memory Management on GPUs with SYCL
Standish, Russell K.
Distributed, Parallel, and Cluster Computing
68W10
D.1.3
Dynamic memory allocation is not traditionally available in kernels running on GPUs. This work aims to build on Ouroboros, an efficient dynamic memory management library for CUDA applications, by porting the code to SYCL, a cross-platform accelerator API. Since SYCL can be compiled to a CUDA backend, it is possible to compare the performance of the SYCL implementation with that of the original CUDA implementation, as well as test it on non-CUDA platforms such as Intel's Xe graphics.
title Dynamic Memory Management on GPUs with SYCL
topic Distributed, Parallel, and Cluster Computing
68W10
D.1.3
url https://arxiv.org/abs/2504.18211