Synergizing CXL with Unified Memory for Scalable GPU Memory Expansion

Synergizing CXL with Unified Memory for Scalable GPU Memory Expansion

Heterogeneous computing, which combines the power of CPUs and GPUs, is increasingly important in today's data-centric era. However, it also presents significant challenges, particularly in terms of GPU memory management. Compute eXpress Link (CXL) technology, which is supported by industry lead...

Full description

Saved in:
Bibliographic Details
Published in:2024 International Conference on Electronics, Information, and Communication (ICEIC) pp. 1 - 4
Main Authors: Lee, Junseung, Kim, Jungrae
Format: Conference Proceeding
Language:English
Published: IEEE 28-01-2024
Subjects:
CXL(Compute eXpress Link)
Electric potential
Graphics processing units
Heterogeneous computing
Industries
Information sharing
Memory management
Programming
Thermal expansion
UM(Unified Memory)
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Heterogeneous computing, which combines the power of CPUs and GPUs, is increasingly important in today's data-centric era. However, it also presents significant challenges, particularly in terms of GPU memory management. Compute eXpress Link (CXL) technology, which is supported by industry leaders, offers enhanced memory expansion capabilities. Alongside this, NVIDIA's Unified Memory (UM) system simplifies the process of sharing data between CPUs and GPUs. We explore the role of CXL memory in various scenarios within the UM framework while simultaneously highlighting issues such as the handling of page faults at the software level.
ISSN:2767-7699
DOI:10.1109/ICEIC61013.2024.10457110