Learn about innovative solutions for GPU memory limitations in this 23-minute conference talk from the 2024 OCP APAC Summit that explores the development and implementation of a CXL Controller capable of achieving two-digit nanosecond latency, addressing one of the most significant challenges in modern GPU computing architectures.
Overcoming the GPU Memory Capacity Limitation with Two-Digit Nanosecond Latency CXL Controller
Open Compute Project via YouTube
Overview
Syllabus
Overcoming the GPU Memory Capacity Limitation with Two-Digit Nanosecond Latency CXL Controller
Taught by
Open Compute Project
Related Courses
-
Advancing CXL Memory Architecture for Enhanced Performance and Flexible Integration
-
Persistent CXL Memory Module with DRAM Performance - A Technical Overview
-
Understanding CXL Technology: Memory Expansion and Storage Integration in Modern Data Centers
-
Case for DRAM Translation Layer - It's Like FTL but for CXL DRAM
-
CXL Shared Memory - Progress, Challenges and Usability
-
Real World Implementation of the OCP Hyperscale CXL TME Specification
