Evaluating CXL-Based Composable Memory Solutions - A TCO Framework for Datacenter Workloads
Open Compute Project via YouTube
The Fastest Way to Become a Backend Developer Online
PowerBI Data Analyst - Create visualizations and dashboards from scratch
Overview
Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Learn to evaluate CXL-based composable memory solutions through a comprehensive TCO framework designed for datacenter workloads in this 24-minute conference presentation from the Open Compute Project. Explore how the OCP CMS group has conducted evaluations across datacenter workloads including content caching, AI-data analytics, and legacy applications, demonstrating the potential benefits of composable CXL technology. Discover CMS Native Expansion strategies covering memory expansion TCO for discrete memory technologies provisioning, bandwidth expansion through comparative analysis of DDR4-DDR5 versus DDR4-5 + CXL for SPARK workloads, and capacity expansion examining single NUMA node configurations with CXL, DRAM, and CXL+DRAM interleaving plus thin memory provisioning with compression. Examine CMS Pooled Memory Expansion through analysis of CXL-attached memory usage for capacity expansion, capability evaluation of co-located storage workloads such as Cassandra, Redis, and MySQL, and efficiency improvements through mitigation of memory stranding via pooled memory utilization. Gain practical insights from industry experts Navneet Rao from Altera and Nilesh Shah from zeropoint technologies on assessing ROI for integrating CMS solutions into datacenter infrastructure.
Syllabus
Evaluating CXL Based Composable Memory Solutions A TCO Framework for Datacenter Workloads
Taught by
Open Compute Project