Regional open-source communities will gather at the Suwon Convention Center in South Korea for a community-powered open-source community event! The OCP APAC Summit is being co-located with the OpenInfra Summit Asia to bring two premier open-source software and hardware infrastructure events together!
https://2024.openinfraasia.org/
https://2024.openinfraasia.org/a/schedule
Day 1
Day 2
Keynote Photos
Day 1 Presentations
| TITLE | EVENT | LINK |
| Advancing CXL Memory Architecture for Enhanced Performance and Flexible Integration | 401 - TALK 1 | Recording |
| CXL Memory Integration in IaaSPaaS | 401 - TALK 2 | Recording |
| CXL Switch for AI Domain Specific Models | 401 - TALK 3 | Recording |
| Effective Data Tiering and Memory Allocation Techniques in Heterogeneous Memory System | 401 - TALK 4 | Recording |
| OCP Hyperscale compliant AI storage servers with compression and deduplication accelerators | 401 - TALK 5 | Recording |
| Project Caliptra Update | 402 - TALK 2 | Recording |
| Advancing Data Center Sustainability | 402 - TALK 3 | Recording |
| Secure System Design on Arm using PA-RoT | 402 - TALK 4 | Recording |
| Leveraging OCP to Create a Purpose Built ARM Processor | 403 - TALK 1 | Recording |
| Open Compute Accelerates the Ecosystem Development of Compute Infrastructure in the Intelligent | 403 - TALK 2 | Recording |
| P4OVN Offloading OVN Pipeline to Intel IPU | 403 - TALK 3 | Recording |
| Open-Source Chips Achievements and Challenges | 403 - TALK 4 | Recording |
| OCP solution for Korean datacenter | 403 - TALK 5 | Recording |
| Data Center Liquid Cooling Trends | 404 - TALK 1 | Recording |
| Direct to chip liquid cooling | 404 - TALK 2 | Recording |
| OCP Standard Quick Connect Couplings for Liquid Cooling AI Data Centers | 404 - TALK 3 | Recording |
| Liquid Cooling When Component Failure is Not an Option | 404 - TALK 4 | Recording |
| Explainable A.I. to Increase Safe Use of A.I. - A Case of Manufacturing | 405 - TALK 1 | Recording |
| Infrastructure Software for Hyperscale AI and GPU Clusters | 405 - TALK 2 | Recording |
| Elice Cloud An era of AI-powered Data Center and Cloud Computing | 405 - TALK 3 | Recording |
| Hybrid AI Operational Strategies for Enterprise AX Realization | 405 - TALK 4 | Recording |
| Overcoming the GPU Memory Capacity Limitation with Two-Digit Nanosecond Latency CXL Controller | 405 - TALK 5 | Recording |
| Case for DRAM Translation Layer (DTL) It’s like FTL but for (CXL) DRAM | 406 - TALK 1 | Recording |
| HMComp Unleashing Hidden Potential - Compression for Faster AI using Hybrid Memory Tiers | 406 - TALK 2 | Recording |
| Chiplet-Based Compressed LLC Cache & Memory Expansion | 406 - TALK 3 | Recording |
| Vector storage for large scale multimodal RAG | 406 - TALK 5 | Recording |
| Paving the Golden Path with Open Source - Accelerating Development Efficiency and Scalability | 407 - TALK 3 | Recording |
| Real world implementation of the OCP Hyperscale CXL TME Specification | 407 - TALK 4 | Recording |
| Rack Scale and Liquid Cooling for AI | 407 - TALK 5 | Recording |