SK Hynix moves AI data centers from processor-centric to memory-centric design
DIGITIMESen

SK Hynix has presented the Memory-Centric AI Rack System (MARS) at the Open Compute Project (OCP) Global Summit. The architecture separates memory from individual servers and pools it at the rack level over Compute Express Link (CXL) fabrics. The company says this addresses the capacity and cost limits that are constraining the growth of large AI models.
This is a short summary published by AI Global Wire. The full article is owned and hosted by DIGITIMES — open it there to read it in full.
Read the full story at DIGITIMESRelated AI news
- Amazon introduces toolkit for running PyTorch workloads on AWS Trainium chipsSiliconANGLE · October 20, 2026
- 從 2021 年的研究路線圖到 Claude Code,Anthropic 如何把 AI 程式設計師變成現實TechNews (TW) · October 20, 2026
- AI 代打面試大翻車!YouTuber 派機器人對決「AI 面試官」,履歷造假、薪水亂談超失控TechNews (TW) · October 19, 2026
- Nscale and Anthropic are reportedly in talks for yet another $45 billion compute dealTechCrunch AI · October 19, 2026
- Sources: OpenAI is racing to secure investor commitments before Anthropic's expected IPO in mid-Nov. and aims to raise ~$30B; the round could value it at $1.4T+ (Reuters)Techmeme · October 19, 2026
- DIGITIMES Insight: AI packaging power reaches kilowatt scale as thermal design moves toward the dieDIGITIMES · October 19, 2026