Tech News
Back to home
Technology

SK hynix completes development of next-gen HBM4 — 2,048-bit interface and 10 GT/s speeds promised for next-gen AI accelerators

Source

Tom's Hardware

Published

TL;DR

AI Generated

SK hynix has announced the completion of the development of its HBM4 memory, featuring a 2,048-bit interface and 10 GT/s data transfer speeds, surpassing JEDEC specifications by 25%. The company's HBM4 stacks utilize advanced DRAM dies and a unique mass reflow molded underfill method for improved performance and heat dissipation. While details like the number of DRAM layers and capacity remain undisclosed, SK hynix is ready for mass production. The move to exceed JEDEC standards in data transfer rate aligns with industry trends, with Micron and Rambus also pushing boundaries in HBM4 technology. This advancement is expected to benefit next-gen AI accelerators from major players like AMD, Broadcom, and Nvidia.

Read Full Article

Similar Articles

SemiEngineering

Panel-Level Packaging’s Second Wave Meets Engineering Reality

Panel-level packaging is gaining traction due to economic pressures and the increasing size of AI accelerators and HPC packages. Glass substrates are being explored to address warpage and dimensional stability issues, but they introduce new failure modes that require material solutions. Challenges in panel-level processing include materials and process integration, not just packaging problems. The industry is moving towards panels driven by economic and technological shifts, but solving these challenges requires a holistic approach.

SemiEngineering•
SemiEngineering

Inside the AI Accelerator: Essential IP Design Solutions: eBook

The eBook delves into how advanced IP, high-speed interconnects, memory interfaces, and multi-die architectures are utilized in next-gen AI accelerators to surpass single-chip limitations. It highlights the role of optical links in enhancing bandwidth and security IP in safeguarding AI data without compromising performance. The eBook also covers how technologies like UALink, PCIe, CXL, and Ultra Ethernet support scaling AI architectures, integrating compute, memory, and accelerators, and enhancing bandwidth density through optical I/O. The focus is on unlocking AI performance at scale and ensuring data security across accelerators.

SemiEngineering•
WEBINAR: HBM4E Advances Bandwidth Performance for AI Training

WEBINAR: HBM4E Advances Bandwidth Performance for AI Training

Rambus is launching its HBM4E memory controller IP product tailored for AI training applications to address the increasing pressure on memory technologies due to the rise of AI applications and high-end GPU platforms. The "memory wall" challenge is highlighted, emphasizing the need for advanced memory architectures to prioritize raw bandwidth for AI training. HBM technology is positioned as a solution for high-performance GPUs targeting AI training servers, offering wider buses, faster transfer rates, and increased stack heights. Rambus leverages its expertise to deliver high transfer speeds with its HBM4E controller, providing significant bandwidth for memory devices. The webinar hosted by Rambus delves into AI use cases, HBM architecture, and the capabilities of the HBM4E controller for those looking to optimize AI training servers and racks.

SemiWiki•
Nvidia demonstrates Rubin Ultra tray, the world's first AI GPU with 1TB of HBM4E memory — new chips will slot into Kyber racks

Nvidia demonstrates Rubin Ultra tray, the world's first AI GPU with 1TB of HBM4E memory — new chips will slot into Kyber racks

Nvidia showcased its upcoming Rubin Ultra tray, the first AI GPU with 1TB of HBM4E memory, set to launch in 2027. This new GPU platform will utilize the Kyber rack design, integrating 144 GPU packages for enhanced performance compared to current models. The Rubin Ultra package features four compute chiplets and a new packaging technology, potentially stacked for efficiency. The Kyber rack will also introduce liquid cooling and a 7th Generation NVLink switch to accommodate more GPUs, promising significant performance improvements.

Tom's Hardware•

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.