News

Samsung Unveils Industry’s First 36GB HBM3E 12H DRAM

Spread the love

Undoubtedly, Samsung is one of the leaders in advanced technology, and to maintain its legacy, it always tries hard. Now it has announced the 12-stack HBM3E DRAM

Samsung finally introduced the very first 12-stack HBM3E DRAM and the highest-capacity HBM product to date. The brand has manufactured it in such a way that it can deliver an all-time high bandwidth of at least up to 1,280 GB per second and also an industry-leading capacity of 36GB. This time, with the 8-stack HBM3 8H, both features have been enhanced by more than 50%. 

Yongcheol Bae, the Executive Vice President of Memory Product Planning at Samsung Electronics, stated that this new memory solution forms part of our drive toward developing core technologies for high-stack HBM and providing technological leadership for the high-capacity HBM market in the AI era.” 

The HBM3E 12H has a bearing on advanced and upgraded thermal compression non-conductive film (TC NCF), which permits the 12-layer products to have the same height specification as 8-layer ones to meet current HBM package requirements. 

This technology is likely to bring advantages, specifically with higher stacks, as the industry seeks to mitigate chip die warping that comes along with thinner dies. The brand has chosen to lower the thickness of its NCF material and get the smallest gap between chips at seven micrometers of industry, while also removing the voids between layers. This all leads to an improved vertical density increase of 20% as compared to its HBM3 8H product. 

Basically, this brand new TC NCF technology allows Samsung to use different-sized bumps between chips while bonding. This leads to improved thermal management. It is expected that the growth of AI applications will help the HBM3E become more ideal for future systems requiring more memory. As compared to HBM3 8H, this time the brand has potentially increased the AI training speed by 34%, which permits more than 11.5 times more concurrent users for AI and also authorizes data centers to manage resources efficiently and reduce costs.

Related Articles

Back to top button