Advanced Packaging Technologies in Memory Applications for Future Generative AI Era
Ki-Ill Moon, Hwa–Young Son, Kangwook Lee
Abstract
It is expected that memory usage will exponentially increase according to the explosive interest in generative AI and its expanded application. Along with this industrial trend, the demand for memory products with high bandwidth, high capacity, and better power efficiency is increasing. In this paper, we would like to introduce key packaging technology of a high bandwidth memory (HBM) and its future challenges which SK Hynix has led its technology as world 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> and world best performance. In addition, a processor-in-memory (PiM) as a promising memory solution for generative AI, will also be introduced focused on its proto-type demonstration for GDDR6 and thermal design of a package.