vix.ing · top · new · best · stats · spec

Challenges and Research Directions for Large Language Model Inference Hardware

2026/01/08 by Xiaoyu Ma, David Patterson · 13 voices · 1 citation
Computer Science · Materials Science · #Advanced Neural Network Applications #Machine Learning in Materials Science #Big Data and Digital Economy

paper · pdf · doi:10.1109/mc.2026.3652916

Abstract

Large Language Model (LLM) inference is hard. The autoregressive Decode phase of the underlying Transformer model makes LLM inference fundamentally different from training. Exacerbated by recent AI trends, the primary challenges are memory and interconnect rather than compute. To address these challenges, we highlight four architecture research opportunities: High Bandwidth Flash for 10X memory capacity with HBM-like bandwidth; Processing-Near-Memory and 3D memory-logic stacking for high memory bandwidth; and low-latency interconnect to speedup communication. While our focus is datacenter AI, we also review their applicability for mobile devices.

Cited by

Discussions

Related