HAAN: A Holistic Approach for Accelerating Normalization Operations in Large Language Models

TL;DR

This paper introduces HAAN, a holistic algorithm-hardware co-design approach that accelerates normalization operations in large language models, significantly improving hardware efficiency and reducing latency and energy consumption.

Contribution

The paper presents a novel co-design method, HAAN, that optimizes normalization operations in LLMs, addressing latency and energy issues more effectively than existing solutions.

Findings

HAAN achieves higher hardware performance than state-of-the-art methods.

Normalization acceleration reduces processing latency in LLMs.

Energy efficiency is significantly improved with HAAN.

Abstract

Large language models (LLMs) have revolutionized natural language processing (NLP) tasks by achieving state-of-the-art performance across a range of benchmarks. Central to the success of these models is the integration of sophisticated architectural components aimed at improving training stability, convergence speed, and generalization capabilities. Among these components, normalization operation, such as layer normalization (LayerNorm), emerges as a pivotal technique, offering substantial benefits to the overall model performance. However, previous studies have indicated that normalization operations can substantially elevate processing latency and energy usage. In this work, we adopt the principles of algorithm and hardware co-design, introducing a holistic normalization accelerating method named HAAN. The evaluation results demonstrate that HAAN can achieve significantly better hardware performance compared to state-of-the-art solutions.