Optimizing Layer-Fused Scheduling of Transformer Networks on Multi-accelerator Platforms

The impact of transformer networks is booming, yet, they come with significant computational complexity. It is therefore essential to understand how to optimally map and execute these networks on modern neural processor hardware. So far, literature on transformer scheduling optimization has been foc...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-06
Main Authors: Colleman, Steven, Symons, Arne, Jung, Victor J B, Verhelst, Marian
Format: Article
Language:English
Subjects:
Hardware
Microprocessors
Networks
Optimization
Schedules
Scheduling
Transformers
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The impact of transformer networks is booming, yet, they come with significant computational complexity. It is therefore essential to understand how to optimally map and execute these networks on modern neural processor hardware. So far, literature on transformer scheduling optimization has been focusing on deployment on GPU and specific ASICs. This work enables extensive hardware/mapping exploration by extending the DSE framework Stream towards support for transformers across a wide variety of hardware architectures and different execution schedules. After validation, we explore the optimal schedule for transformer layers/attention heads and investigate whether layer fusion is beneficial to improve latency, energy or memory requirements. Our study shows that the memory requirements for active feature data can be drastically reduced, by adapting the execution schedule based on the size of the input of the attention head.
ISSN:2331-8422