On-GPU thread-data remapping for nested branch divergence

doi:10.1016/j.jpdc.2020.02.003

2020-05-01

PRE

Huanxin Lin *

DOI：10.1016/j.jpdc.2020.02.003

原文链接

原文求助

摘要

En 中文

Nested branches are common in applications with decision trees. The more layers in the branch nest, the larger slowdown is caused by nested branch divergence on GPU. Since inner branches are impractical to evaluate on host end, thread-data remapping via GPU shared memory is so far the most suitable solution. However, existing solution cannot handle inner branches directly due to undefined behavior of GPU barrier function when executed within branch statements. Race condition needs to be prevented without using barrier function. Targeting nested divergence, we propose NeX as a nested extension scheme featuring an inter-thread protocol that supports sub-workgroup synchronization. We further exploit the on-the-fly nature of Head-or-Tail (HoT) algorithm and propose HoT2 with enhanced flexibility of wavefront scheduling. Evaluated on four GPU models including NVIDIA Volta and Turing, HoT2 confirms to be more efficient. For benchmarks with branch nests up to five-layer-deep, NeX further boosts performance by up to 1.56x. (C) 2020 Elsevier Inc. All rights reserved.

Keyword:

GPGPU

Branch divergence

SIMD

Race condition

AI总结

对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Journal of Parallel and Distributed Computing

IF:

论文数:

3.8K

被引数:

4.8K

机构

University of Hong Kong

学者数:

4.1W

论文数: 3.9W

被引数: 10.1W

引用论文

A chirp-compensated, injection-seeded alexandrite laser

Applied Physics B

IF0

2014-01-23

PREAI

P. Bakule; P.E.G. Baird; M.G. Boshier; S.L. Cornish; D.F. Heller; K. Jungmann; I.C. Lane; V. Meyer; P.H.G. Sandars; W.T. Toner; M. Towrie; J.C. Walling

Necroptosis in Solid Organ Transplantation: A Literature Overview

International Journal of Molecular Sciences

IF0

2022-03-27