SPGPU: Spatially Programmed GPU

Communication is a critical bottleneck for GPUs, manifesting as energy and performance overheads due to network-on-chip (NoC) delay and congestion. While many algorithms exhibit locality among thread blocks and accessed data, modern GPUs lack the interface to exploit this locality: GPU thread blocks...

Full description

Bibliographic Details

Published in:	IEEE computer architecture letters Vol. 23; no. 2; pp. 223 - 226
Main Authors:	Zhu, Shizhuo, Shkirko, Illia, Levinson, Jacob, Wang, Zhengrong, Nowatzki, Tony
Format:	Journal Article
Language:	English
Published:	IEEE 01.07.2024
Subjects:	GPU Graphics processing units Hardware Instruction sets Kernel Message systems on-chip networks Optimization Pipelines Programming programming abstractions Registers shared-data locality Spatial architectures Topology
ISSN:	1556-6056, 1556-6064
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!