A GPU Register File using Static Data Compression

06/10/2020
by   Alexandra Angerd, et al.
0

GPUs rely on large register files to unlock thread-level parallelism for high throughput. Unfortunately, large register files are power hungry, making it important to seek for new approaches to improve their utilization. This paper introduces a new register file organization for efficient register-packing of narrow integer and floating-point operands designed to leverage on advances in static analysis. We show that the hardware/software co-designed register file organization yields a performance improvement of up to 79

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/19/2020

Enabling High-Capacity, Latency-Tolerant, and Highly-Concurrent GPU Register Files via Software/Hardware Cooperation

Graphics Processing Units (GPUs) employ large register files to accommod...
research
03/09/2020

Forensic Analysis of Residual Information in Adobe PDF Files

In recent years, as electronic files include personal records and busine...
research
03/03/2021

Exploitation and Sanitization of Hidden Data in PDF Files

Organizations publish and share more and more electronic documents like ...
research
04/23/2018

Forensic Analysis of the exFAT artefacts

Although keeping some basic concepts inherited from FAT32, the exFAT fil...
research
03/22/2019

Understanding and taming SSD read performance variability: HDFS case study

In this paper we analyze the influence that lower layers (file system, O...
research
09/14/2017

GREENER: A Tool for Improving Energy Efficiency of Register Files

Graphics Processing Units (GPUs) maintain a large register file to incre...
research
05/09/2021

RRCD: Redirección de Registros Basada en Compresión de Datos para Tolerar FallosPermanentes en una GPU

The ever-increasing parallelism demand of General-Purpose Graphics Proce...

Please sign up or login with your details

Forgot password? Click here to reset