Please use this identifier to cite or link to this item:
For citation please use:
Main Title: MEMPower: Data-Aware GPU Memory Power Model
Author(s): Lucas, Jan
Juurlink, Ben
Type: Conference Object
Language Code: en
Abstract: This paper presents the MEMPower power model. MEMPower is a detailed empirical power model for GPU memory access. It models the data dependent energy consumption as well as individual core specific differences. We explain how the model was calibrated using special micro benchmarks as well as a high-resolution power measurement testbed. A novel technique to identify the number of memory channels and the memory channel of a specific address is presented. Our results show significant differences in the access energy of specific GPU cores, while the access energy of the different memory channels from the same GPU cores is almost identical. MEMPower is able to model these differences and provide good predictions of the access energy for specific memory accesses.
Issue Date: 25-Apr-2019
Date Available: 21-Nov-2019
DDC Class: 004 Datenverarbeitung; Informatik
Subject(s): GPU
power modeling
data dependent power
Sponsor/Funder: EC/H2020/688759/EU/Low-Power Parallel Computing on GPUs 2/LPGPU2
Proceedings Title: Architecture of Computing Systems – ARCS 2019 : 32nd International Conference, Copenhagen, Denmark, May 20–23, 2019, Proceedings
Editor: Schoeberl, Martin
Hochberger, Christian
Uhrig, Sascha
Brehm, Jürgen
Pionteck, Thilo
Publisher: Springer
Publisher Place: Cham
Publisher DOI: 10.1007/978-3-030-18656-2_15
Page Start: 195
Page End: 207
Series: Lecture Notes in Computer Science
Series Number: 11479
ISBN: 978-3-030-18656-2
Appears in Collections:FG Architektur eingebetteter Systeme » Publications

Files in This Item:
Format: Adobe PDF | Size: 10.8 MB
DownloadShow Preview

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons