Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-7242
Main Title: ALUPower: Data Dependent Power Consumption in GPUs - Research Data
Author(s): Lucas, Jan
Juurlink, Ben
Type: Generic Research Data
Language Code: en
Abstract: Existing architectural power models for GPUs count activities such as executing floating point or integer instructions, but do not consider the data values processed. While data value dependent power consumption can often be neglected when performing architectural simulations of high performance Out-of-Order (OoO) CPUs, in our related paper we show that this approach is invalid for estimating the power consumption of GPUs. The throughput processing approach of GPUs reduces the amount of control logic and shifts the area and power budget towards functional units and register files. This makes accurate estimations of the power consumption of functional units even more crucial than in OoO CPUs. Using measurements from actual GPUs, we have shown that the processed data values influence the energy consumption of GPUs significantly. This file provides the open research data for this paper, including the coefficients employed by our power model.
URI: https://depositonce.tu-berlin.de//handle/11303/8081
http://dx.doi.org/10.14279/depositonce-7242
Issue Date: 2016
Date Available: 7-Aug-2018
DDC Class: 004 Datenverarbeitung; Informatik
Subject(s): data dependent power
GPU
Power modelling
Sponsor/Funder: EC/H2020/688759/EU/Low-Power Parallel Computing on GPUs 2/LPGPU2
License: http://rightsstatements.org/vocab/InC/1.0/
Referenced By: http://dx.doi.org/10.14279/depositonce-7074
Appears in Collections:FG Architektur eingebetteter Systeme » Research Data

Files in This Item:
File Description SizeFormat 
ALUPower_Data.csvRaw Research Data from ALUPower: Data Dependent Power Consumption in GPUs4.14 kBCSVView/Open


Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.