Haupt-Navigation

Page Content

Publications

MEMPower: Data-Aware GPU Memory Power Model
Citation key	10.1007/978-3-030-18656-2_15
Author	Lucas, Jan and Juurlink, Ben
Title of Book	Architecture of Computing Systems – ARCS 2019
Pages	195–207
Year	2019
ISBN	978-3-030-18656-2
Address	Cham
Editor	Schoeberl, Martin and Hochberger, Christian and Uhrig, Sascha and Brehm, Jürgen and Pionteck, Thilo
Publisher	Springer International Publishing
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.

Download Bibtex entry [1]

Back [2]

Older Publications

Further Publications of Prof. Juurlink are here [3]available.

Repository:

Repository for Research Data and Publications at the TUB-Library: DepositOnce

------ Links: ------

[1] https://www.aes.tu-berlin.de/menue/research/publica
tions/parameter/en/maxhilfe/?no_cache=1&tx_sibibtex
_pi1%5Bdownload_bibtex_uid%5D=4621259&tx_sibibtex_p
i1%5Bcontentelement%5D=tt_content%3A324193

[2] https://www.aes.tu-berlin.de/menue/research/publica
tions/parameter/en/maxhilfe/

[3] https://www.aes.tu-berlin.de/menue/research/publica
tions/older_publications/parameter/en/maxhilfe/