J 2021

Efficiency Near the Edge: Increasing the Energy Efficiency of FFTs on GPUs for Real-Time Edge Computing

ADÁMEK, Karel, Jan NOVOTNÝ, Jeyarajan THIYAGALINGAM a Wesley ARMOUR

Základní údaje

Originální název

Efficiency Near the Edge: Increasing the Energy Efficiency of FFTs on GPUs for Real-Time Edge Computing

Autoři

ADÁMEK, Karel (203 Česká republika), Jan NOVOTNÝ (203 Česká republika, domácí), Jeyarajan THIYAGALINGAM a Wesley ARMOUR

Vydání

IEEE Access, 2021, 2169-3536

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Odkazy

Kód RIV

RIV/47813059:19630/21:A0000169

Organizační jednotka

Fyzikální ústav v Opavě

UT WoS

000615033700001

Klíčová slova anglicky

Graphics processing units;Clocks;Hardware;Libraries;Power demand;Data processing;Real-time systems;Energy efficiency;high performance computing;real-time systems;parallel architectures;fast Fourier transforms

Štítky

Příznaky

Mezinárodní význam, Recenzováno
Změněno: 11. 3. 2022 11:30, Mgr. Pavlína Jalůvková

Anotace

V originále

The Square Kilometer Array (SKA) is an international initiative for developing the world's largest radio telescope with a total collecting area of over a million square meters. The scale of the operation, combined with the remote location of the telescope, requires the use of energy-efficient computational algorithms. This, along with the extreme data rates that will be produced by the SKA and the requirement for a real-time observing capability, necessitates in-situ data processing in an edge style computing solution. More generally, energy efficiency in the modern computing landscape is becoming of paramount concern. Whether it be the power budget that can limit some of the world's largest supercomputers, or the limited power available to the smallest Internet-of-Things devices. In this article, we study the impact of hardware frequency scaling on the energy consumption and execution time of the Fast Fourier Transform (FFT) on NVIDIA GPUs using the cuFFT library. The FFT is used in many areas of science and it is one of the key algorithms used in radio astronomy data processing pipelines. Through the use of frequency scaling, we show that we can lower the power consumption of the NVIDIA A100 GPU when computing the FFT by up to 47% compared to the boost clock frequency, with less than a 10% increase in the execution time. Furthermore, using one common core clock frequency for all tested FFT lengths, we show on average a 43% reduction in power consumption compared to the boost core clock frequency with an increase in the execution time still below 10%. We demonstrate how these results can be used to lower the power consumption of existing data processing pipelines. These savings, when considered over years of operation, can yield significant financial savings, but can also lead to a significant reduction of greenhouse gas emissions.