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

09/13/2020
by   Karel Adámek, et al.
0

The Square Kilometre 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 paper, 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 V100 GPU when computing the FFT by up to 60 than a 10 clock frequency for all tested FFT lengths, we show on average a 50 in power consumption compared to the boost core clock frequency with an increase in the execution time still below 10 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.

READ FULL TEXT

page 3

page 9

research
06/26/2021

On the Impact of Device-Level Techniques on Energy-Efficiency of Neural Network Accelerators

Energy-efficiency is a key concern for neural network applications. To a...
research
04/17/2020

A Data-Driven Frequency Scaling Approach for Deadline-aware Energy Efficient Scheduling on Graphics Processing Units (GPUs)

Modern computing paradigms, such as cloud computing, are increasingly ad...
research
05/19/2018

Partitioning SKA Dataflows for Optimal Graph Execution

Optimizing data-intensive workflow execution is essential to many modern...
research
08/11/2014

Physical Computing With No Clock to Implement the Gaussian Pyramid of SIFT Algorithm

Physical computing is a technology utilizing the nature of electronic de...
research
11/24/2022

Cutting the cost of pulsar astronomy: Saving time and energy when searching for binary pulsars using NVIDIA GPUs

Using the Fourier Domain Acceleration Search (FDAS) method to search for...
research
06/01/2020

Exceeding Conservative Limits: A Consolidated Analysis on Modern Hardware Margins

Modern large-scale computing systems (data centers, supercomputers, clou...
research
06/19/2020

Run-Time Power Modelling in Embedded GPUs with Dynamic Voltage and Frequency Scaling

This paper investigates the application of a robust CPU-based power mode...

Please sign up or login with your details

Forgot password? Click here to reset