Session: ESPM2 2020: Fifth International Workshop on Extreme Scale Programming Models and Middleware

Deploying a Task-Based Runtime System on Raspberry Pi Clusters

Wednesday, November 11, 2020

12:30 PM – 1:00 PM ET

Channel: Track 2

Author/Presenter(s)

NG

Nikunj Gupta

Indian Institute of Technology Roorkee, India
SB

Steve R. Brandt

Louisiana State University, United States of America
BW

Bibek Wagle

Louisiana State University, United States of America
NW

Nanmiao Wu

Louisiana State University, Center for Computation and Technology, United States of America
AK

Alireza Kheirkhahan

Louisiana State University, United States of America
Patrick Diehl

Louisiana State University; Louisiana State University, Center for Computation and Technology, United States of America
FB

Felix W. Baumann

PricewaterhouseCoopers LLP, Germany
Hartmut Kaiser

Louisiana State University; Louisiana State University, Center for Computation and Technology, United States of America

Arm~technology is becoming increasingly important in HPC. Recently, Fugaku, an Arm-based system, was awarded the number one place in the Top500 list. Raspberry Pis provide an inexpensive platform to become familiar with this architecture. However, Pis can also be useful on their own. Here we describe our efforts to configure and benchmark the use of a Raspberry Pi cluster with the HPX/Phylanx platform (normally intended for use with HPC applications) and document the lessons we learned. First, we highlight the required changes in the configuration of the Pi to gain performance. Second, we explore how limited memory bandwidth limits the use of all cores in our shared memory benchmarks. Third, we evaluate whether low network bandwidth affects distributed performance. Fourth, we discuss the power consumption and the resulting trade-off in cost of operation and performance.