Introducing HPC with a Raspberry Pi Cluster

inside-BigData.com
inside-BigData.comPresident of insideHPC Media um inside-BigData.com
Introducing HPC with a Raspberry Pi Cluster
Colin Sauzé <cos@aber.ac.uk>
Research Software Engineer
Super Computing Wales Project
Aberystwyth University
A practical use of and good excuse to build Raspberry Pi Clusters
Overview
● About Me
● Inspirations
● Why teach HPC with Raspberry Pi?
● My Raspberry Pi cluster
● Experiences from teaching
● Future Work
About Me
● Research Software Engineer with
Supercomputing Wales project
– 4 university partnership to supply
HPC systems
– Two physical HPCs
● PhD in Robotics
– Experience with Linux on single
board computers
– Lots of Raspberry Pi projects
Inspiration #1: Los Alamos National Laboratory
● 750 node cluster
● Test system for software
development
● Avoid tying up the real cluster
Inspiration #2: Wee Archie/Archlet
● EPCC’s Raspberry Pi Cluster
● Archie: 18x Raspberry Pi 2’s
(4 cores each)
● Archlet: smaller 4 or 5 node
clusters.
● Used for outreach demos.
● Setup instructions:
https://github.com/EPCCed/w
ee_archlet
Image from
https://raw.githubusercontent.com/EPCCed/wee_archlet/master/images/IMG_20170210_132818620.jpg
Inspiration #3: Swansea’s Raspberry Pi Cluster
● 16x Raspberry Pi 3s
● CFD demo using a Kinect
sensor
● Demoed at the Swansea
Festival of Science 2018
Why Teach with a Raspberry Pi cluster?
● Avoid loading real clusters doing actual research
– Less fear from learners that they might break something
● Resource limits more apparent
● More control over the environment
● Hardware less abstract
● No need to have accounts on a real HPC
My Cluster
● “Tweety Pi”
– 10x Raspberry Pi model B
version 1s
– 1x Raspberry Pi 3 as
head/login node
– Raspbian Stretch
● Head node acts as WiFi
access point
– Internet via phone or laptop
Demo Software
● British Science Week 2019
– Simple Pi with Monte Carlo methods demo
– MPI based
– GUI to control how many jobs launch and show
queuing
● Swansea CFD demo
– Needs more compute power
– 16x Raspberry Pi 3 vs 10x Raspberry Pi 1
● Wee Archie/Archlet Demos
– Many demos available
●
I only found this recently
– https://github.com/EPCCed/wee_archie
Making a realistic HPC environment
● MPICH
● Slurm
● Quotas on home directories
● NFS mounted home directories
● Software modules
● Network booting compute nodes
Network booting hack
● No PXE boot support on original Raspberry Pi (or Raspberry Pi
B+ and 2)
● Kernel + bootloader on SD card
● Root filesystem on NFS
– Cmdline.txt contains:
● console=tty1 root=/dev/nfs 
nfsroot=10.0.0.10:/nfs/node_rootfs,vers=3 ro ip=dhcp 
elevator=deadline rootwait
● SD cards can be identical, small 50mb image, easy to replace
Teaching Materials
● Based on Introduction to HPC with Super Computing Wales carpentry style
lesson:
– What is an HPC?
– Logging in
– Filesystems and transferring data
– Submitting/monitoring jobs with Slurm
– Profiling
– Parallelising code, Amdahl’s law
– MPI
– HPC Best Practice
Experiences from Teaching – STFC Summer
School
● New PhD students in solar
physics
– Not registered at universities
yet, no academic accounts
● 15 people each time
– 1st time using HPC for many
– Most had some Unix
experience
● Subset of Super Computing
Wales introduction to HPC
carpentry lesson
Feedback
● Very Positive
●
A lot seemed to enjoy playing around with SSH/SCP
– First time using a remote shell for some
– Others more adventurous than they might have been on a real HPC
● Main complaint was lack of time (only 1.5 hours)
– Only got as far as covering basic job submission
– Quick theoretical run through of MPI and Amdahl’s law
– Probably have 3-4 hours of material
●
Queuing became very apparent
– 10 nodes, 15 users
– “watch squeue” running on screen during practical parts
Problems
● Slurm issues on day 1
– Accidentally overwrote a system user when creating accounts
● WiFi via Laptop/phone slow
– When users connect to the cluster its their internet connection too
– Relied on this for access to course notes
Experiences from teaching – Supercomputing
Wales Training
● Approximately 10 people
– Mix of staff and research students
– Mixed experience levels
– All intending to use a real HPC
● Simultaneously used Raspberry Pi and real HPC
– Same commands run on both
● Useful backup system for those with locked accounts
● Feedback good
– Helped make HPC more tangible
Future Work
● Configuration management tool
(Ansible/Chef/Puppet/Salt etc) instead
of script for configuration
● CentOS/Open HPC stack instead of
Raspbian
● Public engagement demo which
focuses on our research
– Analysing satellite imagery
– Simulate the monsters from MonsterLab
(https://monster-lab.org/)
More Information
● Setup instructions and scripts -
https://github.com/colinsauze/pi_cluster
● Teaching material -
https://github.com/SCW-Aberystwyth/Introduction-to-HPC-with-
RaspberryPi
● Email me: cos@aber.ac.uk
1 von 18

Recomendados

CUDA-Python and RAPIDS for blazing fast scientific computing von
CUDA-Python and RAPIDS for blazing fast scientific computingCUDA-Python and RAPIDS for blazing fast scientific computing
CUDA-Python and RAPIDS for blazing fast scientific computinginside-BigData.com
1K views82 Folien
State of ARM-based HPC von
State of ARM-based HPCState of ARM-based HPC
State of ARM-based HPCinside-BigData.com
767 views18 Folien
Exploring the Programming Models for the LUMI Supercomputer von
Exploring the Programming Models for the LUMI Supercomputer Exploring the Programming Models for the LUMI Supercomputer
Exploring the Programming Models for the LUMI Supercomputer George Markomanolis
83 views24 Folien
High-Performance and Scalable Designs of Programming Models for Exascale Systems von
High-Performance and Scalable Designs of Programming Models for Exascale SystemsHigh-Performance and Scalable Designs of Programming Models for Exascale Systems
High-Performance and Scalable Designs of Programming Models for Exascale Systemsinside-BigData.com
2.8K views58 Folien
Introduction to GPUs in HPC von
Introduction to GPUs in HPCIntroduction to GPUs in HPC
Introduction to GPUs in HPCinside-BigData.com
1.8K views17 Folien
Programming Models for Exascale Systems von
Programming Models for Exascale SystemsProgramming Models for Exascale Systems
Programming Models for Exascale Systemsinside-BigData.com
671 views62 Folien

Más contenido relacionado

Was ist angesagt?

Lustre Best Practices von
Lustre Best Practices Lustre Best Practices
Lustre Best Practices George Markomanolis
581 views27 Folien
Utilizing AMD GPUs: Tuning, programming models, and roadmap von
Utilizing AMD GPUs: Tuning, programming models, and roadmapUtilizing AMD GPUs: Tuning, programming models, and roadmap
Utilizing AMD GPUs: Tuning, programming models, and roadmapGeorge Markomanolis
209 views30 Folien
Getting started with AMD GPUs von
Getting started with AMD GPUsGetting started with AMD GPUs
Getting started with AMD GPUsGeorge Markomanolis
154 views34 Folien
TAU E4S ON OpenPOWER /POWER9 platform von
TAU E4S ON OpenPOWER /POWER9 platformTAU E4S ON OpenPOWER /POWER9 platform
TAU E4S ON OpenPOWER /POWER9 platformGanesan Narayanasamy
347 views81 Folien
Microsoft Project Olympus AI Accelerator Chassis (HGX-1) von
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)inside-BigData.com
3.6K views28 Folien
A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S... von
A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S...A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S...
A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S...inside-BigData.com
791 views40 Folien

Was ist angesagt?(20)

Utilizing AMD GPUs: Tuning, programming models, and roadmap von George Markomanolis
Utilizing AMD GPUs: Tuning, programming models, and roadmapUtilizing AMD GPUs: Tuning, programming models, and roadmap
Utilizing AMD GPUs: Tuning, programming models, and roadmap
Microsoft Project Olympus AI Accelerator Chassis (HGX-1) von inside-BigData.com
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
Microsoft Project Olympus AI Accelerator Chassis (HGX-1)
inside-BigData.com3.6K views
A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S... von inside-BigData.com
A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S...A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S...
A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator S...
inside-BigData.com791 views
Energy Efficient Computing using Dynamic Tuning von inside-BigData.com
Energy Efficient Computing using Dynamic TuningEnergy Efficient Computing using Dynamic Tuning
Energy Efficient Computing using Dynamic Tuning
inside-BigData.com454 views
OpenCAPI-based Image Analysis Pipeline for 18 GB/s kilohertz-framerate X-ray ... von Ganesan Narayanasamy
OpenCAPI-based Image Analysis Pipeline for 18 GB/s kilohertz-framerate X-ray ...OpenCAPI-based Image Analysis Pipeline for 18 GB/s kilohertz-framerate X-ray ...
OpenCAPI-based Image Analysis Pipeline for 18 GB/s kilohertz-framerate X-ray ...
Welcome to the 2016 HPC Advisory Council Switzerland Conference von inside-BigData.com
Welcome to the 2016 HPC Advisory Council Switzerland ConferenceWelcome to the 2016 HPC Advisory Council Switzerland Conference
Welcome to the 2016 HPC Advisory Council Switzerland Conference
inside-BigData.com1.1K views
Evaluating GPU programming Models for the LUMI Supercomputer von George Markomanolis
Evaluating GPU programming Models for the LUMI SupercomputerEvaluating GPU programming Models for the LUMI Supercomputer
Evaluating GPU programming Models for the LUMI Supercomputer
Huawei’s requirements for the ARM based HPC solution readiness - Joshua Mora von Linaro
Huawei’s requirements for the ARM based HPC solution readiness - Joshua MoraHuawei’s requirements for the ARM based HPC solution readiness - Joshua Mora
Huawei’s requirements for the ARM based HPC solution readiness - Joshua Mora
Linaro3.7K views
SCFE 2020 OpenCAPI presentation as part of OpenPWOER Tutorial von Ganesan Narayanasamy
SCFE 2020 OpenCAPI presentation as part of OpenPWOER TutorialSCFE 2020 OpenCAPI presentation as part of OpenPWOER Tutorial
SCFE 2020 OpenCAPI presentation as part of OpenPWOER Tutorial
Assisting User’s Transition to Titan’s Accelerated Architecture von inside-BigData.com
Assisting User’s Transition to Titan’s Accelerated ArchitectureAssisting User’s Transition to Titan’s Accelerated Architecture
Assisting User’s Transition to Titan’s Accelerated Architecture
inside-BigData.com1.1K views
OpenHPC: A Comprehensive System Software Stack von inside-BigData.com
OpenHPC: A Comprehensive System Software StackOpenHPC: A Comprehensive System Software Stack
OpenHPC: A Comprehensive System Software Stack
inside-BigData.com3.8K views
Deep Learning on ARM Platforms - SFO17-509 von Linaro
Deep Learning on ARM Platforms - SFO17-509Deep Learning on ARM Platforms - SFO17-509
Deep Learning on ARM Platforms - SFO17-509
Linaro5.3K views

Similar a Introducing HPC with a Raspberry Pi Cluster

uWSGI - Swiss army knife for your Python web apps von
uWSGI - Swiss army knife for your Python web appsuWSGI - Swiss army knife for your Python web apps
uWSGI - Swiss army knife for your Python web appsTomislav Raseta
3.1K views20 Folien
Redfish and python-redfish for Software Defined Infrastructure von
Redfish and python-redfish for Software Defined InfrastructureRedfish and python-redfish for Software Defined Infrastructure
Redfish and python-redfish for Software Defined InfrastructureBruno Cornec
2.3K views17 Folien
Spark China Summit 2015 Guancheng Chen von
Spark China Summit 2015 Guancheng ChenSpark China Summit 2015 Guancheng Chen
Spark China Summit 2015 Guancheng ChenGuancheng (G.C.) Chen
431 views16 Folien
HiPEAC 2019 Tutorial - Maestro RTOS von
HiPEAC 2019 Tutorial - Maestro RTOSHiPEAC 2019 Tutorial - Maestro RTOS
HiPEAC 2019 Tutorial - Maestro RTOSTulipp. Eu
127 views59 Folien
[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen... von
[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen...[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen...
[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen...Wong Hoi Sing Edison
315 views74 Folien
PyData Boston 2013 von
PyData Boston 2013PyData Boston 2013
PyData Boston 2013Travis Oliphant
3.9K views44 Folien

Similar a Introducing HPC with a Raspberry Pi Cluster(20)

uWSGI - Swiss army knife for your Python web apps von Tomislav Raseta
uWSGI - Swiss army knife for your Python web appsuWSGI - Swiss army knife for your Python web apps
uWSGI - Swiss army knife for your Python web apps
Tomislav Raseta3.1K views
Redfish and python-redfish for Software Defined Infrastructure von Bruno Cornec
Redfish and python-redfish for Software Defined InfrastructureRedfish and python-redfish for Software Defined Infrastructure
Redfish and python-redfish for Software Defined Infrastructure
Bruno Cornec2.3K views
HiPEAC 2019 Tutorial - Maestro RTOS von Tulipp. Eu
HiPEAC 2019 Tutorial - Maestro RTOSHiPEAC 2019 Tutorial - Maestro RTOS
HiPEAC 2019 Tutorial - Maestro RTOS
Tulipp. Eu127 views
[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen... von Wong Hoi Sing Edison
[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen...[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen...
[HKOSCON][20180616][Containerized High Availability Virtual Hosting Deploymen...
History of Computer Systems - Why we are doing it that way von Leo Lorieri
History of Computer Systems - Why we are doing it that wayHistory of Computer Systems - Why we are doing it that way
History of Computer Systems - Why we are doing it that way
Leo Lorieri1.8K views
Heavy duty Abaqus structural analysis using HPC in the cloud von hpcexperiment
Heavy duty Abaqus structural analysis using HPC in the cloudHeavy duty Abaqus structural analysis using HPC in the cloud
Heavy duty Abaqus structural analysis using HPC in the cloud
hpcexperiment1.9K views
[BarCamp2018][20180915][Tips for Virtual Hosting on Kubernetes] von Wong Hoi Sing Edison
[BarCamp2018][20180915][Tips for Virtual Hosting on Kubernetes][BarCamp2018][20180915][Tips for Virtual Hosting on Kubernetes]
[BarCamp2018][20180915][Tips for Virtual Hosting on Kubernetes]
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod... von inside-BigData.com
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
ICEOTOPE & OCF: Performance for Manufacturing von IceotopePR
ICEOTOPE & OCF: Performance for Manufacturing ICEOTOPE & OCF: Performance for Manufacturing
ICEOTOPE & OCF: Performance for Manufacturing
IceotopePR131 views
[NetApp Managing Big Workspaces with Storage Magic von Perforce
[NetApp Managing Big Workspaces with Storage Magic[NetApp Managing Big Workspaces with Storage Magic
[NetApp Managing Big Workspaces with Storage Magic
Perforce752 views
New Process/Thread Runtime von Linaro
New Process/Thread Runtime	New Process/Thread Runtime
New Process/Thread Runtime
Linaro2.6K views
Sanger OpenStack presentation March 2017 von Dave Holland
Sanger OpenStack presentation March 2017Sanger OpenStack presentation March 2017
Sanger OpenStack presentation March 2017
Dave Holland655 views
Resilience: the key requirement of a [big] [data] architecture - StampedeCon... von StampedeCon
Resilience: the key requirement of a [big] [data] architecture  - StampedeCon...Resilience: the key requirement of a [big] [data] architecture  - StampedeCon...
Resilience: the key requirement of a [big] [data] architecture - StampedeCon...
StampedeCon1.3K views
Cytoscape and External Data Analysis Tools von Keiichiro Ono
Cytoscape and External Data Analysis ToolsCytoscape and External Data Analysis Tools
Cytoscape and External Data Analysis Tools
Keiichiro Ono1.3K views
Integrating Puppet and Gitolite for sysadmins cooperations von Luca Mazzaferro
Integrating Puppet and Gitolite for sysadmins cooperationsIntegrating Puppet and Gitolite for sysadmins cooperations
Integrating Puppet and Gitolite for sysadmins cooperations
Luca Mazzaferro899 views
Pivotal: Operationalizing 1000 Node Hadoop Cluster - Analytics Workbench von EMC
Pivotal: Operationalizing 1000 Node Hadoop Cluster - Analytics WorkbenchPivotal: Operationalizing 1000 Node Hadoop Cluster - Analytics Workbench
Pivotal: Operationalizing 1000 Node Hadoop Cluster - Analytics Workbench
EMC4.3K views

Más de inside-BigData.com

Major Market Shifts in IT von
Major Market Shifts in ITMajor Market Shifts in IT
Major Market Shifts in ITinside-BigData.com
5K views27 Folien
Preparing to program Aurora at Exascale - Early experiences and future direct... von
Preparing to program Aurora at Exascale - Early experiences and future direct...Preparing to program Aurora at Exascale - Early experiences and future direct...
Preparing to program Aurora at Exascale - Early experiences and future direct...inside-BigData.com
1.4K views45 Folien
Transforming Private 5G Networks von
Transforming Private 5G NetworksTransforming Private 5G Networks
Transforming Private 5G Networksinside-BigData.com
1.3K views14 Folien
The Incorporation of Machine Learning into Scientific Simulations at Lawrence... von
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...inside-BigData.com
1.4K views33 Folien
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ... von
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...inside-BigData.com
709 views53 Folien
HPC Impact: EDA Telemetry Neural Networks von
HPC Impact: EDA Telemetry Neural NetworksHPC Impact: EDA Telemetry Neural Networks
HPC Impact: EDA Telemetry Neural Networksinside-BigData.com
310 views18 Folien

Más de inside-BigData.com(20)

Preparing to program Aurora at Exascale - Early experiences and future direct... von inside-BigData.com
Preparing to program Aurora at Exascale - Early experiences and future direct...Preparing to program Aurora at Exascale - Early experiences and future direct...
Preparing to program Aurora at Exascale - Early experiences and future direct...
inside-BigData.com1.4K views
The Incorporation of Machine Learning into Scientific Simulations at Lawrence... von inside-BigData.com
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
inside-BigData.com1.4K views
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ... von inside-BigData.com
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
inside-BigData.com709 views
Biohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring von inside-BigData.com
Biohybrid Robotic Jellyfish for Future Applications in Ocean MonitoringBiohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring
Biohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring
inside-BigData.com460 views
Fugaku Supercomputer joins fight against COVID-19 von inside-BigData.com
Fugaku Supercomputer joins fight against COVID-19Fugaku Supercomputer joins fight against COVID-19
Fugaku Supercomputer joins fight against COVID-19
inside-BigData.com2.1K views
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD von inside-BigData.com
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPODHPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD
inside-BigData.com838 views
Versal Premium ACAP for Network and Cloud Acceleration von inside-BigData.com
Versal Premium ACAP for Network and Cloud AccelerationVersal Premium ACAP for Network and Cloud Acceleration
Versal Premium ACAP for Network and Cloud Acceleration
inside-BigData.com586 views
Zettar: Moving Massive Amounts of Data across Any Distance Efficiently von inside-BigData.com
Zettar: Moving Massive Amounts of Data across Any Distance EfficientlyZettar: Moving Massive Amounts of Data across Any Distance Efficiently
Zettar: Moving Massive Amounts of Data across Any Distance Efficiently
inside-BigData.com496 views
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc... von inside-BigData.com
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
Efficient Model Selection for Deep Neural Networks on Massively Parallel Proc...
inside-BigData.com249 views
Scientific Applications and Heterogeneous Architectures von inside-BigData.com
Scientific Applications and Heterogeneous ArchitecturesScientific Applications and Heterogeneous Architectures
Scientific Applications and Heterogeneous Architectures
inside-BigData.com165 views

Último

Netmera Presentation.pdf von
Netmera Presentation.pdfNetmera Presentation.pdf
Netmera Presentation.pdfMustafa Kuğu
22 views50 Folien
Digital Personal Data Protection (DPDP) Practical Approach For CISOs von
Digital Personal Data Protection (DPDP) Practical Approach For CISOsDigital Personal Data Protection (DPDP) Practical Approach For CISOs
Digital Personal Data Protection (DPDP) Practical Approach For CISOsPriyanka Aash
171 views59 Folien
NTGapps NTG LowCode Platform von
NTGapps NTG LowCode Platform NTGapps NTG LowCode Platform
NTGapps NTG LowCode Platform Mustafa Kuğu
474 views30 Folien
"Node.js Development in 2024: trends and tools", Nikita Galkin von
"Node.js Development in 2024: trends and tools", Nikita Galkin "Node.js Development in 2024: trends and tools", Nikita Galkin
"Node.js Development in 2024: trends and tools", Nikita Galkin Fwdays
37 views38 Folien
Discover Aura Workshop (12.5.23).pdf von
Discover Aura Workshop (12.5.23).pdfDiscover Aura Workshop (12.5.23).pdf
Discover Aura Workshop (12.5.23).pdfNeo4j
20 views55 Folien
Evaluation of Quality of Experience of ABR Schemes in Gaming Stream von
Evaluation of Quality of Experience of ABR Schemes in Gaming StreamEvaluation of Quality of Experience of ABR Schemes in Gaming Stream
Evaluation of Quality of Experience of ABR Schemes in Gaming StreamAlpen-Adria-Universität
44 views34 Folien

Último(20)

Digital Personal Data Protection (DPDP) Practical Approach For CISOs von Priyanka Aash
Digital Personal Data Protection (DPDP) Practical Approach For CISOsDigital Personal Data Protection (DPDP) Practical Approach For CISOs
Digital Personal Data Protection (DPDP) Practical Approach For CISOs
Priyanka Aash171 views
NTGapps NTG LowCode Platform von Mustafa Kuğu
NTGapps NTG LowCode Platform NTGapps NTG LowCode Platform
NTGapps NTG LowCode Platform
Mustafa Kuğu474 views
"Node.js Development in 2024: trends and tools", Nikita Galkin von Fwdays
"Node.js Development in 2024: trends and tools", Nikita Galkin "Node.js Development in 2024: trends and tools", Nikita Galkin
"Node.js Development in 2024: trends and tools", Nikita Galkin
Fwdays37 views
Discover Aura Workshop (12.5.23).pdf von Neo4j
Discover Aura Workshop (12.5.23).pdfDiscover Aura Workshop (12.5.23).pdf
Discover Aura Workshop (12.5.23).pdf
Neo4j20 views
Innovation & Entrepreneurship strategies in Dairy Industry von PervaizDar1
Innovation & Entrepreneurship strategies in Dairy IndustryInnovation & Entrepreneurship strategies in Dairy Industry
Innovation & Entrepreneurship strategies in Dairy Industry
PervaizDar139 views
What is Authentication Active Directory_.pptx von HeenaMehta35
What is Authentication Active Directory_.pptxWhat is Authentication Active Directory_.pptx
What is Authentication Active Directory_.pptx
HeenaMehta3515 views
Transcript: Redefining the book supply chain: A glimpse into the future - Tec... von BookNet Canada
Transcript: Redefining the book supply chain: A glimpse into the future - Tec...Transcript: Redefining the book supply chain: A glimpse into the future - Tec...
Transcript: Redefining the book supply chain: A glimpse into the future - Tec...
BookNet Canada43 views
Mobile Core Solutions & Successful Cases.pdf von IPLOOK Networks
Mobile Core Solutions & Successful Cases.pdfMobile Core Solutions & Successful Cases.pdf
Mobile Core Solutions & Successful Cases.pdf
IPLOOK Networks16 views
PCCC23:日本AMD株式会社 テーマ1「AMD Instinct™ アクセラレーターの概要」 von PC Cluster Consortium
PCCC23:日本AMD株式会社 テーマ1「AMD Instinct™ アクセラレーターの概要」PCCC23:日本AMD株式会社 テーマ1「AMD Instinct™ アクセラレーターの概要」
PCCC23:日本AMD株式会社 テーマ1「AMD Instinct™ アクセラレーターの概要」
Business Analyst Series 2023 - Week 4 Session 8 von DianaGray10
Business Analyst Series 2023 -  Week 4 Session 8Business Analyst Series 2023 -  Week 4 Session 8
Business Analyst Series 2023 - Week 4 Session 8
DianaGray10180 views
This talk was not generated with ChatGPT: how AI is changing science von Elena Simperl
This talk was not generated with ChatGPT: how AI is changing scienceThis talk was not generated with ChatGPT: how AI is changing science
This talk was not generated with ChatGPT: how AI is changing science
Elena Simperl34 views
The Power of Generative AI in Accelerating No Code Adoption.pdf von Saeed Al Dhaheri
The Power of Generative AI in Accelerating No Code Adoption.pdfThe Power of Generative AI in Accelerating No Code Adoption.pdf
The Power of Generative AI in Accelerating No Code Adoption.pdf
Saeed Al Dhaheri44 views
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And... von ShapeBlue
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...
Enabling DPU Hardware Accelerators in XCP-ng Cloud Platform Environment - And...
ShapeBlue120 views
Future of AR - Facebook Presentation von Rob McCarty
Future of AR - Facebook PresentationFuture of AR - Facebook Presentation
Future of AR - Facebook Presentation
Rob McCarty66 views

Introducing HPC with a Raspberry Pi Cluster

  • 1. Introducing HPC with a Raspberry Pi Cluster Colin Sauzé <cos@aber.ac.uk> Research Software Engineer Super Computing Wales Project Aberystwyth University A practical use of and good excuse to build Raspberry Pi Clusters
  • 2. Overview ● About Me ● Inspirations ● Why teach HPC with Raspberry Pi? ● My Raspberry Pi cluster ● Experiences from teaching ● Future Work
  • 3. About Me ● Research Software Engineer with Supercomputing Wales project – 4 university partnership to supply HPC systems – Two physical HPCs ● PhD in Robotics – Experience with Linux on single board computers – Lots of Raspberry Pi projects
  • 4. Inspiration #1: Los Alamos National Laboratory ● 750 node cluster ● Test system for software development ● Avoid tying up the real cluster
  • 5. Inspiration #2: Wee Archie/Archlet ● EPCC’s Raspberry Pi Cluster ● Archie: 18x Raspberry Pi 2’s (4 cores each) ● Archlet: smaller 4 or 5 node clusters. ● Used for outreach demos. ● Setup instructions: https://github.com/EPCCed/w ee_archlet Image from https://raw.githubusercontent.com/EPCCed/wee_archlet/master/images/IMG_20170210_132818620.jpg
  • 6. Inspiration #3: Swansea’s Raspberry Pi Cluster ● 16x Raspberry Pi 3s ● CFD demo using a Kinect sensor ● Demoed at the Swansea Festival of Science 2018
  • 7. Why Teach with a Raspberry Pi cluster? ● Avoid loading real clusters doing actual research – Less fear from learners that they might break something ● Resource limits more apparent ● More control over the environment ● Hardware less abstract ● No need to have accounts on a real HPC
  • 8. My Cluster ● “Tweety Pi” – 10x Raspberry Pi model B version 1s – 1x Raspberry Pi 3 as head/login node – Raspbian Stretch ● Head node acts as WiFi access point – Internet via phone or laptop
  • 9. Demo Software ● British Science Week 2019 – Simple Pi with Monte Carlo methods demo – MPI based – GUI to control how many jobs launch and show queuing ● Swansea CFD demo – Needs more compute power – 16x Raspberry Pi 3 vs 10x Raspberry Pi 1 ● Wee Archie/Archlet Demos – Many demos available ● I only found this recently – https://github.com/EPCCed/wee_archie
  • 10. Making a realistic HPC environment ● MPICH ● Slurm ● Quotas on home directories ● NFS mounted home directories ● Software modules ● Network booting compute nodes
  • 11. Network booting hack ● No PXE boot support on original Raspberry Pi (or Raspberry Pi B+ and 2) ● Kernel + bootloader on SD card ● Root filesystem on NFS – Cmdline.txt contains: ● console=tty1 root=/dev/nfs  nfsroot=10.0.0.10:/nfs/node_rootfs,vers=3 ro ip=dhcp  elevator=deadline rootwait ● SD cards can be identical, small 50mb image, easy to replace
  • 12. Teaching Materials ● Based on Introduction to HPC with Super Computing Wales carpentry style lesson: – What is an HPC? – Logging in – Filesystems and transferring data – Submitting/monitoring jobs with Slurm – Profiling – Parallelising code, Amdahl’s law – MPI – HPC Best Practice
  • 13. Experiences from Teaching – STFC Summer School ● New PhD students in solar physics – Not registered at universities yet, no academic accounts ● 15 people each time – 1st time using HPC for many – Most had some Unix experience ● Subset of Super Computing Wales introduction to HPC carpentry lesson
  • 14. Feedback ● Very Positive ● A lot seemed to enjoy playing around with SSH/SCP – First time using a remote shell for some – Others more adventurous than they might have been on a real HPC ● Main complaint was lack of time (only 1.5 hours) – Only got as far as covering basic job submission – Quick theoretical run through of MPI and Amdahl’s law – Probably have 3-4 hours of material ● Queuing became very apparent – 10 nodes, 15 users – “watch squeue” running on screen during practical parts
  • 15. Problems ● Slurm issues on day 1 – Accidentally overwrote a system user when creating accounts ● WiFi via Laptop/phone slow – When users connect to the cluster its their internet connection too – Relied on this for access to course notes
  • 16. Experiences from teaching – Supercomputing Wales Training ● Approximately 10 people – Mix of staff and research students – Mixed experience levels – All intending to use a real HPC ● Simultaneously used Raspberry Pi and real HPC – Same commands run on both ● Useful backup system for those with locked accounts ● Feedback good – Helped make HPC more tangible
  • 17. Future Work ● Configuration management tool (Ansible/Chef/Puppet/Salt etc) instead of script for configuration ● CentOS/Open HPC stack instead of Raspbian ● Public engagement demo which focuses on our research – Analysing satellite imagery – Simulate the monsters from MonsterLab (https://monster-lab.org/)
  • 18. More Information ● Setup instructions and scripts - https://github.com/colinsauze/pi_cluster ● Teaching material - https://github.com/SCW-Aberystwyth/Introduction-to-HPC-with- RaspberryPi ● Email me: cos@aber.ac.uk