Sign In Start Free Trial

Book Overview & Buying
Table Of Contents
Feedback & Rating

Build Supercomputers with Raspberry Pi 3

By : Carlos R. Morrison

4.3 (7)

Build Supercomputers with Raspberry Pi 3

4.3 (7)

By: Carlos R. Morrison

Overview of this book

Author Carlos R. Morrison (Staff Scientist, NASA) will empower the uninitiated reader to quickly assemble and operate a Pi3 supercomputer in the shortest possible time. The lifeblood of a supercomputer, the MPI code, is introduced early, and sample MPI code provides additional practice opportunities for you to test the effectiveness of your creation. You will learn how to configure various nodes and switches so that they can effectively communicate with each other. By the end of this book, you will have successfully built a supercomputer and the various applications related to it.

Preface

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Reader feedback

Customer support

Free Chapter

1. Getting Started with Supercomputing

1. Getting Started with Supercomputing

Von Neumann architecture

Flynn's classical taxonomy

Historical perspective

The need for greater processing speed

Additional analytical perspective

Sources for reference

Summary

2. One Node Supercomputing

2. One Node Supercomputing

Linux installation

PC processor

Accessing processor technical details

Write/run serial π code

Message passing interface

Summary

3. Preparing the Initial Two Nodes

3. Preparing the Initial Two Nodes

Listing of parts

The Pi2/Pi3 computer

Project overview

Super stack assembly

Preparing the master node

Transferring the code

Preparing the slave node

Summary

4. Static IP Address and Hosts File Setup

4. Static IP Address and Hosts File Setup

Configuring static IP address of the master Pi

Configuring a network switch static IP address

Hosts file setup

Summary

5. Creating a Common User for All Nodes

5. Creating a Common User for All Nodes

Adding a new user for all nodes

ID key generation

ID key transfer

Summary

6. Creating a Mountable Drive on the Master Node

6. Creating a Mountable Drive on the Master Node

Summary

7. Configuring the Eight Nodes

7. Configuring the Eight Nodes

Automating mounting of drives

Setting up the host file for all nodes

Formatting the remaining slave SD cards

Copying the slave1 SD card image to the main computer drive

Copying the slave1 image to the remaining slave SD cards

Summary

8. Testing the Super Cluster

8. Testing the Super Cluster

Wielding the -H command

Pi2 supercomputing

Pi3 supercomputing

Creating bash files

Using unrestrained MPI code logic

Summary

9. Real-World Math Application

9. Real-World Math Application

MPI Taylor series sine(x) function

MPI Taylor series cosine(x) function

MPI Taylor series tan(x) function

MPI Taylor series ln(x) function

Summary

10. Real-World Physics Application

10. Real-World Physics Application

MPI concurrent wave equation and code

Summary

11. Real-World Engineering Application

11. Real-World Engineering Application

MPI Fourier series sawtooth signal

Summary

A. Appendix

A. Appendix

Definitions

Customer Reviews

4.3 (7)

5 star

71.4%

4 star

14.3%

3 star

0

2 star

0

1 star

14.3%

Summary

In this chapter, we learned how to test your supercomputer by first shutting down the entire cluster by using the shutdown -h now command at each node, and then reenergizing the super cluster to reinitialize the machine. This procedure was then followed by commanding multiple nodes simultaneously to demolish the time needed to solve the π equation. Recall that one core on the master Pi2 node, operating in the restrained mode, and using 300,000 iterations, took 58m2.393s to calculate π, and 6m58.102s using all 32 cores in the eight node cluster, while the Pi3 supercomputer, also operating in the restrained mode, took 39m34.726s to calculate π using one core, and only 5m8.955s using 32 cores, and an insanely 3m51.196s using all 64 cores on the 16 node Pi3 machine. The author leaves it up to the reader to determine the time needed to crunch the value of π using 5 million (5,000,000) iterations, and all 64 cores in the unrestrained Pi3...

Search

Your notes and bookmarks