The POWDER Manual

2020-07-23 (06836dc)

← prev up next →

The POWDER Manual

The Powder Team

Powder is a facility for experimenting on the future of wireless networking in a city-scale "living laboratory" environment.

Powder is run by the University of Utah in partnership with Salt Lake City and the Utah Education and Telehealth Network.

The Powder facility is built on top of Emulab and is run by the Flux Research Group, part of the School of Computing at the University of Utah.

1 Powder Overview

1.1 What is Powder

1.2 Powder status

1.3 Roadmap to using Powder

2 Getting Started

3 Powder Over-the-air operation

3.2 Spectrum use

3.3 Step-by-step walkthrough

4 Powder starter profiles

5.1 Register for an Account

5.1.1 Join an existing project

5.1.2 Create a new project

5.1.3 Setting up SSH access

5.1.4 Setting up X11

6 Powder and Repeatable Research

7 Basic Concepts

7.1.1 On-demand Profiles

7.1.2 Persistent Profiles

7.2 Experiments

7.2.1 Extending Experiments

7.4 Physical Machines

7.5 Virtual Machines and Containers

8 Creating Profiles

8.1 Creating a profile from an existing one

8.1.1 Preparation and precautions

8.1.2 Cloning a Profile

8.1.3 Copying a Profile

8.1.4 Creating the Profile

8.1.5 Updating a profile

8.2 Repository-Based Profiles

8.2.1 Updating Repository-Based Profiles

8.2.2 Branches and Tags in Repository-Based Profiles

8.3 Creating a profile from scratch

8.4 Sharing Profiles

8.5 Versioned Profiles

9 Resource Reservations

9.1 What Reservations Guarantee

9.2 How Reservations May Affect You

9.3 Making a Reservation

9.4 Using a Reservation

10 Describing a profile with python and geni-lib

10.1 A single XEN VM node

10.2 A single physical host

10.3 Two XenVM nodes with a link between them

10.4 Two ARM64 servers in a LAN

10.5 A VM with a custom size

10.6 Set a specific IP address on each node

10.7 RF communication

10.8 Specify an operating system and set install and execute scripts

10.9 Profiles with user-specified parameters

10.10 Add storage to a node

10.11 Debugging geni-lib profile scripts

11 Advanced Topics

11.1 Disk Images

11.2 Storage Mechanisms

11.2.1 Overview of Storage Mechanisms

11.2.2 Node-Local Storage

11.2.2.1 Specifying Storage in a Profile – Local Datasets

11.2.2.2 Allocating Storage in a Running Experiment

11.2.2.3 Persisting Local Data

11.2.3 Image-backed Datasets

11.2.4 Remote Datasets

11.2.5 NFS Shared Filesystems

11.2.6 Write-back Storage for Fixed and Mobile Endpoints

11.2.7 Storage Type Summary (TL;DR)

11.2.8 Example Storage Profiles

11.2.8.1 Creating a Node-local Dataset

11.2.8.2 Creating an Image-backed Dataset from a Node-local Dataset

11.2.8.3 Using and Updating an Image-backed Dataset

11.2.8.4 Creating a Remote Dataset

11.2.8.5 Using a Remote Dataset on a Single Node

11.2.8.6 Using a Remote Dataset on Multiple Nodes via a Shared Filesystem

11.2.8.7 Using a Remote Dataset on Multiple Nodes via Clones

11.4 Public IP Access

11.4.1 Dynamic Public IP Addresses

11.6 Introspection

11.6.1 Client ID

11.6.2 Control MAC

11.6.3 Manifest

11.6.4 Private key

11.6.5 Profile parameters

11.7 User-controlled switches and layer-1 topologies

11.8 Portal API

12 Virtual Machines and Containers

12.1.1 Controlling CPU and Memory

12.1.2 Controlling Disk Space

12.1.3 Setting HVM Mode

12.2 Docker Containers

12.2.1 Basic Examples

12.2.2 Disk Images

12.2.3 External Images

12.2.4 Dockerfiles

12.2.5 Augmented Disk Images

12.2.6 Remote Access

12.2.8 ENTRYPOINT and CMD

12.2.9 Shared Containers

12.2.10 Privileged Containers

12.2.11 Remote Blockstores

12.2.12 Temporary Block Storage

12.2.13 DockerContainer Member Variables

13.1 Base-station resources

13.2 Fixed-endpoint resources

13.3 Skylark IRIS Endpoints

13.4 Near-edge computing resources

13.5 Cloud computing resources

14 Planned Features

14.1 Improved Physical Resource Descriptions

15 Powder basic srsLTE Tutorial

15.1 Objectives

15.2 Prerequisites

15.3 Logging In

15.4 Creating a simple srsLTE experiment

15.5 Exploring Your Experiment

15.5.1 Experiment Status

15.5.2 Profile Instructions

15.5.3 Topology View

15.5.4 List View

15.5.5 Manifest View

15.5.6 Graphs View

15.5.8 Web-based Shell

15.6 Using the srsLTE tools

15.7 Digging deeper

15.8 Terminating the Experiment

15.9 Taking Next Steps

16 Powder OAI Tutorial

16.1 Objectives

16.2 Prerequisites

16.3 Logging In

16.4 Building Your Own OAI Network

16.5 Exploring Your Experiment

16.5.1 Experiment Status

16.5.2 Profile Instructions

16.5.3 Topology View

16.5.4 List View

16.5.5 Manifest View

16.5.6 Graphs View

16.5.8 Web-based Shell

16.6 Starting OAI Services

16.7 Connecting the UE

16.8 In-Depth OAI Profile Documentation

16.9 Terminating the Experiment

16.10 Taking Next Steps

17 Powder OpenStack Tutorial

17.1 Objectives

17.2 Prerequisites

17.3 Logging In

17.4 Building Your Own OpenStack Cloud

17.5 Exploring Your Experiment

17.5.1 Experiment Status

17.5.2 Profile Instructions

17.5.3 Topology View

17.5.4 List View

17.5.5 Manifest View

17.5.6 Graphs View

17.5.8 Web-based Shell

17.5.9 Serial Console

17.6 Bringing up Instances in OpenStack

17.7 Administering OpenStack

17.7.1 Log Into The Control Nodes

17.7.2 Reboot the Compute Node

17.8 Terminating the Experiment

17.9 Taking Next Steps

18 Getting Help

← prev up next →