This hands-on tutorial will introduce attendees to GENI, a nationwide virtual laboratory for networking and distributed systems research (http://www.geni.net). GENI is funded by the National Science Foundation and is free to use for research and education.

The tutorial will be split into two sessions and it is strongly recommended that attendees attend both sessions. Attendees can attend only the second session as long as they meet the pre-requisites described below.

Session 1 (Morning):

1. An Introduction to GENI as a Cloud Computing testbed. 
   This talk will include an overview of GENI and GENI concepts including: a demonstration of an experiment on GENI; a description of the various cloud computing, software defined networking, and wireless resources available to experimenters using GENI; and the current deployment status of GENI. Duration: ~60 minutes
2. A Hands-on Introduction to Experimentation with GENI.
   Attendees will gain an understanding of the GENI experimenter workflow by running a simple experiment on GENI. Duration: ~100 minutes

   Pre-requisites for Session 1

   • Comfort with Unix CLI, SSH with public private keys
   • Bring a laptop for the hands-on portion

Session 2 (Afternoon):

1. GENI and Cloud Computing.
   A presentation about how GENI can be used in cloud computing research through example use cases. This presentation will also touch on collaboration efforts with other testbeds. Duration: ~20 minutes
2. A Hands-on SDN Experiment with GENI.
   Attendees will work through an OpenFlow experiment using GENI resources. We assume no prior OpenFlow knowledge and the tutorial will be adjustable for various levels of SDN understanding. Attendees will also get a chance to write their own OpenFlow controller. Duration: ~120 minutes
   
3. Running your own Experiment in GENI. 
   Advice to experimenters for using GENI for their own research: identifying GENI compute and networking resources best suited for their experiment and guidance on debugging their experiments.  A description of available resources and a presentation of further GENI capabilities and tools. Duration: ~20 minutes.

   Pre-requisites for Session 2 without Session 1

   • Comfort with Unix CLI, SSH with public private keys (required)
   • Setup of GENI account, instructions are going to be sent closer to the conference (required)
   • Review the Introductory slides and complete the hands-on tutorial of the first session (optional)
   • GENIEducation/Resources/IntroToGENI.pptx
   • GENIEducation/SampleAssignments/LabZero

Presenters Bios:

Dr. Niky Riga has received her Diploma from NTUA, Greece and her MA and PhD from Boston University. She is a Network Scientist at Raytheon BBN Technologies. Since 2010 she is a member of the GENI Project Office and is responsible for supporting GENI experimenters in integrating and deploying their experiments within the GENI infrastructure and ensuring that the deployment makes the best use of GENI resources. Her research focuses on designing and prototyping pioneering transport services for Mobile Ad-hoc Networks. Dr. Riga has led several tutorials in major networking and system conferences.

Sarah Edwards received her BS in Electrical Engineering from Rice University and her MS from the University of California, San Diego. She is a Scientist at Raytheon BBN Technologies. Since 2010 she has been a member of the GENI Project Office where she helps experimenters and instructors use GENI and where she writes software to help experimenters use GENI more easily. Previously, Ms Edwards spent several years designing experiments to validate the design of large systems.

This tutorial will present how the Grid’5000 platform can be used to deploy and experiment your own Cloud infrastructure at a large scale. Grid’5000 is a scientific instrument designed to support experiment- driven research in all areas of computer science related to parallel, large-scale or distributed computing and networking. Grid’5000 is supported by French major scientific institutions (INRIA, CNRS, ...). Key features of Grid’5000 platform are ability to conduct large-scale and reproducible experiments, while providing to users a high capacity of customization at every level of computer stack (from hardware to high level applications). Further information on https://www.grid5000.fr

During the morning, the tutorial will show how to get started using Grid’5000, and will describe the virtualization and Cloud-related tools it provides. In the afternoon, the tutorial will present how to perform large-scale deployment of virtual machines and Cloud infrastructures, thanks to reconfigurability capability of Grid’5000.

Session 1 (Morning):

The first part of the morning session will be dedicated to the general Grid’5000 usage. After a quick presentation of the testbed, users will learn how to obtain a Grid’5000 account and connect to the platform by configuring their SSH environment. They will learn how to connect to different Grid’5000 sites and to discover, visualize and reserving Grid’5000 resources. Then the tutorial will present how to get root access on nodes by deploying and customizing their own experimental environment.

The second part of the morning session will present how to use KVM on Grid’5000 nodes production environment. First, attendees will learn how to reserve subnets dedicated to virtual machines. Then, it will be demonstrated how to deploy, run and connect to the virtual machines. Contextualization, and cross-site experiments will also be discussed.

Session 2 (Afternoon):

During the afternoon session, the tutorial will show how to deploy OpenStack in an isolated network. First, people will learn how to reserve an isolated VLAN and to deploy the nodes in this VLAN. Then, automatic tools for OpenStack deployment will be presented and people will learn how to access to its Dashboard.

The last tutorial topic will be dedicated to the automatic deployment of custom physical hosts and virtual machines to let attendees perform experiments with their own environments. An automatic experimental engine that performs reservation, deployment and user-defined workflow will finally be presented.

Prerequisites: Participants that do not follow the morning session require a Grid’5000 account and basic knowledge of platform usage. An on-line tutorial is available at https://www.grid5000.fr/mediawiki/index.php/Getting_Started. A laptop with SSH client is also required.

Presenters Bios:

Dr. Simon Delamare is a research engineer from CNRS, working inside the LIP laboratory at ENS Lyon, France. He received his Master at Université Paris 6 and his Ph.D. at Telecom ParisTech in 2010. He is the current operational manager of the technical team of Grid’5000, the large computer science experimentation tested originated from France. His favorite research topics include quality of service and reliability of distributed computing infrastructures.

Dr. Laurent Pouilloux is a research engineer from INRIA, also working inside the LIP laboratory at ENS Lyon, France. He received his Ph.D. at Institut de Physique du Globe de Paris in 2007 and has a strong experience in teaching computer science. He is mainly involved in helping Grid’5000 users to setup large-scale experiments, especially for virtual machines deployment and cloud computing experiments automation.