= GENI Quarterly Status Report - March 2010 = 

'''Project:''' GENI Experiment Control Using Gush

'''PIs:'''  	 Jeannie Albrecht, Williams College; Amin Vahdat, UC San Diego


== 1. Major Accomplishments ==
   
The main accomplishments this quarter were the first official release of Nebula, the Gush GUI, and the acceptance of a Gush experiences paper to TridentCom.  One of my students, Danny Huang, and I will travel to Germany in May to present the paper.  The paper highlights the general functionality of Gush, and describes our early integration efforts with ORCA and ProtoGENI.  In addition, we added Raven integration to Gush and demoed that functionality at GEC 7.

=== 1.1. Milestones Achieved ===

a.  Demo - Although my demo did not go exactly as planned at GEC 7 due to technical difficulties with my laptop, I was fortunate to be sitting right next to the NetKarma group, who were also demoing Gush.  I was also able to demo Raven integration in Gush until my laptop stopped working.

b.  Raven integration - Gush now has the ability to use Raven for package management.  Users simply have to specify what Raven package they want installed, and Gush/Raven takes care of configuring the slice automatically when an experiment begins.

c.  Gush release - Nebula (the Gush GUI) was released (it is available through svn in the usual place).

=== 1.2 Deliverables Made ===

The past three months of work on Gush have resulted in progress towards three of our deliverables: code development, code release, and documentation.  In terms of code development, we finished integrating Raven into Gush for more efficient package management.  In terms of code release and documentation, we continued to update our website with additional details for developers, and officially released Nebula.

In addition to code development and release, I also had a paper accepted to TridentCom.  The paper describes how we have integrated Gush with resources from three control frameworks: PlanetLab, ORCA, and ProtoGENI.  (Note that this integration is still very much a work in progress, but basic functionality does exist.)  I am excited to have the opportunity to present this work in Germany in May.


== 2.  Description of Work Performed ==

=== 2.1 Activities and Findings ===

The main activities this quarter involved integrating with Raven, releasing Nebula, and writing the camera-ready version of the paper for TridentCom.  I also have been in communication with Justin Cappos at UW about possibly sending my student (Danny Huang) to Seattle for the summer to work on the Million Node GENI project.  I hope that Danny's experience with Justin will help facilitate future integration.

=== 2.2 Project Participants ===

The PIs are Jeannie Albrecht and Amin Vahdat.  Other participants to date include three undergraduates at Williams and a graduate student at UCSD.

=== 2.3  Publications ===

Jeannie Albrecht and Danny Huang.  Managing Distributed Applications Using Gush.  In Proceedings of the International Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities, Testbeds Practices Session (TridentCom), May 2010.

Jeannie Albrecht and Ryan Braud. Application Management and Visualization. In Demo Session Proceedings of the IEEE Conference on Peer-to-Peer Systems (P2P), September 2009. 

Elliot Jaffe and Jeannie Albrecht. PlanetLab - P2P Testing in the Wild. In Demo Session Proceedings of the IEEE Conference on Peer-to-Peer Systems (P2P), September 2009. 

Jeannie Albrecht. Bringing Big Systems to Small Schools: Distributed Systems for Undergraduates. In Proceedings of the Fortieth ACM Technical Symposium on Computer Science Education (SIGCSE), March 2009.

=== 2.4 Outreach Activities ===

No outreach activities were completed this quarter, although I am planning to collaborate with Mark Berman regarding a demo that highlights the educational benefits of GENI.

=== 2.5 Collaborations ===

Our main cluster collaborations to date are with John Hartman at the University of Arizona, the PlanetLab group at Princeton, the GpENI group, MAX, and NetKarma.    

In addition, we still remain in contact with groups at UMass and Duke University about Gush-related collaborations with ORCA, DOME, and ViSE.  

We have also been communicating with Rob Ricci regarding ProtoGENI support in Gush.

=== 2.6 Other Contributions ===

I attended GEC 2, GEC 3, GEC 4, GEC 5, GEC 6, and GEC 7 and presented at all conferences.  I am also current serving as the Experiment Services Working group co-chair.  Note that due to the pending birth of my son, I will not be attending GEC 8.