Changes between Initial Version and Version 1 of MillionNodeGEIN-qsr-Nov10

03/29/11 15:51:53 (13 years ago)
Vic Thomas



  • MillionNodeGEIN-qsr-Nov10

    v1 v1  
     1= Technical Status Report for Million Node GENI project for June 2010 through Nov 2010 =
     4PI: Thomas Anderson
     6Co-PI: Justin Cappos
     9== Major accomplishments ==
     11=== Milestones achieved ===
     15=== Deliverables made ===
     18e. Integrate with ProtoGENI so that calls that get resources and
     19release resources work across both frameworks.    This milestone is
     20contingent on agreement with ProtoGENI of what the interface should be
     21no later than 18 mo.
     25f. Specification of Clearinghouse API for end-hosts:   A description
     26of the interface by which the clearinghouse (SeattleGENI) interacts
     27with end-hosts.   This involves describing how the node manager API is
     28manipulated by SeattleGENI to track the state of resources.
     32g. Reference implementation v0.1 of end-host clearinghouse
     37h. Integrate Spiral 2 identify management mechanisms into SeattleGENI.
     38 }}}
     42i. Collaborate with O&M team on methods to share non-sensitive O&M
     43data with other GENI groups for Spiral 2.
     48a. Demonstration of Million Node GENI at GEC9. (Nov 5, 2010)
     50•Demonstration will include common use scenarios along with alpha
     51versions of a clicker program written in Seattle and an automatic
     52deployment system.
     56== Description of work performed during last quarter ==
     59We spent a large amount of time performing a proof-of-concept
     60integration with ProtoGENI's component manager.   As far as we can
     61tell, the service has not been used.
     64We have also been very involved with outreach.
     68== Project participants ==
     71Tom Anderson (PI)
     73Justin Cappos (Co-PI)
     75Arvind Krishnamurthy (Senior Personnel)
     77Monzur Mohammad (Open Source Developer)
     79Armon Dadgar (Undergraduate)
     81Jeff Rasley (Undergraduate)
     83Alan Loh (Undergraduate)
     85Sebastian Morgan (Undergraduate)
     87Danny Y. Huang (Undergraduate Intern visiting from GUSH group)
     89Alex Hanson (Undergraduate)
     91Shurui Sun (Undergraduate)
     94== Publications (individual and organizational) ==
     96We presented a paper at CCS (Computer and Communications Security)
     97related to the security architecture of Seattle's sandbox.   We also
     98presented a paper to the 5th International Workshop on Systems
     99Software Verification (SSV '10) on the techniques we are using to
     100validate the portability of Seattle programs.   One student has an
     101honor's thesis on Seattle ongoing.
     105== Outreach activities ==
     107Seattle is currently being used in four classes which represents a
     108significant support load.     We hosted a workshop / tutorial session
     109located with GEC 8 which we hope will increase both educational and
     110research adoption.
     112Justin Cappos co-taught a class on GENI for Ph.D. students with Kurt
     113Tutschku at the University of Vienna.  The class was attended by about
     11415 students from all around Europe.  GENI (in particular Seattle and
     115OpenFlow) were very well received.
     117== Collaborations ==
     119We have a deployment working on DOME, PlanetLab, GpENI, and done
     120automated deployments on Emulab.   We are using the Digital Object
     121Registry in production now.    Researchers at Nokia have ported
     122Seattle to some of their newer phones.   We are working with
     123researchers at the University of Victoria and HP Labs to support
     124Google Native Client with Seattle.
     126We worked with the Raven / Stork folks on a proof-of-concept port of
     127Owl to Seattle.   We hosted a student intern from the GUSH group who
     128is working on Seash (our GUSH-like shell) and network heterogeneity
     129masking.   As a result some of our networking code may be used in GUSH
     130and / or DOME.
     132We are using the TUF updater on Seattle's beta network.   We expect to
     133use it on the production network in the near future.
     135We are working with researchers at U Vienna to build an end-host
     136overlay on Seattle.
     138We are working with an expert in model-based testing at the University
     139of Washington to explore formal verification of Seattle's portability
     143== Other Contributions ==