| 1 | Cluster D conference call on Friday, June 18, 2010 |
| 2 | |
| 3 | |
| 4 | Attendees: [[BR]] |
| 5 | |
| 6 | Ilia Baldine (RENCI)[[BR]] |
| 7 | Anirban Mandal [[BR]] |
| 8 | Chris Heermann[[BR]] |
| 9 | |
| 10 | Harry Mussman (GPO) [[BR]] |
| 11 | |
| 12 | |
| 13 | |
| 14 | Mike Zink (UMass) [[BR]] |
| 15 | Emmanuel Cecchet [[BR]] |
| 16 | Brian Lynn[[BR]] |
| 17 | David Irwin[[BR]] |
| 18 | |
| 19 | Denise Gurkan (Univ Houston) |
| 20 | |
| 21 | Michael Wang (Columbia) [[BR]] |
| 22 | |
| 23 | Onjin (Oklahoma) [[BR]] |
| 24 | Wenjie[[BR]] |
| 25 | |
| 26 | Wenjie (Ohio State) [[BR]] |
| 27 | |
| 28 | Hongwei (Wayne State)[[BR]] |
| 29 | |
| 30 | Rudra Dutta (NCState) [[BR]] |
| 31 | |
| 32 | Agenda: [[BR]] |
| 33 | |
| 34 | 1) Taking stock of known unresolved problems:[[BR]] |
| 35 | + Brian Lynn: Still has issues with pre-mature close[[BR]] |
| 36 | + Ilia: Study more.[[BR]] |
| 37 | + David: Eucalyptus cluster at UMass; need to connect via new clearinghouse, before GEC8 |
| 38 | + Ilia: Have clearinghouse up by next week. |
| 39 | + David: Needs to check new broker, explain issues with policy and make request. |
| 40 | |
| 41 | 2) New networking driver structure[[BR]] |
| 42 | + Denise: Shade working on driver for 3400 |
| 43 | + Ilia: New document on drivers, quite detailed. |
| 44 | |
| 45 | 3) GEC8 demos [[BR]] |
| 46 | + DICLOUD: yes; ticket |
| 47 | + ViSE: yes; no ticket |
| 48 | + DOME: no |
| 49 | + IMF: yes |
| 50 | + ORCA/BEN: |
| 51 | - OpenFlow demo RENCI and GIST Korea |
| 52 | - Programmable router, on multiple site hosts, practice for GEC9 |
| 53 | + Kansei: yes; un cross-site demo, sensor nodes and latency through |
| 54 | + OKGEM: yes, possibly controllable robot |
| 55 | + LEARN: Integration of 3400 into LEARN, at Univ Houston and Rice |
| 56 | |
| 57 | 4) GENI Solictation 3: [[BR]] |
| 58 | + Ilia: Considering GENI racks |
| 59 | - Many Cluster D sites |
| 60 | - Each site with separate network connectivity |
| 61 | - Spaces in lab or university data center? |
| 62 | - Connectivity via many (10+) VLANs? |
| 63 | - Need good connectivity into sites |
| 64 | + David: Space, power, cooling requirements for each rack? |
| 65 | - Ilia: 19" full height rack |
| 66 | - GENI enabled open-flow switches |
| 67 | + Ilia: Purchased by project, sent out to each site |
| 68 | - Agreement with manufacturer, who sends it out. |
| 69 | - Harry: See current WiMAX mesoscale project |
| 70 | + Ilia: Two CFs: ORCA and protoGENI Can be accessed by either |
| 71 | - Local interface, but also managed by central site (e.g., RENCI) |
| 72 | + Ilia: How many sites? 40 racks too many for level of funding |
| 73 | - Perhaps 15 sites, each with one or two racks |
| 74 | - Need sites on the west coast |
| 75 | + Ilia: Connectivity to backbone costs |
| 76 | - Campus? |
| 77 | - Regional networks? |
| 78 | - Perhaps: FrameNet wherever possible, otherwise Starlight |
| 79 | |
| 80 | |
| 81 | Last call A) Connectivity to Wayne State and Ohio State[[BR]] |
| 82 | + Joe: Ready to go from Starlight to Merit racks; [[BR]] |
| 83 | + Hongwei: Merit to OARnet to Ohio State; also Merit to Wayne State[[BR]] |
| 84 | + Ilia: Working with MAX on interoperability[[BR]] |
| 85 | + Ilia: working toward translation from protoGENI to NDL [[BR]] |
| 86 | - Harry: Srikanth also working on topic[[BR]] |
| 87 | |
| 88 | Last call B) GEC9 demo proposals, accepted: [[BR]] |
| 89 | a) Joe (Northwestern): iGENI, HP and Ericsson: video transcoding on clouds over dynamically provisioned networks. Prototype at GEC8. HP wants to include OpenFlow. [[BR]] |
| 90 | b) David (UMass): Now-casting (short-term weather forcasting), GEC7 demo plus more computation [[BR]] |
| 91 | c) Jeff Chase (Duke): Dynamically stand-up ASs with custom click routers [[BR]] |