| 28 | === Session Summary === |
| 29 | |
| 30 | Links to appropriate section of the [http://groups.geni.net/geni/wiki/GEC12MesoScaleMonitoring/DetailedNotes Detailed Notes] are included below. |
| 31 | |
| 32 | '''Sarah Edwards''' of the GPO introduced the session. |
| 33 | |
| 34 | There were two goals for the session: |
| 35 | 1. Keep talking about day-to-day issues and problems, as we did at GEC11 |
| 36 | 2. Look ahead to where GENI is headed. We need to be able to provide inputs |
| 37 | so that other parts of GENI will address monitoring needs, and so |
| 38 | we don't ignore any important pieces. |
| 39 | |
| 40 | '''Josh Smift''' of the GPO [http://groups.geni.net/syseng/wiki/SarahEdwards/GEC12MonitoringNotes#FOAM spoke] about FOAM, the new !OpenFlow Aggregate Manager (AM). |
| 41 | * Version 0.6 will migrate to campuses on November 21. |
| 42 | * FOAM addresses three pain points identified at the last GEC: |
| 43 | * improved data mapping between FV slices, GENI slivers, GENI slices, and experimenters allows much improved monitoring; |
| 44 | * an automated approval engine make approval (aka !OpenFlow Opt-In) easier; and, |
| 45 | * FOAM reduces the size of the flowspace (eg the GPO saw an improved by a factor of 30). |
| 46 | |
| 47 | '''Camilo Viecco''' (of Indiana U/GMOC) did a [http://groups.geni.net/syseng/wiki/SarahEdwards/GEC12MonitoringNotes#SNAPPandthePortal demonstration] of how to use SNAPP to find and visualize monitoring data from the GMOC DB. |
| 48 | * Camilo demonstrated the new [http://gmoc-db.grnoc.iu.edu/measurement/portal.cgi Portal] interface to SNAPP which allows you to browse the data by meso-scale participant. |
| 49 | * In addition, Camilo demonstrated how to use search in [http://gmoc-db.grnoc.iu.edu/measurement SNAPP] (eg to find all data related to a single slice). |
| 50 | |
| 51 | '''Chaos Golubitsky''' of GPO [http://groups.geni.net/syseng/wiki/SarahEdwards/GEC12MonitoringNotes#Topology spoke] about Topology in the Mesoscale. |
| 52 | |
| 53 | Chaos discussed two solved problems which can cause an experimenter's topology to not match their expectations. Monitoring in both the near and long terms needs to address problems like these. |
| 54 | |
| 55 | '''Sarah Edwards''' of the GPO provided an [http://groups.geni.net/syseng/wiki/SarahEdwards/GEC12MonitoringNotes#Monitoringandmanagementrequirements overview] of top-down GENI monitoring requirements. |
| 56 | Sarah presented a top-down analysis of monitoring and management requirements, to ensure we cover all the key areas and focus on areas that will become problems, not just areas that are currently a problem. Monitoring and management in GENI must support |
| 57 | * cross-GENI monitoring, |
| 58 | * troubleshooting and event escalation, |
| 59 | * quickly identifying and addressing security events, and |
| 60 | * emergency stop, |
| 61 | * all while ensuring all parties follow defined policies, |
| 62 | * and preserve both experimenter and end user privacy. |
| 63 | |
| 64 | |