Ecoinformatics site parent site of Partnership for Biodiversity Informatics site parent site of REAP - Home




Meeting Notes Feb 07

This is version 1. It is not the current version, and thus it cannot be edited.
[Back to current version]   [Restore this version]

Baru -- NEON presentation

PDR (Prelim design Review) -- May 2007 FDR (Final Design Review) -- May 2008 MRE construction, 5 years MRE operations + maintenance from there

Currently focused on Single String testbed -- risk management/assessment

Compact RIO controller (NI) POP in SDSC (moving to James Reserve this week) -- runs ORB -- RBNB (Ring Buffered Network Bus) (Data Turbine) -- CREARE agreed to go open-source with DataTurbine

RDFA --> RDF Annotations

Refinement of deliverables

D1: web service/ecogrid/opendap integration

-- need technical use cases to drive development -- e.g., import opendap data into a kepler workflow

a. reconcile data models of kepler/opendap/eml/sensorml/other sensor models?/OGC data sources/(V)ORB/RBNB b. develop opendap datasource adapter for kepler c. transport: shared transport library usable by several datasource adapters d. expose opendap/ODC metadata search in Kepler? e. determine how to express queries (server-command language?) -- subsetting on remote servers -- aggregation on remote servers -- server side processing (how does this relate to workflow processing?) f. determine how to handle authentication/authorization to opendap services g. ability for opendap and others servers to self-register with a service registry

Outreach deliverables

O1: web interfaces, ilkay

Monthly Updates -- target scientists to keep people updated

Kepler Deliverables

K1: Scientist's view a) running models -- features we need -- for SST use case need ability to compare fields statistically -- advanced search capability for spatio-temporal data discovery -- adaptive models, change during course of execution -- data assimilation? e.g., the COMET examples -- hybrid models? -- yes, for 2 different physical models -- different temporal behaviors? multiscale? -- different types of modeling (e.g., IBMs, CT, stochastic/deterministic) -- checkpointing for long running models -- asynchronous execution -- detachable from GUI -- everybody in project should be able to check status of execution, e.g., from the web -- new components needed -- spatial statistics components (statistical comparison of fields) -- spatial and temporal buffering (e.g., convex hull buffer) (matchups) -- new components for GAs, simulated annealing, NN, etc. -- fix MATLAB expression actor to support data output

K2: Engineer's view (possibly lead by SDSC)

Workshop 1: Requirements and use cases

1. many people or many workshops? 2. eco + oceans workshops separate? 3. Workshop in late June or July in Santa Barbara 4. Conference call to be scheduled to plan with eco people 5. Peter C. to possibly present mockup at GHSST meeting in Melbourne in May to start getting people involved 6. Mark S. to plan ws and get rooms, hotels, etc. 7. Action item: develop wf mockups to show current and candidate features and components

Timeline + Milestones

  • Year 1:
    • Deliverable D1
      • Data model reconciliation (Nathan)
      • Web services interface(s)
      • OPeNDAP actors/libraries in Kepler

      • Streaming data from DataTurbine (Ilkay)
        • sensor calibration workflows

      • Deploy eco test case sensors (Matt)

      • Workshop wf mockups

      • Advanced search UI in Kepler
      • Rapid data exploration/viz in Kepler

Go to top   More info...   Attach file...
This particular version was published on 28-Feb-2007 11:14:21 PST by uid=jones,o=NCEAS.