2006.11.10

Attendees

• Julien Barde
• John Graybeal
• Duane Edginbton
• Luis Bermudez
• Mike Godin
• Brian Schlining
• Kevin Gomes
• Mike McCann

Purpose of the meeting

Discuss in a technical group the data aggregation concept at MBARI to have a coherent strategy to propose to Francisco Chavez.

Minutes

• Luis B. presented these 2 documents to frame the discussion:
  • Data aggregation ideas: MDA.pdf
  • MBARI-Related Data and Infrastructure Projects (by John G.): DataProjects.doc
• Aggregation is a function of the scientific audience. It depends on the use case. Need to further investigate this topic with Francisco.
• Thredds has limitations in terms of aggregation. A dimension in all the data sets have to be compatible in order to aggregate by that dimension. Need to perform interpolation. Thredds uses the NectCDF aggregation library, so Thredds is really not neccesary.
• When two heterogeneous data sets are put together like CODAR and AUV data it makes good impressions to scientists
• Aggregate by platform makes sense.
• Andrew Chase did previous work using WMS - which we could leverage
• Mike Godin is working on MoQua to not be depended on LAS.
• As a first step taking aggregating data using OPenDAP and NetCDF makes sense. This will be a baby step in the design of a more general aggregation framework.

Actions

• Discuss next meeting with Franciso the concept of Aggregation - using NetCDF files and set a use case

MDA.pdf (application/pdf)
DataProjects.doc (application/msword)