[netcdf-java] Aggregation on an inner time dimension

• To: netcdf-java@xxxxxxxxxxxxxxxx
• Subject: [netcdf-java] Aggregation on an inner time dimension
• From: Martin Price <mpricemetoffice@xxxxxxxxxxxxxx>
• Date: Thu, 13 Aug 2009 14:56:07 +0100

Does anyone know if it's possible to do an Aggregation on an inner dimension
(preferably ForecastModelRunCollection but a plain-vanilla aggregation would
be fine).  We are using NCML to aggregate a lot of model run files (every 6
hours for ~10 years), and most often using the aggregated dataset to extract
full length time series.  This works fine with time as the outer dimension,
but the disk is spending nearly all its time seeking, and we know from tests
using hard-coded file access that this runs a lot faster if we permute the
files so that time is the inner dimension.

I've tried using 'union' or 'joinexisting' ncml to do it - but as you'd
expect from the documentation it doesn't work.  Is there a way to do this?

cheers,
Martin

--
Martin Price
Ocean Forecasting Research and Development
Met Office, FitzRoy Road, Exeter, EX1 3PB, United Kingdom
Tel: +44 (0) 1392 886982   Fax: +44 (0) 1392 885681
email: mpricemetoffice at googlemail dot com
http://www.metoffice.gov.uk

• Follow-Ups:
  • Re: [netcdf-java] Aggregation on an inner time dimension
    • From: John Caron