Re: [thredds] How do ncss/wcs/wms deal with multiple time axis from large FCs?

  • To: THREDDS community <thredds@xxxxxxxxxxxxxxxx>
  • Subject: Re: [thredds] How do ncss/wcs/wms deal with multiple time axis from large FCs?
  • From: John Caron <caron@xxxxxxxx>
  • Date: Wed, 24 Jun 2015 17:22:54 -0600
Hi Dan:

Yes, this explosion of time axes with inhomogeneous collections is causing
new problems. The opendap form should have all of the axes, if you look
close. The ncss may be getting confused, however, esp if different grids
use different axes.

Im working on a new version of Grids (in 5.0) that should help solve it,
and i will test it on some datasets with many time axes. In the meanwhile,
keep your eyes out for a reproducible example, we may be able to mitigate
it in 4.6.

John

On Wed, Jun 24, 2015 at 2:50 PM, Christian Ward-Garrison <cwardgar@xxxxxxxx>
wrote:

> Hi John,
>
> Dan has a question about coordinate systems in GRIB FeatureCollections. I
> think it would take me a long time to figure out the answer, so I'm going
> to throw it over to you.
>
> -Christian
>
> ---------- Forwarded message ----------
> From: Dan Swank - NOAA Affiliate <dan.swank@xxxxxxxx>
> Date: Tue, Jun 23, 2015 at 7:44 AM
> Subject: Re: [thredds] How do ncss/wcs/wms deal with multiple time axis
> from large FCs?
> To: Christian Ward-Garrison <cwardgar@xxxxxxxx>
>
>
> With the exception of the unusual cases where
> the error message is thrown outright, NCSS manages to work,
> I just do not know how it is constructing its single time
> axis from the multiple axis present in many of our FCs.
>
> Would it help if I passed along our entire stack of gbx9/ncx3 ?
>
> On Mon, Jun 22, 2015 at 5:19 PM, Christian Ward-Garrison <
> cwardgar@xxxxxxxx> wrote:
>
>> Hi Dan,
>>
>> I wouldn't be surprised at all if this occurs. A related issue is that
>> NCSS tries to present ALL values on the time axis. Of course, this doesn't
>> scale in a useful way; imagine a dataset with 50,000 times.
>>
>> We may be able to extend these services to allow the user to choose the
>> right time axis. However, it's critical that we be able to reliably
>> reproduce the conditions first.
>>
>> Cheers,
>> Christian
>>
>> On Mon, Jun 22, 2015 at 3:08 PM, Dan Swank - NOAA Affiliate <
>> dan.swank@xxxxxxxx> wrote:
>>
>>>
>>> This is in regards to the latest THREDDS 4.6.2
>>>
>>> Some of our GRIB datasets from NCEI-NOMADS, when
>>> placed into huge TwoD featureCollections, present me with
>>> many time axis:  I've seen on the order of refTime15, time15, etc.
>>> Resulting (I suspect) from non-homogeneities from dataset changes over
>>> time
>>> (years) that cannot easily be dealt with.
>>>
>>> ncss and wcs/wms services supposedly work with GRIB featureCollections,
>>> I've seen various results.   In daily directory pratitions, which are
>>> almost always
>>> homogenous, these services generally work well.   If you try them on
>>> the top level feature collections (spanning years)  you get a flood of
>>> different time axis, mentioned above.
>>>
>>> My main question is... if ncss / wcs / wms is
>>> used on these multi-year aggregations, how do/can
>>>  these services select and use only one from the
>>> various time axis presented on the OpenDap form?
>>>
>>> In some cases, our ncss on Full Collection top level, will
>>> outright fail, printing nothing more then a text string, like this:
>>>
>>>
>>> gribFC/srrsGrib1fc/TwoD/LatLon_181X360-p5N-180p0E/dataset.html
>>>
>>> and terminate,  leaving no ncss form.
>>>
>>>
>>> John Caron has claimed ot not be able to replicate this.
>>> And I have not been very successful in isolating what conditions
>>>    cause it to happen.   But it certainly does.
>>>
>>>
>>> Dan Swank
>>> NOMADS System Support, STG Inc. - Federal Government Contractor
>>> NOAA's National Centers for Environmental Information (NCEI)
>>> Data Stewardship Devision (DSD)
>>> 151 Patton Ave
>>> Asheville, NC 28801
>>> dan.swank@xxxxxxxx
>>> Phone: 828-271-4007
>>> Customer Support:  828-271-4800 or http://www.ncdc.noaa.gov/contact
>>>
>>> The newly formed NCEI merges the National Climatic Data Center (NCDC),
>>> the National Geophysical Data Center (NGDC), and the National Oceanographic
>>> Data Center (NODC).
>>>
>>> Facebook: http://www.facebook.com/NOAANationalClimaticDataCenter
>>> Twitter: @NOAANCDC @NOAAOceanData
>>>
>>> ======================= Disclaimer ===========
>>> I am not a Federal Employee.
>>> Any opinions expressed in this message are mine personally and
>>> do not represent official positions from STG Inc, ERT, NCEI, or NOAA
>>> ============================================
>>>
>>> _______________________________________________
>>> thredds mailing list
>>> thredds@xxxxxxxxxxxxxxxx
>>> For list information or to unsubscribe,  visit:
>>> http://www.unidata.ucar.edu/mailing_lists/
>>>
>>
>>
>
>
> --
>
> Dan Swank
> NOMADS System Support, STG Inc. - Federal Government Contractor
> NOAA's National Centers for Environmental Information (NCEI)
> Data Stewardship Devision (DSD)
> 151 Patton Ave
> Asheville, NC 28801
> dan.swank@xxxxxxxx
> Phone: 828-271-4007
> Customer Support:  828-271-4800 or http://www.ncdc.noaa.gov/contact
>
> The newly formed NCEI merges the National Climatic Data Center (NCDC), the
> National Geophysical Data Center (NGDC), and the National Oceanographic
> Data Center (NODC).
>
> Facebook: http://www.facebook.com/NOAANationalClimaticDataCenter Twitter:
> @NOAANCDC @NOAAOceanData
>
> ======================= Disclaimer ===========
> I am not a Federal Employee.
> Any opinions expressed in this message are mine personally and
> do not represent official positions from STG Inc, ERT, NCEI, or NOAA
> ============================================
>
>
  • 2015 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the thredds archives: