Re: [las_users] Save As function on LAS

[Anton Yoel Buenavista <tmsbayd@xxxxxxxxxx>]

Hi Roland,

I'm a little confused about the use case here.  If you are downloading the full resolution for multiple time steps via LAS for a data collection for which you have local access there might be some more efficient ways to achieve your goal.

Our goal is basically to provide access to users who do not have or who are not allowed to have local access to our server. LAS is the only  means for the user to interact with our data.

Thank you for the hints. Looks like TDS is indeed requesting a large amount of memory, so we'll look into increasing memory instead for our server. This should solve the problem.

Thank you for your help.

--
Regards,
Anton Buenavista

-------- Original Message --------
Subject: Re: [las_users] Save As function on LAS
From: Roland Schweitzer - NOAA Affiliate <roland.schweitzer@xxxxxxxx>
To: las_users@xxxxxxxx <las_users@xxxxxxxx>
Date: 11/12/2012 00:25

Hi,

Yes, it it very likely a memory problem on the TDS server which is preventing it from handling the request.  Certainly looking though the $TOMCAT_HOME/content/thredds/logs will give you some clues as to what's going wrong.

I'm a little confused about the use case here.  If you are downloading the full resolution for multiple time steps via LAS for a data collection for which you have local access there might be some more efficient ways to achieve your goal.  

Roland

On Mon, Dec 10, 2012 at 1:53 AM, Anton Yoel Buenavista <tmsbayd@xxxxxxxxxx> wrote:

Dear LAS mailing list,

I am able to use the Save As function for LAS but only for a limited number of time steps for my dataset (1320x1487 grid size). I can only save as NetCDF of up to only 20 time steps for the full domain. Any more than that and I get the following error:

 **netCDF error
             NetCDF: Malformed or inaccessible DAP DATADDS (OPeNDAP/netCDF Error code -73)
             Data set: http://localhost/thredds/dodsC/las/data.nc

Our data is hosted using thredds, and contains about 1008 time steps via aggregation. My hunch is that this is a memory problem, as the machine where our thredds is hosted only contains 4GB of RAM, but I'm hoping the problem could be something else.

Could anyone advise if this is indeed a memory problem, or if there is any workaround for downloading large sets of data over thredds/las. Your suggestions are highly appreciated.
--
Regards,
Anton Buenavista