[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #SRG-800720]: IDV2.9u2: http vs. RAMADDA data source problem

> Julien and/or Yuan,
> 
> I'm running IDV v2.9u2 on various Macs running Mac OS X v10.6.5.
> 
> I ran into a problem a year ago with an earlier iteration of the IDV (which I 
> didn't report) that is still present in v2.9u2. The other day I discovered an 
> interesting twist on the problem that might offer some insight to help solve 
> it.
> 
> I have students create a display, the data source for which is NAM model 
> output in a file residing on one of our local computers (virga.sfsu.edu). 
> Virga is a Web server, and students load the data source from virga using its 
> URL.
> 
> The display itself comprises (a) mean sea-level pressure (a color-filled 
> contour plot); (b) surface wind vectors; and (c) 500-1000 mb thickness (a 
> contour plot). Both the wind vectors and 500-1000 mb thickness are derived 
> quantities. The thickness is calculated from the 500 mb and 1000 mb 
> geopotential heights.
> 
> Students create the display successfully, then save it as a bundle.
> 
> When the IDV opens the bundle, loads the data source, and tries to create the 
> display, it creates the MSLP and wind vector fields successfully. However, it 
> can't load the geopotential height fields needed to derive the 500-1000 mb 
> thickness field and plot it, and for that plot it generates an error message.
> 
> However, virga is not only a Web server but also a RAMADDA server. When I 
> edit a typical student bundle and change the data source URL to a RAMADDA 
> reference to the same file on the same machine, the bundle works fine--no 
> problem with any of the three fields, including the 500-1000 mb thickness.
> 
> So the question is, what is it about accessing a data source via the Web 
> server that causes a problem for some fields (or at least the thickness field 
> derived from geopotential heights in a NAM forecast output file) but not 
> others, but that isn't a problem when the same file is accessed on the same 
> computer using RAMADDA?

Hi Dave,
         It is very different accessing a data file via the web server or the 
RAMADDA server, the RAMADDA is operating and/or sending the data  via OPeNDAP. 
But, in the web server, it doesn't have any better way to operate process the 
data, it will treat the data the same as a html file, in many situation when 
there is a compression involved, it will be more trouble. Anyway, it is not 
recommended to use web server for large grid data file.


Yuan
,  > 
> I've attached a copy of each bundle, which are identical except for the three 
> (identical) data source references in the bundle.  Let me know if you need 
> anything else, such as the error message produced when the IDV tries to 
> download the geopotential height fields needed to computer thickness.
> 
> No big rush on this. I hope you're enjoying the holidays!
> 
> -- Dave
> 
> ************************************************************
> *   Dr. Dave Dempsey                        |       ^   ___       \|/     *
> *   Dept. of Geosciences                    |  ) ^     /||_||\    --0--  *
> *   San Francisco State University   | )  )    /  ||_||  \    /|\     *
> *   1600 Holloway Ave.                      |  )  )  /   ||_||    \          *
> *   San Francisco, CA   94132          |  )  ) /    ||_||      \   ^   *
> *                                                             | )  )  )    
> ||_||        \      *
> *   Phone:  (415) 338-7716               |  )  )  )~||~||~~~~ \~~*
> *   FAX:      (415) 338-7705               | )  )  )  ) ~  ~  ~ ~ ~ ~*
> *   Email:   address@hidden        |  )  )   )  ) ) ~  ~   ~ ~ *
> ************************************************************
> 
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: SRG-800720
Department: Support IDV
Priority: Normal
Status: Closed