[gradsusr] dods server for CMIP5 data?

[Jennifer Adams]

There is no free lunch when it comes to CMIP5 data. There is a thredds data server running at each data node, but the authentication issues make access a bit awkward. There is also (in some cases) extreme granularity in the files -- some variables have as many as 2000 files to cover the complete time series -- and each file is served separately. Also, there is no standard naming convention for the directory paths on each of those servers, so it is difficult to automate access from one data node to another. For CMIP data, most of the valuable metadata exists as attributes in the data files, and that would likely be lost when accessing and collecting data via OPeNDAP. And OPeNDAP would be very slow, slower than a wget job, that's for sure. The new PeerToPeer system for getting CMIP5 data is working well enough, and it is possible to automate search and download jobs. I speak from personal experience -- I have spent much time over the past year collecting and managing about 30Tb of CMIP5 data. 
--Jennifer

On Nov 15, 2012, at 9:39 PM, Haibin Li wrote:

> Hello there,
> 
> Does anyone know whether there is a dods server for recent CMIP5 data 
> that can be accessed using GrADS?
> Thanks.
> 
> Haibin
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr

--
Jennifer M. Adams
IGES/COLA
4041 Powder Mill Road, Suite 302
Calverton, MD 20705
jma at cola.iges.org



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20121116/afdf773d/attachment-0003.html