[Go-essp-tech] Options for implementing DRS URLs in CMIP5

[stephen.pascoe at stfc.ac.uk]

Bryan,

I confined myself to the current ESG software stack -- the datanode and
gateway.  I could have said "Internal redirect filter in front of
download services" instead of talking explicitly about THREDDS.  I
believe the architectural decision is the same which ever services we
consider.

I've tried to set out the options in a neutral manner so that we can
evaluate the advantages and disadvantages.  My assumption has always
been that we just put the data in the DRS hierarchy on the filesystem
(Option 1) but Bob believes datanode admins won't want to do this.

Stephen.

---
Stephen Pascoe  +44 (0)1235 445980
British Atmospheric Data Centre
Rutherford Appleton Laboratory

-----Original Message-----
From: Bryan Lawrence [mailto:bryan.lawrence at stfc.ac.uk] 
Sent: 14 December 2009 13:29
To: go-essp-tech at ucar.edu
Cc: Pascoe, Stephen (STFC,RAL,SSTD)
Subject: Re: [Go-essp-tech] Options for implementing DRS URLs in CMIP5

Hi Stephen

I am still given to understand that CMOR will write out the appropriate
directory heirarchy ... so why would a node manager want to arrange it
differently? (One of the reasons I have been against symlinks is that
does impact on how one would organise the data in ways that do impact on
physical layout).

In your document there appear to be some assumptions:
 - that a rewrite filter on the thredds interface would achieve the
result (is the thredds interface the only one that matters)?
 - that (in some options) we'd all not follow the DRS, but follow some
other common pattern? 

I don't believe either to be true (once you throw out the DRS), but I'm
not sure they're assumptions either :-)

Cheers
Bryan
 
On Monday 14 December 2009 11:21:49 stephen.pascoe at stfc.ac.uk wrote:
> Hi all,
>  
> So far we haven't agreed on how we are going to support DRS URLs 
> across the CMIP5 ESG federation.  Several options were discussed 
> briefly at NCAR last week but we didn't talk details.  I've attached a

> document with 4 options for how we could do it.  I'd appreciate input 
> on extra options and advantages/disadvantages of what's already there.
>  
> I think we need to make a decision about this very soon or DRS URLs 
> won't become a reality.  Can we discuss this at the telco on Tuesday?
>  
> Cheers,
> Stephen.
>  
> ---
> Stephen Pascoe  +44 (0)1235 445980
> British Atmospheric Data Centre
> Rutherford Appleton Laboratory
>  
> 



--
Bryan Lawrence
Director of Environmental Archival and Associated Research (NCAS/British
Atmospheric Data Centre and NCEO/NERC NEODC) STFC, Rutherford Appleton
Laboratory Phone +44 1235 445012; Fax ... 5848;
Web: home.badc.rl.ac.uk/lawrence
--
Scanned by iCritical.
-- 
Scanned by iCritical.