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

[THREDDS #WVA-198552]: NAS Hardware for Thredds Server



Christof,

Thanks for the kind words--it's great to know that our community is making 
great use of our work.

Regarding the NAS, on the surface I think this should work, but is dependant 
upon:

1) The network connection to/from the NAS
2) The types of files--GRIB (due to index files) and NetCDF (good headers) 
should work well. Other files (which require downloading/reading the entire 
file) may tax your bandwidth more.

Would you be willing to post this question to the mailing list at 
address@hidden ? This may help you tap into the experience that the rest of the 
thredds user community has with network storage for data.

Thanks,

Ryan

> Dear Sir or Madam,
> 
> first of all, I would like to thank you for all your efforts w.r.t. to the 
> TREDDS Server Software. It is really amazing how easy and convenient you can 
> set up such a server and
> I am really impressed by all the different access methods and services, which 
> work directly out of the box.
> 
> Currently, I’m running the server on my own working machine. But within the 
> next few weeks, our working group would like to set up a small NAS, where we 
> can move our
> often used data (mainly global and regional hydrometeorological data such as 
> climate simulations, forecasts, …). As we would like to still use all the 
> different services (especially
> OPeNDAP and NCSS, which is extremely helpful for accessing large data), it 
> would be really great if you could give me some advice about the 
> configuration of such a server.
> 
> In fact, I’m wondering if the performance of a NAS (e.g. Synology 415+) is 
> sufficient for running a THREDDS-Server or if it would make more sense to 
> just save the data
> on the NAS and run the software on a powerful workstation. Some „benchmark“ 
> application would be to access regional climate simulations (which are stored 
> in e.g.
> monthly files, where each file contains hourly data) and use the NCML to 
> aggregate all files and access only a subset of the full mode runs.
> 
> Currently, we’re assuming that about 10 persons will have access to the 
> server, i.e. our solution is not indented as a public data server (which, I 
> think, already somehow
> reduces the required performance of the system).
> 
> Do you have any experience and suggestions w.r.t. such server solutions?
> 
> Best regards and many thanks in advance,
> Christof Lorenz
> 
> 
> 
> 
> ---------------------------------------------------------------------------
> Dipl.-Ing. Christof Lorenz
> 
> Karlsruhe Institute of Technology (KIT)
> Institute for Meteorology and Climate Research (IMK-IFU)
> 
> Kreuzeckbahnstr. 19
> 82467 Garmisch-Partenkirchen (Germany)
> 
> Phone:   +49 8821 183 214
> Fax:         +49 8821 183 243
> 
> E-mail: address@hidden
> Web: http://www.imk-ifu.kit.edu
> 

Ticket Details
===================
Ticket ID: WVA-198552
Department: Support THREDDS
Priority: Normal
Status: Closed