Grib data

[Andres Sepulveda]

 Quite a good idea, indeed. Could there be a test of it? Could the
performance be measured?

On Tue, August 29, 2006 2:42 pm, Frank Bandle said:
> Hi Wesley
>
> BIT-Torrent could be a good idea to speed up  performance for gfs
> forecast data
>
> regards Frank
>
>
> Dipl. Met. Frank Bandle
> frank at bandle.com * FeMax Wetteragentur IT & Medienberatung
> www.wetter365.net * www.femax.de
>
>
>
> Am 21.08.2006 um 19:45 schrieb Wesley Ebisuzaki:
>
>> Scot Lilly wrote:
>>>
>>> My meteorologist and I are looking for two things.
>>>
>>>
>>>
>>> 1.)     We currently use a Linux virtual server to download the GFS
>>> operational data off of the NOAA http server.  This data seems to be
>>> somewhat quick in manner, however we find that sometimes it can get
>>> hung up and even not work to some extent slowing down our time to
>>> analyze the data.  Are there any known locations and/or methodologies
>>> to get this data quicker so that we can get the most current model
>>> runs.
>>>
>>>
>>>
>>
>> Using the http server to download specific grib records is about as
>> efficient as you can get.
>> For others, the reference is,
>>
>>   http://www.cpc.ncep.noaa.gov/products/wesley/
>> fast_downloading_grib.html
>>
>> The problem is that many people want to download the GFS forecasts as
>> soon as they are available and
>> do it within a few minutes.  In a quest to get the data quickly as
>> possible, the users will overload the servers.
>>
>> Any comments about using bit-torrent for gfs forecast data?
>>
>>       Wesley Ebisuzaki
>>
>>
>>> 2.)     We are trying to find some kind of consistent format for
>>> getting hurricane model tracks so that we can feed it to Grads.  Our
>>> problem is that we have tried to use the text file that comes from
>>> the
>>> NOAA and feed it into a spreadsheet to create a csv-ascii output
>>> file.  However, the problem is that the file format often changes
>>> requiring manual modifications to create the proper output.  We are
>>> looking for any type of consistent sheet (that does not change
>>> formats) and/or a grib file.  Any suggestions would be extremely
>>> helpful.
>>>
>>>
>>>
>>> We are open to any new ides for getting data in the most efficient
>>> and
>>> timely manner, less spending thousands of dollars.
>>>
>>>
>>>
>>> Thanks for your time and look forward to any and all comments.
>>>
>>>
>>>
>>>
>>>
>>> Scot Lilly
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>