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

[TIGGE #XLU-757491]: tigge ldm server specs

Hi Baudouin,

I apologize for not answering this on Friday.  I was on my way out of
the door when I saw your message arrive.

> what would be the impact (locks, i/o...) of sending fewer larger
> products (i.e. 1000 files of 100MB)? One obvious drawback we need to
> resend larger amount of data. But apart from that?

There would be three impacts:

- a decrease in the number of write locks per unit of time (good)

- the need to have larger portions of the LDM queue "expired" (scoured)
  to make room for new products (this could be bad)

- as you say, the need to resend a larger block of data when it is missed
  (this would be bad)

Since the initial data transfer tests between ECMWF and NCAR/Unidata were
very successful when using larger products (10, 20, 30, 60 MB), I would have
no doubt that the effect on the ECMWF-NCAR transfers would work fine.
It would probably be a very good idea to perform transfer tests to/from CMA
to see what the potential impacts would be there.

Cheers,

Tom
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: XLU-757491
Department: Support IDD TIGGE
Priority: Normal
Status: Closed