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

Re: CONDUIT latencies--My two cents

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: CONDUIT latencies--My two cents
  • Date: Fri, 16 Jul 1999 10:12:00 -0500 (CDT)
We had a similar problem with the NMC2 feed when we tried it
this spring.  We were feeding from U. of Arizona and even though
the connection to them was not great it was acceptable. It turns
out that by the time the data filtered to them it was getting
old and the last leg to us sent it over the 3600 sec mark.
My feeling is that if you were really far from the main feed
like we were there is no way to receive all of the data
especially the large sets like the AVN/MRF NMC2 feed.

On another note..just my two cents.  I would highly recommend
considering Sun Solaris 7 x86 for those having problems with Linux.  It is
free except for shipping for education/non-commercial use.
We have been running the full NOAAPORT feed with all the GEMPAK
and McIDAS decoders (except NIDS and NLDN) for over a year plus
serves as the McIDAS and GEMPAK workstation.
It also serves as a web server and generates images(CGI scrips) for
our use from remote sites.  The speed at which it generates
the charts is great.  I have never had any trouble with,
and this is with IDE drives(400Mhz PII, 256 MB RAM).  I have no doubt that
in terms of raw performance Linux is faster, but I have had no complaints
at all.  Never a crash, or anything strange going on with the
LDM(except things I caused, of course). 

Robert Mullenax