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

20030515: LDM use at CICESE (cont.)

>From: Luis M Farfan <address@hidden>
>Organization: CICESE - BCS
>Keywords: 200305121843.h4CIhU7U016942 LDM IDD rtstats

Hi Luis,

>greetings from CICESE in Baja California, Mexico.

It is good to hear from you!

>We have experienced several problems in our computer
>department and this is why we are not ingesting data from
>our source: Scripps.

OK.  No worries.

>However, we are still interested in getting the real-time data
>and I will work with our support office to get the operations
>back to normal.
>
>Let me, briefly, mention 2 aspects for your comments:
>
>1) we have the LDM version released by September 2002
>   and have not updated the software since then. As you,
>   suggest we can go for the upgrade soon. We seem to have
>   ldm-5.2 at the moment.

It would be best to do the LDM upgrade as part of getting data
ingestion working again.  The machine at Scripps, aeolus.ucsd.edu, has
been running LDM-6 for some time now, and it has been performing _much_
better since this upgrade.

>2) The machine that receives the IDD data is a SunOS 5.8
>   with 8 processors and is a production unit. This is, is used
>   to perform jobs that requires of large resources (time,
>   memory, etc). In the last few weeks, the number of users
>   and jobs has increased to a higher level of activity. Then, I
>   will monitor how this affects the LDM/IDD performance.
>   Please let me know if you have suggestions on this issue.

The LDM system use for a site that is not relaying data is very low.  I
do not expect the processing on your machine to be affected by the
installation and use of LDM-6.  Also, I don't expect that the system
use will affect the performance of the LDM.

I seem to remember that the feeds that you are requesting are pretty
modest (global observation data in IDS|DDPLUS; NOAAPORT model data in
HDS; satellite imagery from UNIWISC (also known as MCIDAS)).  I expect
that you will find that receipt of these data will not be noticable.
If you decide that you want to get higher volume datasets like the
NOAAPORT imagery (NIMAGE) and NCEP high resolution model data
(CONDUIT), then paying attention to system use may become more
important.

>Thanks, Luis M Farfan.

Thanks for your reply!

Tom