[WAM-IPE] Fwd: Re: WAM-IPE ops - RESTART folders....: new idea to test restart compset

Naomi Maruyama naomi.maruyama at noaa.gov
Tue Apr 11 19:52:09 UTC 2017


Hi Houjun and Weiyu,


here is the failed run Bob tried with the 1hr restart compset that we 
discussed during our meeting:

/scratch3/NCEPDEV/swpc/scrub/Robert.Oehmke/rt_64184/swpc%20090115_dumprs_1hr_spacewx_gsm%wam%T62_ipe%80x170

We would really appreciate it if you could kindly give us any suggestions.

Thanks.

naomi




-------- Forwarded Message --------
Subject: 	Re: WAM-IPE ops - RESTART folders....: new idea to test 
restart compset
Date: 	Mon, 3 Apr 2017 13:12:37 -0600
From: 	Robert Oehmke <robert.oehmke at noaa.gov>
To: 	naomi.maruyama at noaa.gov
CC: 	George Millward - NOAA Affiliate <george.millward at noaa.gov>, Tim 
Fuller-Rowell - NOAA Affiliate <tim.fuller-rowell at noaa.gov>, Rashid 
Akmaev <Rashid.Akmaev at noaa.gov>



Hi Naomi,

    I originally started with a 1hr run, but when I tried to do a 1hr 
run with the WAM restart interval set to 1hr, the run stopped almost 
immediately. Valery speculated that maybe WAM won’t work with that short 
of a restart interval (he said that he’s never tried it and maybe it 
causes a problem with the digital filters), so I switched to the 1day 
run, where it at least ran for 18 hours.

If you’re curious here’s that 1hr run:
/scratch3/NCEPDEV/swpc/scrub/Robert.Oehmke/rt_64184/swpc%20090115_dumprs_1hr_spacewx_gsm%wam%T62_ipe%80x170

If you see anything that I should change, let me know. I’m happy to try 
that run again.

Thanks,

- Bob

On Apr 3, 2017, at 12:03 PM, Naomi Maruyama <naomi.maruyama at noaa.gov 
<mailto:naomi.maruyama at noaa.gov>> wrote:

> George and Bob,
>
> I understand that Bob's new compset cannot be tested until the problem 
> of failing WAM-IPE run at 18UT will be solved (which I am working on 
> right now). It seems both Bob and Weiyu are having the same problem.
>
> I checked the Weiyu's results as you saw in the other thread: it looks 
> like the IPE electron density looks ok until 18UT (when the runs 
> crashed) by comparing the two runs between before vs. after Weiyu's 
> new commit. Also I compared to my old run on jan 13: The Weiyu's 
> results appeared identical until 16UT.
>
> If running WAM-IPE is holding the test, how about testing the new 
> compset with using 1hr WAM-IPE run, instead of 1day run(i.e., restart 
> at 1UT instead of 24UT)? It would be also helpful for our future debug 
> task because it takes much shorter to run only for 1hr.
>
> Just a thought.
> Thanks.
> naomi
>
>
>
>
> On 3/30/17 4:02 PM, Tim Fuller-Rowell - NOAA Affiliate wrote:
>> Good.
>>
>> On Thu, Mar 30, 2017 at 11:58 AM, George Millward - NOAA Affiliate 
>> <george.millward at noaa.gov <mailto:george.millward at noaa.gov>> wrote:
>>
>>     All
>>
>>     One thing we did with SWMF was, for operations, we wrote all of
>>     the relevant restart data into
>>     a directory that was timestamped - like this:
>>
>>     SWMF_RESTART.20161013_122800
>>
>>     These were produced every minute - whether we do it that frequent
>>     - or less (5 mins) I think this is a good approach
>>
>>     ..so in the directory:
>>
>>     WAM-IPE_RESTART.20161013_121500
>>
>>     we then have other subdirectories for the IPE and WAM needed
>>     datasets.
>>
>>     What this means is this: IPE output files like plasma00, plasma01
>>     etc. just contain data from a single time....this is not how it
>>     currently works (plasma00 can contain multiple times and these
>>     are referenced by another file ut_rec). T
>>
>>     This timestamped RESTART folder system might only be applicable
>>     to operations, so we may need to be introducing another new
>>     variable here such as:
>>
>>     $RUN_MODE=OPERATIONAL.
>>
>>     Anyhow, any comments?  Having a set of files clearly timestamped
>>     is a much better system for ops than one where files can contain
>>     multiple times - it yields to the sort of order that is needed.
>>
>>     Cheers
>>
>>     George.
>>
>>
>>
>>     -- 
>>     -- 
>>     ----------------------------------------------------------------
>>     Dr. George Millward
>>     Cooperative Institute for Research in Environmental Sciences (CIRES)
>>     University of Colorado, Boulder
>>
>>     Visiting scientist at
>>     NOAA Space Weather Prediction Center
>>     325 Broadway, Boulder, CO 80305
>>     Tel: 303-497-6775
>>
>>     george.millward at noaa.gov <mailto:george.millward at noaa.gov>
>>     ----------------------------------------------------------------
>>
>>
>
>
> -- 
> --------------
> Naomi Maruyama
> CIRES, Univ. of Colorado Boulder
> NOAA Space Weather Prediction Center
> phone: +1 303.497.4857
> email:naomi.maruyama at noaa.gov

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www.lstsrv.ncep.noaa.gov/pipermail/wam-ipe/attachments/20170411/67e949b7/attachment.html 


More information about the WAM-IPE mailing list