[Hybrid baryons] [clas12_rgk] Fwd: RGK Decoded files

[Nick Markov]

Dear Raffaella,
As I say, we will need 10 (up to 15 in the very worst case) TB for the RGK.
I am looking exactly at the decoded files right now and from what I see
they all can go being rather old. I have not personally used any of them in
the last months.
So, again, we need nothing for the decoded files while 10-15 TB for 6.5 GeV
and 7.5 GeV combined.
Again, if we can use /work/clas12 it will be ideal.
Nick.


On Fri, Aug 30, 2019 at 9:53 AM Raffaella De Vita <
Raffaella.Devita at ge.infn.it> wrote:

> Dear Nick,
> Thanks for the information.
> I think the best would be if you could come up with an estimate of the
> disk space needed first of all and of how much you could clean up from
> /work/clas12/rg-a/ (I see 45 TB are decoded files that I’m not sure are in
> use), and then we can decide what would be the best location for the RG-K
> cooking.
> /Work has indeed 10 TB free but we need to leave some space available for
> RG-B, simulations in support of the analysis and users who are working on
> analysis right now.
> Thanks,
> Raffaella
>
> On 30 Aug 2019, at 15:32, Nick Markov <markovnick at gmail.com> wrote:
>
> Dear all,
> the rga pass 1 cook (66 runs) occupies about 10TB and is currently stored
> on volatile (and copied to cache).
> Depending on how much runs we will need for the RGK, I'd say we will have
> roughly the same amount of data. Currently rg-a occupies 83TB on work,
> while rg-k is about 9TB. I am not sure about the quote right now, but
> hallB/clas12 is 10TB under the quote according to the scicomp page. I will
> clean some more today (mostly rga) but I'd say we can try to squeeze it all
> on work (unless there are reasons not to).
> Decoded files are stored in the cache/silo.
> Nick.
>
> On Fri, Aug 30, 2019 at 8:07 AM Latifa Elouadrhiri <latifa at jlab.org>
> wrote:
>
>> Dear Raffaella
>>
>> From our side Will will check on the old files from RGA and cleanup as
>> necessary
>> We will need RGK processed as soon as possible
>>
>> Best regards
>>
>> Latifa
>>
>>
>> On Aug 30, 2019, at 8:01 AM, Raffaella De Vita <
>> Raffaella.Devita at ge.infn.it> wrote:
>>
>> Dear Annalisa and Latifa,
>> before starting a massive data processing, it will be necessary to have
>> an estimate of the disk space needed. Currently /volatile/clas12 is fully
>> used by rg-a and rg-k and there is very limited space left on /work/clas12.
>> Specifically, I would like to ask you to give an estimate of:
>> - disk used or needed for decoded files
>> - disk used or needed for DSTs
>> indicating the space and file server you intend to use.
>> In parallel, I also would like to ask you to check the files that are
>> currently under /work/clas12/rg-a and /work/clas12/rg-k to see if there are
>> old files that can be deleted to clear space.
>> I’m confident that with some planning, we’ll find the resources needed.
>> Best regards,
>> Raffaella
>>
>> On 30 Aug 2019, at 13:28, Latifa Elouadrhiri <latifa at jlab.org> wrote:
>>
>> We need to start cooking everything as soon as possible
>>
>> Best regards
>> Latifa
>>
>>
>>
>> On Aug 30, 2019, at 5:48 AM, Annalisa D'Angelo <
>> annalisa.dangelo at roma2.infn.it> wrote:
>>
>> Dear All,
>>
>> please start to think about priorities about which runs to cook for
>> pass1. We may want to discuss it at today's meeting.
>>
>> All the best
>>
>> Annalisa
>>
>>
>> -------- Messaggio Inoltrato --------
>> Oggetto: RGK Decoded files
>> Data: Thu, 29 Aug 2019 19:12:18 -0400
>> Mittente: markov <markov at jlab.org> <markov at jlab.org>
>> A: Annalisa D'Angelo <annalisa.dangelo at roma2.infn.it>
>> <annalisa.dangelo at roma2.infn.it>
>> CC: Latifa Elouadrhiri <latifa at jlab.org> <latifa at jlab.org>
>>
>> Dear all,
>> since we are approaching the cooking stage,
>> I have checked if we have enough decoded files handily available from
>> both 6.5 GeV and 7.5 GeV.
>> It looks like we can have at least 30 runs in each of the sub periods,
>> I have attached short summary (in reality we have more decoded files from
>> each of the sub periods, not sure if we will be able to cook all of them on
>> time).
>> I think it is about the time to decide what and when we cook.
>> Best regards, Nick.
>>
>> <rgkDecoded.xlsx>
>>
>> _______________________________________________
>> clas12_rgk mailing list
>> clas12_rgk at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/clas12_rgk
>>
>> _______________________________________________
>> clas12_rgk mailing list
>> clas12_rgk at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/clas12_rgk
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/hybrid_baryons/attachments/20190830/23b5483d/attachment.html>