I think that If we have time we can try:
why we need 4 different collections?
Now we run all four seeds in parallel and we can drop this
distinction and run on only one collection:
what do you think?
alessio
(We have time just before sunday morning)
______________________________________________________
Alessio Sarti Universita' & I.N.F.N. Ferrara
tel +39-0532-781928 Ferrara
roma +39-06-49914338
SLAC +001-650-926-2972
-----------------------------------------------------------------------
La vita non e` un prodotto da essere rifiuti consumati da fare parte di
una lavata di omicidio l'anima dal vostro vegan in tensione delle mani
-----------------------------------------------------------------------
M. Pierini "That's why do you really want to hit me"
---------- Forwarded message ----------
Date: Fri, 19 Jul 2002 18:11:29 -0700
From: Yury G. Kolomensky <[log in to unmask]>
Reply-To: [log in to unmask]
To: Alessio Sarti <[log in to unmask]>
Cc: [log in to unmask], Alessandra Forti <[log in to unmask]>,
Giuseppe Finocchiaro <[log in to unmask]>,
Urs Langenegger <[log in to unmask]>, [log in to unmask],
[log in to unmask]
Subject: Re: skimData
Hi Alessio,
On Fri, Jul 19, 2002 at 05:27:43PM -0700, Alessio Sarti wrote:
>
>
>
> > Hi Alessio,
> >
> > if you are looking for *skims*, there are two problems with your
> > query:
> >
> > a) The collections are in skim table, not objy table
> > b) The *skim* collections are stored, not *streams*
>
> What i have to do in order to have all the various skims contained in a
> given stream?
>
You need to get separate tcl files for each skim.
Which brings us to an important question. Somebody (perhaps Ric and
Daniele) set up semi-exclusive collections by mode:
declarePath BToDstarSemiExclPath Breco/BToDstarSemiExcl Merged/Stream06 BrecoStream
declarePath BToDcSemiExclPath Breco/BToDcSemiExcl Merged/Stream06 BrecoStream
declarePath BToDstar0SemiExclPath Breco/BToDstar0SemiExcl Merged/Stream06 BrecoStream
declarePath BToD0SemiExclPath Breco/BToD0SemiExcl Merged/Stream06 BrecoStream
This means that if you want to run over all these events, you need to
run 4 jobs (or configure your job to read 4 collections). Issues of
event duplication etc. will clearly come up. I am not sure why it was
set up this way -- historic, perhaps, or maybe despite my numerous
talks and posts on the subject, the difference between skims and
streams was not clearly understood. Basically, streams were an
artificial construction, for data distribution: built originally for
Kanga, and migrated into Objy temporarily in release 10. You should
have never relied on all these events being in one stream. If you
wanted one giant collection for all semi0-exclusive modes, you should
have defined one collection in FilterTools.
So my question is -- do you want 4 collections, or 1 collection ? If
you want a single collection, there is still time (until early Sunday)
to change this setup for release 12 (Padova) reprocessing. Otherwise,
you will be getting 4 collections for each run next year.
Do I need to forward this question to vub-recoil list ?
Cheers,
Yury
> Alessio
>
> >
> > In other words, I'd do this:
> >
> > $BFROOT/bin/skimData -I "<R10.3.1a" --basename old2000 -g 600000-749000 -s BtoDsD -m "B+B- generic"
> > skimData: Used 429 of 429 inputs (0 superceded)
> > skimData: Selected 963173 of 9077194 unique events (10.61%)
> > <snip>
> >
> >
> > If you are looking for collections produced directly in SP production,
> > I believe you need to select AllEvents stream -- I do not think other
> > streams were ever produced. --tableprefix objy is appropriate then.
> >
> > Yury
> >
> >
> > On Fri, Jul 19, 2002 at 04:55:13PM -0700, Alessio Sarti wrote:
> > > Hi,
> > > we are experiencing a problem using skimData...
> > > We are trying to get tcl files for objy SP4 MC collections skimmed
> > > (stream6). We are intrested in generic MC events (and also cocktail):
> > >
> > > Here is what I'm trying to use for generic MC :
> > > # MC 2000
> > > ~/perl/skimData -I "<R10.3.1a" --basename old2000 -g 600000-749000
> > > --tableprefix objy --s Stream6 -m "B+B- generic" --tcl 20k
> > >
> > > This is not working:
> > > skimData: Used 0 of 0 inputs (0 superceded)
> > > skimData: Selected 0 of 0 unique events (0.00%)
> > >
> > > Release | P10.2.3i, Out/In(Frac) P10.2.3h, Out/In(Frac)
> > > P10.2.3g, Out/In(Frac) P10.2.3f, Out/In(Frac) P10.2.3e,
> > > Out/In(Frac) P10.2.3c, Out/In(Frac) P10.2.3b, Out/In(Frac)
> > > P10.2.2, Out/In(Frac) P10.1.0a, Out/In(Frac) P10.0.3b,
> > > Out/In(Frac) P10.0.3a, Out/In(Frac) P10.0.2b, Out/In(Frac)
> > > P9.10.3c, Out/In(Frac) P9.10.3b, Out/In(Frac) P9.10.3a,
> > > Out/In(Frac) P9.10.2a, Out/In(Frac)
> > > -------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > > -------------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > >
> > >
> > > The same happens if I try on specific samples of cocktail MC.
> > > For example:
> > > # MC 2000
> > > ~/perl/skimData -I "<R10.3.1a" --basename old2000 -g 600000-749000
> > > --tableprefix objy --s SP -m "Weighted B0 -> DX and DstarX cocktail" --tcl
> > > 20k
> > >
> > >
> > > this is working,
> > > but this is not
> > > ~/perl/skimData -I ">=R10.3.1a" --basename new2000 -g 1200000-1369999
> > > --tableprefix objy --s SP -m "Weighted B0 -> DX and DstarX cocktail" --tcl
> > > 20k
> > >
> > >
> > > Where can I found information about all the MC skimmed? Shall I use a
> > > different script to get this information (getData)?
> > > Why those commands are not working? Thanks much for you help.
> > > Alessio
> > >
> > > ______________________________________________________
> > > Alessio Sarti Universita' & I.N.F.N. Ferrara
> > > tel +39-0532-781928 Ferrara
> > > roma +39-06-49914338
> > > SLAC +001-650-926-2972
> > >
> > > -----------------------------------------------------------------------
> > > La vita non e` un prodotto da essere rifiuti consumati da fare parte di
> > > una lavata di omicidio l'anima dal vostro vegan in tensione delle mani
> > > -----------------------------------------------------------------------
> > > M. Pierini "That's why do you really want to hit me"
> > >
> >
> > ----------------------------------------------------------------------
> > Yury G. Kolomensky
> >
> > UCB Campus: 439 LeConte, (510)642-9619
> > LBNL: 50A-2145, (510)486-7811
> > SLAC: Bldg 280, room 172, (650)926-5188
> > Fax: (510) 495-2957
> > Cell: (925)437-2425
> > Pager: (650)846-0289
> > LBL mail: Physics Division/BaBar Group
> > Lawrence Berkeley National Laboratory
> > Bldg 50A2160
> > 1 Cyclotron Road
> > Berkeley, CA 94720-8143
> >
>
|
