sPHENIX discussion of software

[pinkenburg <pinkenburg AT bnl.gov>]

Hi Christof,

I'll have a look a bit later - just quickly, the run 3 files have been 
replaced by run 4 (which should include the new TPOT). I changed the 
gpfs dir to be consistent with luster which does not speak caps for 
directories when using minio. They are now under 
/sphenix/sim/sim01/sphnxpro/mdc2/shijing_hepmc/fm_0_20 with names like:

DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000004-00000.root

The CreateFileList.pl (with -run 4) will give the usual list, segments 
0-9 are in gpfs and will be picked first.

Chris

On 4/14/2022 8:21 AM, Christof Roland via sPHENIX-software-l wrote:
> Hi Chris,
>
> I have started reshuffling the MDC2 macros.
> We wanted to split the pass4 tracking macro into two separate macros
> 1) run the clustering and drop the hitsets
> 2) run the tracking on clusters only
>
> At the same time we wanted to remove all unneccessary libraries that are not 
> needed.
>
> I have 2 macros stripped more or less to the bone:
> /phenix/u/bogui/data/MacrosFileSize/macros/detectors/sPHENIX/
> ReadClustersRunTrackingMDC2.C
> ReadHitsRunClusteringMDC2.C
>
> What is missing still is the switch in the G4_Tracking.C to not load the acts 
> fieldmap for the clustering step.
> Not sure what the status of Joe's switch is. I'll add this as soon as the 
> pull request is merged.
>
> I would like to start benchmarking these macros.
> Unfortunately my benchmark samples have dissappeard from gpfs.
> In /sphenix/sim/sim01/sphnxpro/MDC2/sHijing_HepMC/fm_0_20/
> we used to have files with 50kHz pileup: 
> DST_TRKR_HIT_sHijing_0_20fm_50kHz_bkg_0_20fm-0000000003-0000?.root
> They are unfortunately not there anymore. Could we restor those, please?
>
> Regarding dropping the hitsets we have a conceptual problem.
> We use the hitsetcontainer to navigate the ClusterMapContainer.
> If we drop the hitsetcontainer we can't access the cluster map anymore.
>
> As a workaround I would propose to add the functionality to create an empty 
> hitsetcontainer and keep it on the node tree to use as navigation tool.
> Does anyone have a problem with this?
>
> Cheers
>
>     Christof
> _______________________________________________
> sPHENIX-software-l mailing list
> sPHENIX-software-l AT lists.bnl.gov
> https://lists.bnl.gov/mailman/listinfo/sphenix-software-l

--
*************************************************************

Christopher H. Pinkenburg       ;    pinkenburg AT bnl.gov
                                ;    http://www.phenix.bnl.gov/~pinkenbu

Brookhaven National Laboratory  ;    phone: (631) 344-5692
Physics Department Bldg 510 C   ;    fax:   (631) 344-3253
Upton, NY 11973-5000

*************************************************************