[Halld-physics] Analysis Launch: 2017-01 ver01

[Alexander Austregesilo]

Dear Colleagues,

Due to constraints by the JLab tape library, I merged the trees for each 
plugin into run-wise chunks. You may find the these trees in the 
subfolder "merged", similar to the trees produced in the reconstruction 
launch. I will unpin the individual files now and just keep the merged 
versions on the cache disk.

In the future, we will not keep the analysis trees per raw file at all. 
They will only be stored temporarily and deleted after the run is 
completely processed and the trees are merged.

Best regards,

Alex



On 08/04/2017 04:05 PM, Alexander Austregesilo wrote:
>
> Dear Collaborators,
>
> The first analysis launch over all of 2017-01 ver01 REST files is 
> completed. Only 20 out of 42108 jobs showed problems and could not 
> produce any output trees.
>
> 60 different channels were processed. The total processing time was 
> only about 50h, since we were able to run with up to 700 jobs in 
> parallel with 12 threads each. This huge speed-up compared to previous 
> launches was made possible by a complete rework of the analysis 
> library (Paul M.) and tighter cuts on the accepted combinations.
>
> All trees (about 27TB) and histogram files (3TB) can be found here:
>
> /cache/halld/RunPeriod-2017-01/analysis/ver01/
>
>
> Feel free to consult the following webpage for information about the 
> available channels:
>
> https://halldweb.jlab.org/wiki-private/index.php/Spring_2017_Analysis_Launch#Version01
>
>
> Detailed information about the processing can be found here:
>
> https://halldweb.jlab.org/data_monitoring/analysis/summary_swif_output_analysis_2017-01_ver01_batch01.html
>
>
> A second launch with channels including displaced vertices, missing 
> decay products and detected massive neutral particles will be started 
> as soon as the library is thoroughly tested.
>
> Best regards,
>
> Alex
>
>
> _______________________________________________
> Halld-physics mailing list
> Halld-physics at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-physics

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-physics/attachments/20170811/75e92dd3/attachment.html>