Last Friday, the S&C Team management met with the PWGC for the second "PWGC / S&C Exchange meeting". The discussion was productive and we coverred lots of grounds including progrss and status since the last meeting we had, a overall status of data production, recent developments in tracking and vertex finding as well as the potential to run a massive data production on NERSC/Cori as part of recently acquired 25 Million hours allocation. We also asked of the PWGC to prepare a new data production plan and review the datasets on distributed disks.

For the data production, the good news is that we are doing very well - Most of the 2016 Au+Au 200 GeV that was asked have been produced and we even started to produce the AuAu200_production2_2016 trigger setupname. A reminder that we have a Production Dataset Inventory page showing what is complete and what not. From the past samples, we rediscussed the fate of the st_sst sream (it will be produced as the rest of the data) as well as the Run 2015 p+p 200 GeV ssdmb which we will run with the latest calibrations available and a different tracking tune. The 200 GeV d+Au Run 16 is also moving forward (in fact, at the time of this newsletter it is already 22% done). Finlly, we are targetting the whole st_mtd as a sample to run on our NERSC/Cori allocation - despite a new production plan, slow event processing is best for this Hogh Performance Computing infrastructure (where IO capabilities are limited).

For the NERSC/Cori exercise, we are also in the proces of upgrading and tuning our grid gatekeeper infrastructure. A this point, the data transefr from BNL to NERSC seem to show a performance above 600 MB/sec data transfer rate. We are working with our NERSC colleague to understand the link speed in the reverse direction (not symetrical). In all cases, the good news is that the transfer rate is above of requirement needs and we should be in good shape to re-start the full end-to-end production workflow.


 

Now that the run is in full ramp-up, many members of the team are also focused and shifting attention to its support. All databse daemons and information gathering procsses were restored and FastOffline resumed . We wish all a productive shift and as usual, if there are run-related problems, please use the rts Hypernews forum for communication.