NPPS Leadership Team

[Torre Wenaus <wenaus AT gmail.com>]

FYI. Informing Phil Marshall of Rubin of Sergey's departure (not a false alarm this time).

---------- Forwarded message ---------
From: Torre Wenaus <wenaus AT gmail.com>
Date: Tue, Dec 14, 2021 at 2:48 PM
Subject: Re: Rubin Ops and PanDa etc
To: Phil Marshall <dr.phil.marshall AT gmail.com>
Cc: Nomerotski, Andrei <anomerotski AT bnl.gov>, Richard Dubois <dubois AT stanford.edu>

Hi Phil, (adding Richard)

Well the other shoe dropped, Sergey tells us he belatedly got an offer from another of his August interviews and this one he cannot pass by. He will leave at the end of this year. He's initiated meetings with his BNL and PanDA team colleagues to transfer his responsibilities.

Better now than three months ago! Sergey says that Rubin's use of PanDA has become much more robust and production-ready, with many issues fixed and with new features development now supposed to be frozen for an extended period. He has also been more directly involving (in addition to Shuwei) Wen Guan, the developer of iDDS which is a key component of how Rubin uses PanDA. Finally he says "So, on my understanding, everything that’s needed is to keep operations is in good shape and I think Shuwei and Wen are the right persons to keep things running. Having in mind that the expected hiring of a new PanDA developer got much closer I think my leave will not jeopardize the project and the replacement seems to be quite natural."

Also while three months ago the concurrent adoption of PanDA by sPHENIX was a liability in that it also was demanding of core developers' time, now it's an advantage. PanDA for sPHENIX is operating and ready for their imminent Mock Data Challenge, using the same CERN based service Rubin currently uses. The further work there is mainly in establishing a PanDA instance at BNL, backed by PostgreSQL and deployable with Kubernetes, ie the same as PanDA@SLAC. So good synergy between the two efforts going forward. We have someone working with us in the BNL facility (SDCC) on the BNL PanDA installation now underway; we'll need the same at SLAC when you're ready to begin PanDA@SLAC.

The PanDA position we have open at present requires that the person be based at BNL (as opposed to CERN) so they are not 9 time zones away from west coast Rubin folks. We so far have one applicant who makes the threshold to interview. But we have at least 2 and possibly 3 very good PanDA team members based at CERN (non-BNL) who would be interested in a BNL PanDA position at CERN but do not want to relocate to the US, at least at present. Sergey leaving means we have a second PanDA position to open, one that is not constrained to be BNL based. Our plan is to hire "quickly" (I work at a national lab, it has to be in quotes ;-) one of the PanDA experts at CERN, and have them come up to speed as the Sergey replacement. The way US based applications are going, we can probably open and fill that quicker than the existing opening. Once we do fill the existing US based opening, that person will also come up to speed on Rubin, and will become the primary Rubin PanDA person, with the new CERN based person giving the redundancy of Rubin PanDA expertise (in addition to Shuwei) that you want.

Sorry for the long mail and the unwelcome news, but my own feeling is that we'll all come out OK. Questions and comments welcome and I'm ready to discuss.

  Torre

--

-- Torre Wenaus, BNL NPPS Group, ATLAS Experiment

-- BNL 510A 1-222 | 631-681-7892 |  wenaus AT gmail.com | npps.bnl.gov | wenaus.com

-- NPPS Mattermost room: https://chat.sdcc.bnl.gov/npps/channels/town-square