NPPS Leadership Team

[pinkenburg <pinkenburg AT bnl.gov>]

    Hi Torre,
    
    I really think there is a misconception in the administration in a
    sense that computing is "just software". At least for PHENIX rcf
    hasn't coded up a lot (the production system and the hpss batch
    system being the exceptions I can recall). All they do is provide
    services (and they used to be good at it - no criticism). Now that
    some key people left it is more problematic.
    
    My main concern is that this will just aggravate my current problem
    with responsiveness and working closely together. Under Michael it
    was certainly different in terms of having resources to try stuff
    and actually hashing out solutions for our problems together. Now
    it's just about "money first" which will have a severe impact on the
    capabilities of the physics dept.
    
    I got a taste of this during our covid effort. NP: we know how to do
    high throughput, we got a few nodes, we can run tests, here is some
    TB temporary disk space - took us 2 days to have something running.
    CSI (who formally had the lead): we need to see how this fits into
    our accounting scheme to run this, some quota needs to be allocated,
    jobs get terminated because someone ran out of quota. It took weeks.
    The vast majority of the jobs were run by Martin on our resources
    (with permission from DOE) in the end, CSI didn't contribute a lot
    to this in the end.
    
    I am horrified at the prospect of having to set up sPHENIX computing
    under these conditions. It's just not going to work. The mdc gives
    me a first real glimpse of what we are up against and it ain't
    pretty. It's solvable (I would say) but not if rcf turns into some
    remote appliance which sees us only as a source of funding.
    
    Chris

On 12/11/2020 5:39 PM, Torre Wenaus via Phys-npps-mgmt-l wrote:

Hong's comments to me on Eric's plan...

                It is considered inevitable that RACF as part of SDCC will leave Physics, because of the new computing facility, the lab’s plan for expanding computing, and Eric’s view of future.   Forming NPPS is to counterbalance it, so that there is a home for software development in Physics.  Your concern over SDCC expanding into software is valid.    We should certainly manage this risk.     I hope we have the advantage of you as the NPPS group leader, as opposed to the other organization with leaders who are eager to depart.

 

                It is not clear how it will work out in the near future for RACF.  Berndt hoped that by agreeing to move RACF out of Physics to NPP as a Center, it will at least remain in NPP.  But this has caused other complications.   We will see whether the plan will converge or not next week.    

  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

_______________________________________________
Phys-npps-mgmt-l mailing list
Phys-npps-mgmt-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/phys-npps-mgmt-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

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