Weekly SubMIT project team meeting

→ America/New_York
ZOOM

ZOOM

David Walter (Massachusetts Institute of Technology), Mariarosaria D'Alfonso (Massachusetts Institute of Technology)
    • 09:00 → 09:20
      Discussion 20m
      Speakers: David Walter (Massachusetts Institute of Technology), Luca Lavezzo (MIT), Mariarosaria D'Alfonso (Massachusetts Institute of Technology), Matthew Heine (Massachusetts Institute of Technology)

      scratch:

      increase the 'buffer' size of another 0.5 or 1T to avoid slow reading when too full

      small cleanup:

      1. mariadlf/david_w/ bendavid/plugato/srothman already cleaned stale data
      2. to do cleanup of Gaia folder /scratch/submit/gaia/ with 1.1T copied from Paus and root since 2022; a copy already seems to exist in /ceph/submit/data/group/gaia/; Matt contact Lina
      3. Maria contact karich (Kate Richardson) since she left MIT for industry and has 3.44 T (also report on the security audit )

       


      new Nodes from T3 (David):

      1. upgraded to alma9 (only HTcondor before, now also slurm)
      2. on slurs is a new partition since it's old hardware and things do not run
      3. all disk mounted (home work submit scratch)

       

      soon old nodes also from T2 (next couple of weeks)

      TO DO:

      1. David write article (maybe wait for T2nodes)
      2. user guide:  update so that limitation are clear to everybody https://github.com/mit-submit/submit-users-guide/pull/119
      3. admin guide: the static page should be updated with these new nodes, at the same time think how to re-organize since those are better of read from the https://github.com/mit-submit/submit-admins-guide/blob/main/source/server.rst
      4. ganglia update: David add in the ganglia out of the box , Matt add the customization

      Matt: reorganization of those alarms ongoing 


      AOB:

      1. Paul -- need add new MOIRA group (this request pending 10 days https://cleo.mit.edu/issues/6322)

      2. received the request from Holger Witte and Risa Hocking, both from Bates, on two sw: 

        1. BELFEM, need OpenMPI --> can run out of the box

        2. COMSOL, interface with IntelMPI or MPICH2.
      3. atm noOne is planning to join the submit-project group