Last modified 12 years ago Last modified on 01/20/09 17:31:50

Tier I dCache deployment Phone conference Jan 20, 2009

Participants,Gerd,Patrick), IN2P3(), SARA(), GridKa(Doris), PIC(Gerard), Triumf(), BNL(Iris), Fermi(Timur), NDGF(Mathias), CERN(Flavia)

Special Topics


In the daily WLCG operations meeting, general overload of grid components (SE) are reported. There has been a discussion on how to cope with the fact of overloading SE's. This could essitially be done by :

  • Limit the number of requests, which is possible with known clients but could be challenge doing analysis.
  • Making clients smarter, e.g. backing off if load increases.
  • Let the SE hardware grow with the number of CPU's deployed at a site.


Tier I's which would like to have access to the dCache support ticket system should please send an e-mail to patrick.

Site reports

  • Gerard : Nothing except sometimes high load, but not dangerios yet.
  • Iris : nothing
  • Timur (for USCMS) : nothing
  • Mathias : nothing for the system.
  • Doris : high load as discussed during the workshop.

Last Modified Paul @ Wed Mar 3 07:25:39 2021