ATLAS SCCS Planning 22Aug2007
-----------------------------
9am, SCCS Conf Rm A, to call in +1 510 665 5437, press 1, 3935#
Present: Wei, Booker, JohnB, Randy, Stephen
Agenda:
1. DQ2 Status/Web Proxy
Has been some problems with DQ2, not sure if it is at SLAC or
not. There have been problems both at CERN and BNL. PANDA has
registered to transfer many files from tape at BNL to SLAC so we
are waiting for so long that FTS cuts off the transfer. This is
either a tape problem or we need to change the software to wait
longer. Michael Ernst is doing something else to bypass the DQ2
system, he'd like SLAC to participate in this test. Not starting
really from scratch, will still use the catalogue system. Will only
use gsiftp not SRM. Worries about SRM being production quality.
There wasn't any intervention on DQ2 till yesterday. Booker
followed Wei's instructions for during the power outage and it
seemed to work. Wei yesterday cleaned all the FTS pipe etc. Things
looked better since.
Would be interested to know which type of tape libraries and drives
(and how many) are in use at BNL.
2. Tier-2 Hardware
The only thing we can imagine ordering right now are Dell 1950s. As
our code ran slightly better by clock than the AMD the ATLAS code
is able to take advantage of the larger caches on Intel but not be
too disadvantaged by the lower memory bandwidth.
We have some Dell 1950s on order for BaBar to replace nomas and
toris, there will be 120 to be installed in the water cooled
racks. Not sure if there will be another order for BaBar by the end
of the calender year. Thing there might be but not sure.
3. AOB
Action Items:
-------------
070905 Wei Determine BNL contact for tape system information
|