[Attendee: Booker, John Bartelt, Randy, Chuck, Len, Wei]
1. DQ2/web proxy/DDM
DQ2 is running OK. Panda team still haven't got the new pilot working at SLAC. This problem has been there for a few weeks and only seen some progress in the last few days. AOD replication to WT2 is continuing.
2. Tier 2 hardware
Chuck said that if water cooling rack plan goes well, we could have all ATLAS computer nodes and storage installed by May/June. Plan B would be to seek the freed power and space of memfs machines. There is a gap of around 40 A of power in Plan B.
The current schedule is to bring online one of the thumper boxes (18TB) when they arrive in April. Wei proposed to retire sulky05 (1 Sun T3, 500GB) and sulky51 (2 Sun T3, each is 1.2TB, 1.7A measured). This will return 6-8A for another thumper box (5A) ASAP. We can use the current DQ2 space (2.3TB netapp) for an even exchange for the lost disk space for local users (sulky51 storage). Both Chuck and Randy endorse this plan so we will talk to local users.
Also want to keep some retired nomas and toris for griddev machines and ATLAS usage.
3. HPSS/gsiftp/xrootd
gsiftp is done. This item could be changed to "xrootd/srm". See below
4. Tier 2 meeting at San Diego
AOD size is 5 times bigger than estimated, to 500KB/event. So we need addition storage boxes online ASAP.
Michael Ernst emphasized the need to have a matured storage system with the needed complexity to handle the real data. SRM interface is also important.
There will be a Full Dressed Rehearsal between June and November.
Some T2s are interested in xrootd, but want SRM. They are also looking at/using dCache. There is a serious problem with dCache SRM interface. Michael is leading an effect to address it.
Torre explained that pAthena jobs can do remote read from storage
SLAC need to work with DDM team (Patrick) on security issues.
a) prefer to use system provided mysql and apache. Otherwise, a seemly harmless access to unpatched mysql and apache from outside is still dangerous.
b) http interface to LRC has write access to LRC DB. SLAC security team would prefer that writing and reading URLs to use different ports (so that writing port can be blocked by firewall). This involve major change to Pool File Catalog.
Wei will research a) and will discuss both a) and b) with Patrick. b) will be a US ATLAS effect. We do not expect b) is easy.
5) AOB
PetaCache project invited ATLAS and CMS to test their jobs against a DRAM-based 1TB system.
Action Items:
-------------
061108 Richard Discuss with SLAC Security longterm approach to ATLAS VO
061115 No information.
061213 Nothing happened yet.
070103 No information.
070110 Richard & BobC in Denver, Stephen will email them.
070124 Don't know the status.
070131 Don't believe this happened.
070207 Have not done this. Randy has talked to Heather, didn't
have any time to comment today but she is aware about
it. Will treat each VO as an enclave, if you are using
anonymous accounts need to be able to show how ran a job
and when. The main issue is that VOs are not legal
entities and anyone can declare themselves a VO. Would
need to actually test that the required information can
actually be found.
070221 No info.
070228 No info.
070314 No info
--
Wei Yang | [log in to unmask] | 650-926-3338(O)
