 |
 |
I attended the above meeting at FNAL. Rough notes from the meeting can be found at http://www-iepm.slac.stanford.edu/about/status/trip-lhc-fnal-oct06.htm. About 40 people were present representing ATLAS, CMS, CERN, Internet2, ESnet, NLR, DoE, T0, T1 and T2 sites. It was a very successful meeting in exposing many impotant issues. Action items include: getting an AUP (Acceptable Use Policy) for LHCnet; getting traffic estimates for T1-T2s; to assist in helping ISPs understand LHCnet make it closely associated as part of ESnet. The immaturity of PerfSONAR was exposed (it is really an architecture into which many people are contributing to fit in measurements, analysis, visualization, diagnosis etc.) and the need to figure out where MonALISA and PerfSONAR fit together was exposed. However, open source nature of PerfSONAR, the big buy-in to PerfSONAR from GEANT, Esnet and LHCOPN, the deployment and access to PerfSONAR measurement points at startegic points for LHC networking, and the large community working on PerfSONAR make it the chosen path for LHC (as compared to say MonALISA which is more mature, but not open source and thus more centralized). Also the need to better understand the T1-T2 traffic requirements became apparent. There will be T1-T2 network challenge April next year to ensure all T1-T2 links are performing adequately. Of the 12 Tier2 (7 with CMS, 5 with ATLAS), only 3 are not at 10Gbits/s yet, SLAC is one of the three. There will be another meeting next year.