 |
 |
Hi Maria, Patricia, The point of having separate VOs and individual VO AUPs is to allow site administrators: 1) to gauge the impact supporting a given VO will have on their local resources, 2) to authorize use conforming to local policies, and 2) to account correctly for the resources used. Having a scientific collaboration use the dteam VO violates all three of these: The dteam use may far exceed that expected from the AUP, support of Geant4 might not be in keeping with local policies (however laudable the activity is), and the use gets lumped in with "testing" instead of identified with a specific experiment. Moreover as sites must support the dteam VO, this puts the Geant4 collaboration in a preferential situation in that it can use every resource on the LCG production service. I'm sure that's a situation may other VOs would like to have. A case could be made to use Geant4 as a test package for low-level tests, as Patricia stated in her second point. But this is an activity very far from the large scale production that was described. Her other two arguments are completely beside the point. First, MAGIC and Planck both at the moment contain very few people, yet they have gone through the effort to be officially recognized have a separate VO. Second, I'll grant that there is a software distribution problem. However this can be attacked and tested without having to grant priviledged access to a specific collaboration. I request that this collaboration be setup as a separate, proper VO and that the dteam VO not be used for similar activities in the future. Cal Maria Dimou-Zacharova wrote: > On Tue, 17 May 2005, Charles Loomis wrote: > > >>Hi Maria, >> >>Can you point me to the dteam AUP? I ask because of a message that Patricia >>posted on the LCG ROLLOUT list implying that Geant4 is continuing to use the >>dteam VO for large-scale production. My impression was that the dteam VO was >>uniquely for diagnosing problems and testing sites. Perhaps I've >>misunderstood the purpose of this VO. > > > DTEAM AUP: > > This VO is created, in order to include members associated with a > registered site, who are involved in its operation (Site Administrators). > > This appears indirectly in the Site Registration Requirements' document > http://edms.cern.ch/document/503198 as well. > > However, I discussed the issue with Patricia and I understand the > arguments in favour of GEANT4 running, for now, at least, as DTEAM. > Cheers > - maria > >>Cal >> >>... >> >>The request did not come from the site but from the experiment. In this >>case it was a Geant4 requirement and at this point they are running as >>dteam. Geant4 wanted to make a huge production and before they have to >>split their software in the largest possible number of sites. So its >>software has been installed in more than 20 sites in the DTEAM Experiment >>software area of the WNs. >> >> > >