Choose timezone
Your profile timezone:
Feasibility Study for a Common Analysis Framework
→
Europe/Zurich
28/S-029 (CERN)
Description
To join both the web and audio conference (recommended), click here.
If you only want to participate in the audio conference, you may dial in with the following numbers: +41227676000 (Main), and enter access code: 0166696
Meeting on a Common Analysis Framework between ATLAS and CMS - Pilot, Pilot factories and GlideInWMS discussions (13 April 2012)
Attendees:
- CERN: Alexei Klimentov, Daniel van der Ster, Stephen Gowdy, Maria Girone, Nicolo Magini, Paul Nilsson, Domenico Giordano, Tadashi Maeno, Mattia Cinquilli, Fernando Barreiro Megino, Torre Wenaus
- Remote: Rodney Walker, Jose Caballero, Claudio Grandi, Daniele Bonacorsi, Daniele Spiga, John Hover, Eric Vaandering
Dan gave an overview of the PanDA pilot and the PanDA Pilot Factories and then an overview of the GlideInWMS and the current integration state in ATLAS and the usage in CMS.
- There was discussion about the three presented scenarios for the integration of the GlideInWMS in the experiment WM frameworks. Ian asked about the PanDA rebrokerage capabilities under these 2 scenarios and pointed out that this is one of the interesting features in PanDA. Dan answered that in scenario B this would be more complicated, Rod added that PanDA would need to interact with Condor in this case. More thought needs to be invested in the scenarios.
- Alexei asked about SSH-to-the-job feature and if sites can decide to turn it off/block it. Rod answered that only outbound connectivity is needed on the WNs and that it can't be blocked by sites. SSH-to-the-job is considered a nice feature, but it is not essential.
- Rod reported to be running 4k jobs in his large scale glideInWMS tests and at the moment it seems stable. He needs more sites and further collaboration with UCSD. Also Quattor is restarting Condor randomly, which does not help him.
- On Thu April 19 in the ADC TIM in Annecy there is a 45+min block to discuss the Common Analysis Framework. Fernando will give summary presentation. The same day Rod will present the status of his tests to integrate PanDA and the glideInWMS. Alexei will ask Stephane Jezequel about the possibility for remote participation. If not, the next meeting will be held Friday April 27.
Attendees:
- CERN: Alexei Klimentov, Daniel van der Ster, Stephen Gowdy, Maria Girone, Nicolo Magini, Paul Nilsson, Domenico Giordano, Tadashi Maeno, Mattia Cinquilli, Fernando Barreiro Megino, Torre Wenaus
- Remote: Rodney Walker, Jose Caballero, Claudio Grandi, Daniele Bonacorsi, Daniele Spiga, John Hover, Eric Vaandering
Dan gave an overview of the PanDA pilot and the PanDA Pilot Factories and then an overview of the GlideInWMS and the current integration state in ATLAS and the usage in CMS.
- There was discussion about the three presented scenarios for the integration of the GlideInWMS in the experiment WM frameworks. Ian asked about the PanDA rebrokerage capabilities under these 2 scenarios and pointed out that this is one of the interesting features in PanDA. Dan answered that in scenario B this would be more complicated, Rod added that PanDA would need to interact with Condor in this case. More thought needs to be invested in the scenarios.
- Alexei asked about SSH-to-the-job feature and if sites can decide to turn it off/block it. Rod answered that only outbound connectivity is needed on the WNs and that it can't be blocked by sites. SSH-to-the-job is considered a nice feature, but it is not essential.
- Rod reported to be running 4k jobs in his large scale glideInWMS tests and at the moment it seems stable. He needs more sites and further collaboration with UCSD. Also Quattor is restarting Condor randomly, which does not help him.
- On Thu April 19 in the ADC TIM in Annecy there is a 45+min block to discuss the Common Analysis Framework. Fernando will give summary presentation. The same day Rod will present the status of his tests to integrate PanDA and the glideInWMS. Alexei will ask Stephane Jezequel about the possibility for remote participation. If not, the next meeting will be held Friday April 27.
There are minutes attached to this event.
Show them.