The Caltech High Energy Physics Department |
Shown above is a system level description of the GAE architecture that was discussed at the workshop. (Click on the image to get the full size version.)
A few notes about the diagram:
1) The "process wrapper" service is used to simplify the application level interface to grid activities. Not all applications will need to know the details of how to run the job planner and the job execution service.
2) I've separated the services into many Grid Service Hosts to make clear the separation of services. Some services, such as MonaLisa and the condor pool, need to live on the same server. There are still some cases where I think we could split it up even more (RMS and JPES don't need to live on the same host).
3) I've mapped projects onto the diagram to show what's being worked on and what's missing. I know that I'm missing some pieces (RMS, for example). Feel free to help me fill in the blanks.
4) I still need to add some sort of execution flow to show the order in which things happen. Some connections aren't shown to keep from cluttering up the diagram (Steering <-> JPES, Data Collection <-> JPES).
The entire Workshop was available in the VRVS "Island" room.
Time | What | Who | What | Where |
8:15-8:45 | Breakfast | 248 Lauritsen (Building 48) Campus map |
8:45-9:15 | Welcome, Opening Remarks, Goals and Agenda | Harvey Newman (video) |
9:15-9:30 | Overview of GAE work at Caltech | Julian Bunn (video) |
9:30-9:45 | MonALISA | Iosif Legrand (video) |
9:45-10:00 | Simulating GAE | Iosif Legrand |
10:00-10:15 | Clarens | Conrad Steenberg (video) |
10:15-10:30 | SOCATS | Eric Aslakson (video) |
10:30-11:00 | Coffee break | All |
11:00-11:15 | PROOF | Maarten Ballintijn (video) |
11:15-11:30 | Java Analysis Studio | Tony Johnson (video) |
11:30-11:45 | DIAL | David Adams (video) |
11:45-12:00 | VDT | Rick Cavanaugh (video) |
12:00 - 1:15 | Lunch will be served al fresco | "Beef and Chicken Shawarma" |
1:15-1:30 | GANGA | Craig Tull (video) |
1:30-1:45 | US-CMS analysis plans | Lothar Bauerdick (video) |
1:45-2:00 | Event Selection Manager: Enabling File-Transparent Event Access from the Grid | John Wu (video) |
2:00-2:15 | Distributed Analysis at LCG | Torre Wenaus (video) |
2:15-2:30 | LCG - Simulations | Kathrin Paschen |
2:30-2:45 | Distributed Heterogeneous Data Warehouses for Analysis | Saima Iqbal |
2:45-3:00 | dCache for GAE | Ian Fisk (video) |
3:00-3:15 |
HEPCAL (and DAWN) The DAWN Proposal www.cern.ch/fca/HEPCAL-prime.doc
|
Ruth Pordes (video) |
3:15-3:45 | Coffee/Tea Break | |
3:45-6:00 |
DemonstrationsThe demonstrations will all be given concurrently, and attendees will be encouraged to mill around the room and look at each. We will install a 4-screen analysis station in the meeting room (shown to the right) for demos: if you want to use this setup, please contact Michael Thomas Clarens (Conrad Steenberg, Michael Thomas) DIAL (David Adams) Chimera/VDT (Dimitri Bourilkov) SOCATS (Eric Aslakson) PROOF (Maarten Ballintijn) Java Analysis Studio (Tony Johnson) |
l |
6:00 | Adjourn |
Time | What | Who | What | Where |
8:30-9:00 | Breakfast & Talks Unfinished from the Day Before!!! | 248 Lauritsen (Building 48) Campus map |
9:00-9:15 | PDA/JAS client Notes and Screen Shots |
Ashiq Anjum |
9:15-10:30 | Discussion of Goals and Workplan | All |
10:30-11:00 | Coffee break | |
11:00-12:45 | Candidate Topic 1: GAE Architectures for LHC global analysis and
remote working CS11 references for today's workshop:
|
All |
12:45 - 2:00 | Lunch will be served al fresco | "Super Mexican Feast" |
2:00-4:00 | Candidate Topic 2: Simulating GAE Architectures (task characteristics,
persistency mechanisms, monitoring, manual and automatic task execution
systems, etc. etc.)
Blackboard showing task life: Full Image and separate image1, image2. Motivational slides from Harvey
|
All |
4:00-4:30 | Coffee/Tea Break | |
4:30-5:30 | Candidate Topic 3: Existing Analysis software, Grid production
systems, integration in candidate Architecture from Topic 1. Blackboard showing boxology of architecture: Full Image and Notes |
All |
5:30 | Adjourn | |
6:30- | Workshop Dinner | Athenaeum
(on campus)
|
Time | What | Who | What | Where |
8:30-9:00 | Breakfast | 248 Lauritsen (Building 48) Campus map |
9:00-10:30 | Candidate Topic 3 (continued) | All |
10:30-11:00 | Break | |
11:00-12:15 | Future Activities, relationships with LHC Experiments, LCG, US-Grid
Projects, CrossGrid, etc. etc. Blackboards on Work to be done: Text version of Work to be done
Grid Service Primer (from Globus).
|
All |
12:15-12:30 | Conclusions, comments, Workshop Wrap-up Summary (HTML Version) (Excel Version) (Torre Wenaus)
|
All |
12:30 | Adjourn |