A common GRID approach for the LHC experiments

A common GRID approach for the LHC experiments F. Carminati PEB 15 January 2002 1

Current situation The four collaborations are collaborating within EDG-WP 8 A lot of ground has been covered toward a constructive dialog and coordination However n n We still are building a global vision of what we need from GRID We still do not have a completely defined GRID architecture to adapt our applications to January 15, 2002 PEB 2

M ar 20 ch 01 9 What we want from EDG Specific application layer VO common application layer Data. GRID ATF GLOBUS team ALICE ATLAS CMS LHCb HEP WP 9 WP 10 Earth Obs. Biology WP 8 -910 TWG Data. GRID middleware PPDG, Gri. Phyn, Data. GRID Bag of Services (GLOBUS) OS & Net services January 15, 2002 PEB 3

What we have with TB 1 Specific application layer ALICE ATLAS WP 1 CMS LHCb WP 2 WP 3 WP 9 WP 4 WP 10 WP 5 Data. GRID middleware GLOBUS team Bag of Services (GLOBUS) OS & Net services January 15, 2002 PEB 4

The long term view TB 1 is the result of a tremendous effort done in a very short time The original design was based on our use cases (deliverable 8. 1 b) However our four use cases still lack coherence n n In reality we are doing more or less the same thing! We produce read only files with atomic entities (events) We refine them progressively We do data mining We have to be able to express this in a much more uniform way We could be able also to associate EO and Biology January 15, 2002 PEB 5

How to proceed CMS Biology ATLAS Core Common common use cases case ALICE January 15, 2002 EO LHCb PEB 6

A proposal It If Orwill we even manage bebetter easiertofor define them to arrive at Specific application layer VO commonlayer application layer ALICE ATLAS CMS LHCb HEP WP 9 WP 10 Earth Obs. Biology WP 8 -910 TWG Common core case Common useuse cases WP 1 WP 2 WP 3 WP 4 WP 5 Data. GRID middleware GLOBUS team Bag of Services (GLOBUS) OS & Net services January 15, 2002 PEB 7

A proposal The current cooperation should be extended beyond EDG to the global GRID projects of the four experiments in the LCG level n This would ensure coordination with the US side for ALICE, ATLAS and CMS We should build on n n The current coordination we have in WP 8 The testbed plans we are discussing since more than one year in Data. GRID A common set of use-cases would help n n Clarify and rationalize our demands to the different GRID MW projects (i. VDGL, Gri. Phy. N, PPDG, EDG, …) Run trans-Atlantic tesbeds making meaningful comparisons January 15, 2002 PEB 8

How to get there To be proposed on Friday 18 at the ATWG The EIP should n n Read carefully 8. 1 b Interview the experiment architects Prepare a set of common use cases that must be implementable by the experiment frameworks Iterate with the architects till there is agreement w Ideally a couple of months We will finalise this proposal at an architect – GRID expert meeting n Have the different experiments accept this proposal as their GRID baseline January 15, 2002 PEB 9

GRID interoperation Application Gri. Phy. N – PPDG – i. VDGL EDG January 15, 2002 PEB 10

Why this is important? The LCG will require experiments to work on common projects n We need a common set of requirements / use cases to define common deliverables The HICB expects proposal from the experiments n n n It would be effective to provide a common set of use cases Instead of competing one with the other We will try to get to Toronto (HIJTB+GGF) with a roadmap on how to get to the common use cases The different GRID MW activities risk to diverge n n Common use cases could help them to develop coherent solutions Or ideally complementary elements January 15, 2002 PEB 11

Conclusion The idea of coordination between GRID activities of the four experiments came naturally out of the positive experience of WP 8 We are confident that we can extend this experience to the overall GRID activities of the four experiments This would help n n Focusing the requirements to the different MW projects Perform meaningful trans-Atlantic tests and comparisons Developing common tools for the four experiments Reinforcing the partnerships that are being established between different GRID MW projects The four experiments support this activity January 15, 2002 PEB 12