Summary Introduction: vhe -ray astronomy and magic

Summary

• Introduction: VHE -ray astronomy and MAGIC

• Data handling at MAGIC

• GRID at MAGIC

• Virtual observatory

• Conclusions

VHE astronomy

Scientific highlights

• MAGIC has discovered 10 new VHE -ray sources (7 extragalactic + 3 galactic)

• New populations unveiled (radio-quasar, micro-quasar)

• Farthest ever observed object at these energies (z = 0.54)

• GRBs observed (not detected) during the prompt emission

• Big flares used to test Lorentz invariance

• So far, we have published ~30 papers and many more are in the pipeline

Raw data volume

Data flow

MAGIC/PIC Data center

Trends foreseen for 2008

• Philosophy:

• Adopt Grid to allow MAGIC users to do better science
• “If it ain’t broken, don’t fix it”

• Leverage worldwide mutual trust agreement for Grid certificates to simplify user ID management for:

• Interactive login (ssh → gsi-ssh or equivalent)
• Casual file transfer (https via Apache+mod_gridsite or gridftp)

• Move to batch submission via Grid tools in order to unify CPU accounting with LHC

• Setup the Grid utility “reliable File Transfer Service” to automate file distribution between MAGIC Datacenter @ PIC and sites which regularly subscribe many datasets

• PIC/IFAE will have specific resources to help with this, partially thanks to funding from the EGEE-III project

• Integrate into the procedure for opening an account in the Datacenter the additional steps for a user to get a Grid certificate and to be included as member of the MAGIC VO.

Monte Carlo simulation

• The recorded data are mainly background events due to charged cosmic rays (CR)

• Background rejection needs large samples of Monte Carlo simulated -ray and CR showers

• Very CPU consuming (1 night of background > 106 computer days )

• Access to simulated samples, MC production coordination, scalability (MAGIC II, CTA...)

• GRID can help with these issues

Status MC production

• Last data challenge (from September 2005) produced ~15000 simulated -ray showers, ~4% failures

• After that H. Kornmayer left the project, has been stalled since.

• New crew taking over the VO (UCM Madrid + INSA + Dortmund).

• Plan to start producing MC for MAGIC-II soon

Virtual observatory

Summary

• MAGIC scientific program requires large computing power and storage capacity

• Data center at PIC/IFAE (Barcelona) is up and start official operation in September 2008 with MAGIC-II

• Massive MC production for MAGIC-II will involve GRID

• (Some) Data will be release through virtual observatory

• Good benchmark for other astro-particle present and future projects