Overview
- The GIOD (Globally Interconnected Object Databases) joint project is addressing the data storage and access problems posed by particle collider experiments at CERN in 2005.
- 100 MBytes/sec from online systems => several PetaBytes per year.
- Raw data will be reconstructed to particle tracks, energy clusters, etc. in near-real time by large processor farms of ~10^7 MIPS.
- Reconstructed data will be stored in ODBMS.
- Need to make available object data from around 10^9 particle collisions.
- Replicate significant fractions of the ODBMS amongst world-wide "regional centres”.