LOFAR will have an extensive System Health Management function to make the instrument self-diagnosing and (where possible) self-healing.

LOFAR produces very large data streams, especially for the astronomy application. One month of observing in this mode results in PetaBytes of data. Systematic long-term storage for such data volumes thus becomes extremely expensive.

Therefore LOFAR has adopted a processing model in which final data products for routine observations are formed in a highly automated mode with user interaction at well-defined moments. Only final data products are exported. Currently, a LOFAR archive architecture is defined and constructed. Of course it will only be possible to operate the instrument this way when the conditions for “routine” observations are available.

LOFAR post-processing can take place either at the Central Processor or locally with the users. If the available Internet capacity is sufficient, intermediate dataproducts can be transported to the user, and local processing can be done. Otherwise processing resources at the Central Processor are available for further data reduction (within the limits of the Central Processor processing budget). In the early years of operation all data processing and analysis will be highly interactive, requiring more intensive access to lower-level dataproducts. It is very well possible to export selected observations to another site, even on a regular basis. It should be kept in mind that the receiving party should have access to processing resources that are matched to the incoming data-rate.