HOME

HOME → LOFAR → 3.2 Data processing

3.2 Data Processing

1. Introduction	LOFAR will produce large data streams, especially for radio astronomical applications (e.g. 6 TeraByte of raw visibility data for 8 beams, a 4 hour synthesis observation, after integration for 1 sec and over a 10 KHz bandwidth). One month of observations in this mode results into a PetaByte of data. Systematic long-term storage for such data volumes thus would become too expensive. Therefore LOFAR has adopted a processing model in which final data products for routine observations are formed in a highly automated and almost totally independent mode with user interaction only at well-defined moments. In this way only final data products of elaborations are exported. Archiving is in principle the responsibility of the user or the Science Center. This model has been worked out in considerable detail for the radio astronomical applications; on the contrary the data rates and processing needs for the other applications are currently being defined.
2. The Core
3.Technical Details
3.1 The System
3.2 Data processing
3.3 Antennas spec.
4.Lofar at Medicina
4.1 Baseline extension
4.2 Lofar Kit
4.3 Superstation
5. Science

LOFAR post-processing can take place either at the Central Processor or at the user home country (in particular at Science Centers). If the available network capacity matches the data output rate, such data can be transported to the user workplace and local processing can be done. Otherwise processing resources at the Central Processor will be available for further data processing (within the limits of the Central Processor processing capabilities).

The core of LOFAR Central Processor (CEP) is represented by IBM supercomputer, the BlueGene/L system, that in November of 2004 has achieved the machine elaboration peak speed of 70 TFlops/sec and it has been the fastest supercomputer in the world. Efforts will be done in order to obtain the 180/360 TFlops/sec.

Webmaster

G. Bianchi