Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.mrao.cam.ac.uk/projects/skads/lofar/summaries/discussion.1.pdf Дата изменения: Wed Apr 8 01:09:43 2009 Дата индексирования: Tue Aug 18 10:27:52 2009 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п

MKSP Workshop Cambridge 2009 March 25/26 Discussion Part 1
Observation time: · MKSP can only observe a representative subsample of galaxies · Piggybacking needed wherever possible · MKSP members should also ask for open time · Task for commissioning: How much daytime can be used? Data processing and storage: · Processing time with Groningen cluster may exceed observation time, especially if thermal noise or high dynamic range or high resolution (international baselines) is requested · Other resources for data analysis should be investigated · Involvement of FZ JЭlich (1 PB storage and cluster processing) is considered and will be tested, data transfer between Groningen and JЭlich should not be a problem · JЭlich should be informed that the MKSP wishes to use its storage (Action Item to Enno Middelberg) · SKA project planning needs the LOFAR experience on data processing Data volumes: · 100h of observations gives a data cube of ~2PB; reduction by factor ~4 probably possible without major losses, but still too much · Further reducing of data volume needs smearing in time or frequency, or reduce field, or restrict to subfields task for commissioning observations, and task for the MKSP Working Groups to argue for the number of square degrees of sky they really need to accomplish their science · Averaging in frequency is not possible if RM Synthesis needs large RM ranges (± 1000-10000 rad/m2 ), but averaging in time is still possible.