Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/vlbi/dokuwiki/lib/exe/fetch.php/difx/socorro2010/socorro_difx.pdf Дата изменения: Wed Nov 17 00:47:39 2010 Дата индексирования: Tue Apr 12 12:36:58 2016 Кодировка:

DiFX vs Mark IV: A Geodetic User`s Perspective

Alessandra Bertarini, IGG University of Bonn Helge Rottmann, MPIfR Bonn

Hier kann nochmal der Titel stehen


Alessandra's Background Mark IV Experience: 8 years DiFX

3 months
2

Learning Curve:

1 year
(starting from zero knownledge)

1 week
(after 8 years exp.)

Hier kann nochmal der Titel stehen


Geodesy Needs. Part I The International VLBI Service delivers a service and not research. Geodesy correlation needs to be fast & reliable. Fast -> Data reduction should happen within two weeks from the observation. DiFX pro: Correlation of a 24 hour 6 station geodetic experiment runs in ~ 12 hours instead of ~ 30 hours with the Mark IV. DiFx contra: In case of problems, Mark IV support comes from developer (fast), DiFX support comes from mailing list (slow).
Hier kann nochmal der Titel stehen 3


Geodesy Needs. Part II Reliable -> Every upgrade must be tested. DiFX contra: Upgrades happen more often for DiFX than for hardware correlators (Mark III to Mark IV ~ 10 years). Solution: A test experiment needs to be correlated and analyzed (up to EOP delivery) for every upgrade.
Hier kann nochmal der Titel stehen 4


Error Space DiFX: More opportunity of screwing up everything but happily correlate . Whatever real number given to a double array won't prevent a software code from running. Mark IV: usually it either correlates or it does not. Reality: Geodesy always uses the same mode (i.e. same v2d file) for correlation, so after learning on the first experiments, correlation should run smoothly.
Hier kann nochmal der Titel stehen 5


Know-How

The user needs to develop sysop skills too (not really required for the Mark IV).
6

The user needs to develop C++ skills (opening a program to check its functionality happens more often for DiFX than for the Mark IV). The user needs to develop radio interferometry knowledge (whilst for the Mark IV undergraduate physics knowledge is enough).
Hier kann nochmal der Titel stehen


Conclusion

Geodesy correlation was proven to work (different geodetic setups were succesfully correlated), but still requires some amount of work to become operational. Operational geodesy requires product support (present with the Mark IV) to resolve problems rapidly. Who is going to take up this role?

7

Hier kann nochmal der Titel stehen