9 Jul
2015
9 Jul
'15
12:50 a.m.
Hi Thomas,
We are getting ready to run two years worth of correlations and we are wondering how you
can achieve faster cc computations. Could you advise us if this is possible? Does the
"-t" command do this? We have tried it, but we see no appreciable gain, so
perhaps we are using it incorrectly with "msnoise compute_cc".
Thank you in advance.
Cheers,
Hunter
Hunter A. Knox, PhD
Geophysics and Atmospheric Sciences
MS 0750
Sandia National Laboratories
1515 Eubank SE
Albuquerque, NM, 87123
(O) 505-844-6724
9 Jul
9 Jul
3:46 p.m.
Hi Hunter,
No, sorry, the -t option only works with the scan_archive procedure
currently. The only way to "parallelize" the CC is to run multiple
instances of the code. Note: if the number of stations is large & you do
only ZZ component, you might fill the RAM pretty fast. This is because
of the "faster mode" precalculating the whitened slices. I should remove
this mode on the next version, it's not very useful anyway...
Let me know how it goes
Cheers
Thomas
Le 09/07/2015 00:50, Knox, Hunter Anne a écrit :
Hi Thomas,
We are getting ready to run two years worth of correlations and we are wondering how you
can achieve faster cc computations. Could you advise us if this is possible? Does the
"-t" command do this? We have tried it, but we see no appreciable gain, so
perhaps we are using it incorrectly with "msnoise compute_cc".
Thank you in advance.
Cheers,
Hunter
Hunter A. Knox, PhD
Geophysics and Atmospheric Sciences
MS 0750
Sandia National Laboratories
1515 Eubank SE
Albuquerque, NM, 87123
(O) 505-844-6724
_______________________________________________
MSNoise mailing list
MSNoise(a)mailman-as.oma.be
http://mailman-as.oma.be/mailman/listinfo/msnoise
3586
days inactive
3587
days old
1 comments
2 participants
participants (2)
-
Knox, Hunter Anne -
Thomas Lecocq