[OpenSPIM] Deconvolution on cluster

[Marquart, Gregory (NIH/NICHD) [F]]

We have been trying to run deconvolution on a cluster at the NIH without much luck.  Although we have run the process on a 256GB node and Fiji says the deconvolution should only take ~80GB, Fiji still crashes.  How accurate are these memory estimates?  Is it possible that we are still running out of memory?  Or is it more likely that our settings just aren't right?  Any help or suggestions would be much appreciated.

Cheers,
Graham

Gregory (Graham) Marquart
Unit on Behavioral Neurogenetics, NICHD
Building 6B, 3B-308
6 Center Drive, Bethesda, MD 20892