Speed up Sen2Cor batch processing

Optical Toolbox sen2cor
#1

Hi,
I’m using Sen2Cor thru command line “L2A_Process” . Since I want to process 45 Sentinel-2 products I’m using that process in a for loop (UNIX code for linux):

for file in ./*.SAFE; do L2A_Process --resolution=10 file; done

this code are working fine, but seem it does not use al the machine capability. the CPU usage are less than 20% and the RAM occupied is 20 Gb with 40 Gb free.

Since the processing time is very long, I’m asking if there is a way to speed up the process using more machine hardware.

Thank you all,
Best.

Missing image file after conversion from L1C to L2A
Best practice in acquiring and atmospherically correcting Sentinel-2 data
How to save the command prompt code after processing using sen2cor
#2

I do not know the particulars of Sen2Cor but large EO raster-data processing is very often I/O-limited, therefore SNAP on SSD (or RAMDISK) is way faster than on HDD.

#3

try for file in ./*.SAFE; do L2A_Process --resolution=10 file & ; done

#4

it doesn’t work at all.
error: unexpected ;

#5

If I were you I would script the processing using Python.
Python also has a good multiprocessing libary.

#6

Hi,

Do you mind sharing your UNIX code for batch processing with Sen2Cor? I also want to batch processing L1C files for atmospheric correction in Linux.

Thank you!
Bests,
Han