Optimize gpt for production use.


We are using gpt on console to process radar images, this is how we run it:

gpt -x -q 8 -c 5G graph.xml -Pinfile=radarProductPath -Poutfile=outPathDir -Ppolygon=bbox

Its working ok, but the problem is that we can only have one or two instances running at the same time.

I think one of the problems is that one gpt command starts a lot of processes, and I think its because the -q flag is not working, I saw that this bug still open (https://senbox.atlassian.net/browse/SNAP-632)

We have 3 machines on creodias:
1 Machine: eo2.xlarge 4vCores RAM 16GB local disc SSD 64GB
2 Machines: eohm.2xlarge 16vCores RAM 128GB local disc SSD 384GB

So my question is, how can we optimize the gpt process so I can process more radar images at the same time, we already are using Area of interest, but we are not being able to run more than one or two processes at same time.