SNAP 6.0 very slow saving sentinel 2 to BEAM-DIMAP

downloaded a sentinel 2 file of Lourdes, France (S2B_MSIL1C_20190613T105629_N0207_R094_T30TYN_20190613T120957 approx 300mb).

when I do some remote sensing tasks in Erdas the task is completed in a few seconds o minutes. When in SNAP, it takes almost an hour only the mandatory resampling of the file. When later saving a subset or clipped file it takes me again a large amount of time.

What i want to do with SNAP:

  1. resamsple the sentinel 2 file to 10m pixels (needed for creating a subset)
  2. create a subset
  3. create NDVI image of subset

Computer specs:
win 10 64, 8gb ram
SNAP 6.0
Erdas 2018 V16.5


  1. is it normal that SNAP takes its time time?
  2. SNAP uses 3000mb ram is that enough?
  3. does someone know how to speed up things?

I want to transfer to SNAP when using sentinel 2 imagery and let go Erdas, but only when things can be done quicker.


Similar topic is discussed here,

Source of the thread

The RAM 8 GB should be enough for some processing, but simultaneously is not enough to other processing and it should be increased, I don’t know about ERDAS at the moment, to compare, But still you could increase the virtual SNAP RAM as it’s mentioned in the thread I shared,

Do you mean the cache size?

I meant the following,

increasing the SNAP RAM ,For the SNAP Desktop application, you can increase the amount of memory available to SNAP.

In the ‘etc’ folder of the SNAP installation directory, you’ll find a file named snap.conf. Open it in a text editor.

There is the line which starts with ‘default_options=’

In this line you’ll find an option like -J-Xmx5G . Increase the value. You could use something like -J-Xmx13G , if you have enough memory in your computer. By default, it is set to ~75% of the maximum value. This is usually a good choice.

Also you could set the cache size ,

Hi @bearhunt,
only some comments that perhaps will help you:

  • In the first step of resampling, do not save the data to the disk. When doing that, SNAP needs to compute all the data and you do not need it, since you are going to apply a subset later. If you don’t save to disk, the computation will be done only when the data is needed, so it will be very fast.
  • When you apply the resampling all the bands are resampled, including the angle bands (they are around 30). Initially these bands are 22x22 pixels, so if you want to write them, it is very fast, but when resampling you have 30 new bands of 10980x10980 pixels, which takes a lot of time to be saved.
  • When applying the subset, try to keep only the bands you need.

Thank you for clarification.

@ obarrilero yes, i’m working with a subset of a few square km’s.

can i save the subset quickly