Dear All,
Ermldet is only partially working in the docker.
It takes a tremendous amount of time before failing.
Does anyone know the fix?
Thank you in advance.
All the best,
Matteo
Ermldet in the Docker
-
- Posts: 24
- Joined: Wed Jun 23, 2021 11:46 am
Re: Ermldet in the Docker
Dear Matteo,
Could you be more specific? For example, which data set are you using? The different options you are using in ermldet and the error you get when the task fails.
Best wishes,
Miriam
Could you be more specific? For example, which data set are you using? The different options you are using in ermldet and the error you get when the task fails.
Best wishes,
Miriam
- AlainGueguen
- Posts: 36
- Joined: Mon Jun 28, 2021 12:06 pm
Re: Ermldet in the Docker
Dear Matteo,
Ermldet uses a large amount of memory, and if the memory allocated to the docker is not enough then ermldet can be killed by the system itself when running out of memory .
my first suggestion would be to increase ( as much as possible) the memory allocated to the Docker engine ( in the preferences of the docker server itself)
also to reduce the memory usage of ermldet , i would suggest to launch it only on one energy band
if you need the result for more than one energy band i would suggest to launch one instance per energy band.
let us know if this help .
regards
Alain
Ermldet uses a large amount of memory, and if the memory allocated to the docker is not enough then ermldet can be killed by the system itself when running out of memory .
my first suggestion would be to increase ( as much as possible) the memory allocated to the Docker engine ( in the preferences of the docker server itself)
also to reduce the memory usage of ermldet , i would suggest to launch it only on one energy band
if you need the result for more than one energy band i would suggest to launch one instance per energy band.
let us know if this help .
regards
Alain
Re: Ermldet in the Docker
Dear Miriam, dear Alain,
Thank you for your replies. Indeed running it on a single energy band works and also allocating more memory helps.
When trying to run the WHOLE demo script though, it looks like it gets stuck in one of the multiple iterations (macOS - DOCKER).
This is not really an issue for my work though, I just wanted to make you aware about it.
Thanks again for he support.
All the best,
Matteo
Thank you for your replies. Indeed running it on a single energy band works and also allocating more memory helps.
When trying to run the WHOLE demo script though, it looks like it gets stuck in one of the multiple iterations (macOS - DOCKER).
This is not really an issue for my work though, I just wanted to make you aware about it.
Thanks again for he support.
All the best,
Matteo