on a 448 thread and 6TB RAM, Ubuntu 22.04 machine.
My dataset.txt contains 21273 continuous variables and 3500 samples.
The installation of causal-cmd was done through maven.
However, I noticed that only 3 TB of RAM and approximately 1/3 of the threads are being used every moment.
I was wondering if there is an issue with the java version that I am currently using (OpenJDK Runtime Environment (build 21.0.2+13-Ubuntu-122.04.1)) not being able to use more resources that what there are available, or if there is anything I can do (java configurations), to make sure that the total resources of my machine are utilised by causal-cmd.
I would be indebted if you could provide me with a better command line arguments for what I am doing.
Hello,
I am using causal-cmd for a causal inference study, with the following command:
java -Xmx5000G -jar causal-cmd-1.13.0-SNAPSHOT-jar-with-dependencies.jar --data-type continuous --delimiter tab --parallelized yes --algorithm fges --dataset dataset.txt --score sem-bic-score --penaltyDiscount 32.0 --maxDegree 3 --verbose yes
on a 448 thread and 6TB RAM, Ubuntu 22.04 machine.
My dataset.txt contains 21273 continuous variables and 3500 samples.
The installation of causal-cmd was done through maven.
However, I noticed that only 3 TB of RAM and approximately 1/3 of the threads are being used every moment. I was wondering if there is an issue with the java version that I am currently using (OpenJDK Runtime Environment (build 21.0.2+13-Ubuntu-122.04.1)) not being able to use more resources that what there are available, or if there is anything I can do (java configurations), to make sure that the total resources of my machine are utilised by causal-cmd.
I would be indebted if you could provide me with a better command line arguments for what I am doing.
Thank you in advance, Bill