But that only runs the first time, and gives me the following error the rest of the time :
Error occurred during initialization of VM
Could not allocate metaspace: 268435456 bytes
From what I can tell, the problem is caused by “ulimit -v”, but I don’t have privileges to edit this.
39m heapsize and 64m class size seem very small, is there any way that I can get more memory? The project I’m attempting to run is a fairly expensive (and long-running) job for timeseries classification.
As a side note, is it possible fore me to avoid having to type out the long command every time?
The issue is likely that you are running your application on the development login node. This system is only for code development, and workloads should go to the compute nodes.
Please use the Torque queue to get your java workload on to one of the compute nodes; they will have the memory space you need.
The access page that you download the key from will have instructions on how to use the Torque queue.
As for typing the command, you can press up arrow key to go back in your bash history. Though in this case, I recommend writing a simple bash script.