Basic rules
- always check the location where you read/write files and space availability. See df command.
- AVOID using nfs file systems/mounts for your simulations, especially when running on multiple nodes.
This will double sometimes your execution time since for every read/write operation an extra IOwait time is needed to transfer information over the network.
Use a location mounted on a local disk, usually /export/work, /export-b/work/ or /scratch (on HPC nodes) and move your data after finishing in a desired location.
To learn
Commands
basic commands !!!
df (Check Disk Space) especially usefull options:
show mounts including Type: df -T
show mount location and type of current working directory: df -T .du (disk usage) and especially useful option:
show total size, in human readable format, of current working forlder: du -hs- ncdu if it's installed, the simplest way to find out what and much space it occupies. RECOMMENDED instead of du
- ssh command
Local Linux Cluster
- info about the cluster status using ganglia (works only internally). If you choose then "cluster linux" , you see in "choose a node" the available computers.
- from outside, you have first to connect via ssh to login.stud.ntnu.no (or login.ansatt.ntnu.no). From there you can connect in the next step to a local computer.
- use nice/renice to lower the priority of your processes
- you can ran processes in parallel using MPI. Multithread libraries in C++, such as pthread, are also available.
...