translated by Google

Machine-translated page for increased accessibility for English questioners.

Operation of calculations

General information

This text regulates the conditions under which it is possible to run processes that are more demanding on system resources (memory, CPU, disk operations, etc.) or are of a longer-term character, etc. on generally available shared faculty machines.

When running such processes, it is generally important to ensure that you do not have exclusive access to the faculty machines and that other users can also use them. Therefore, it is necessary to run such processes considerably, especially in order not to restrict interactive users. This can be achieved by following the instructions below.

For larger scale calculations (duration in days, more machines, CPU intensive) it is also necessary to report them with a brief explanation and duration to unix@fi.muni.cz .

You do not need to follow these instructions for unrepeated short-term processes or for interactive debugging of your programs. Also running calculations on Aura it works in a different mode and not all of the restrictions described here apply.

The administrators of faculty machines continuously monitor the use of machines. In case of notification to administrators it is necessary to reduce the generated load. Processes that are not reported or overload the machine (s) may be terminated without notice. Also, in the event of repeated or severe non-compliance with the instructions here, administrators may temporarily block an account.

Restrictive measures

It is undesirable to run calculations at the time of teaching or potential examinations, generally between 07: 45-20: 15, on machines in classrooms that are directly used in teaching (Nymfe23-74 and sometimes Luna01-12). In general, however, it is best to run calculations only at night, if possible.

If the calculations are of a longer-term nature and are part of the research, respectively. thesis, it is advisable to use the Aura computing server for this purpose. If you do not have access to it, you can request a grant at justified reasons unix@fi.muni.cz .

For CPU-intensive processes it is necessary to set them the lowest priority +19 by eg command nice :

nice -n +19 ./program

It is also possible to restrict the process threads to specified CPU cores by command taskset :

taskset -c 1-2,4 ./program

If the running process generates higher I / O load (random access, large transfers) by disk operations, it is a good idea to place files with such access in a folder on the local disk ( /tmp ), not the home directory ( /home/xlogin ), since it is available over the network over NFS and the volume on which the data resides is also used by other machines.

You can use the command to limit the amount of memory used by processes running from a single shell instance when running more memory-intensive calculations ulimit -v to prevent exhaustion or excessive memory usage. You can also set a limit for the consumed process time using the command ulimit -t ; if the limit is exceeded, the process will be terminated by a KILL signal. For more information about it and its other options, see the command help ulimit or system call documentation setrlimit .

Load monitoring

During the calculations, it is advisable to continuously monitor the status, respectively. machine utilization (disk full, memory, I / O, load, state of processes), for example the following tools:

  • top, htop - summary information about the system and its processes
  • ps ux - listing information about my processes
  • iostat - listing information about I / O operations
  • who - list of logged in users (relevant only for classroom and hall machines)
  • df - utilizing disk capacity