Ubuntu – command or a serie of commands to make the computer use as much resources as possible

command lineresourcesUbuntu

I am looking for one or possibly more commands, or a combination of commands, to get my PC to use as much resources as possible.

I want to check how my computer behaves when subject to the maximum amount of data it can handle.

I've tried by running multiple programs such as browsers, graphic and system tools one by one and all together, I've also tried to download big files to monitor its network and storage performance but every time depending on what program(s) I run, I get different results in terms of RAM, CPU, I/O.

Often my whole system crashes and I have to reboot or struggle to close some programs. To monitor, I'm using different commands such as iostat, iotop, htop, vmstat, lsof, iftop, iptraf but also some other little programs. (I'm using ubuntu)

I would very much appreciate an answer that could list some way to exploit GPU, CPU and RAM and a way to write a file (even with zeros) in the quickest way possible to see how fast my computer can produce output and write it on the HDD.

Best Answer

You could probably use stress:

stress: tool to impose load on and stress test systems

If you wnat to stress memory you could use :

stress  --vm 2 --vm-bytes 512M --timeout 10s

to use 2 vm using both 512MB of ram for 10 seconds.
If you want to stress CPU add:

stress --cpu ## -t 10s

with ## equal to your number of core to simulate a 100% cpu usage on all core at the same time for 10 seconds.
and if you want to simulate IO use the option :

stress --io 4 -t 10s

It will add thread that do sync call to you disk but you could also write to your disk with this option:

stress --hdd 4 --hdd-bytes 256M

This would create 4 thread writing each 256 MB of data to your disk, this could of course adjusted to simulated either lots of small file write or huge file write.

This will need some adaptation form you if you wan to stress thing one by one or all together or for longer time like this:

  stress --hdd 4 --hdd-bytes 256M --io 4 --cpu ## --vm 2 --vm-bytes 512M -t 60s

As for the GPU you could use glmark2 which should be available in Ubuntu. It's a basic GPU benchmark you could run it forever to simulate a gpuload:

glmark2 --run-forever

Related Solutions

Ubuntu – How to open tty or execute some script when the shortcut Ctrl+Alt+F[1-6] doesn’t work and X11 is partially frozen but system is still working

GNOME cannot interfere with Ctrl-Alt-F[N], since this is caught by the kernel, which implements VT switching. I.e., those key events never reach userspace, which is why you/your DE/any DE cannot override them or use them for some other purpose.

So if you can't switch VTs, there is a system level issue. In my first comment I said this sounds like panicing/busy looping, which it does. However, having that last 8 hours is pretty weird -- my next thought is it is I/O failure. Since you are using ubuntu 10.10, I presume this is an older netbook and the problem is new, even though you are still running the same old software. You also mention the problem has happened before but perhaps has been getting more serious, which again points to (disk) I/O failure -- your harddrive is dying. This creates a very nasty, unavoidable snag if the root filesystem is involved.

Unfortunately, you won't be able to tell for sure until you can look at the system logs. Right now, even if you had sshd running, I suspect you would not be able to get in because, again, it's the whole system that's frozen.

You can try shutting down and rebooting (moral of story: save work frequently, and save to some backup medium -- cloud, usb stick, whatever -- often as well). If it really is disk failure, you may be lucky to get anything off the disk. If you can reboot, the first thing you should do is:

cd /var; tar -cjf log

Then save /var/log.tar.bz2 to something so you can look at the logs on another computer, because (once again) if it is disk failure, this one is not going to be much good until you replace it. You may be able to duct-tape the problem by running e2fsck -cc (see man e2fsck) on the partitions, but that requires mounting the disk without the system on it running (e.g., if you can boot a liveCD from the external CD-ROM).

Bash – Interrupt shell command line expansion

As of GNU bash, version 4.4.19(1)-release (x86_64-pc-linux-gnu) and I am not using a VM:

echo $(yes)

exists the shell and does not freeze the system, and:

ls /*/../*/../*/../*/../*/

returns

bash: /bin/ls: Argument list too long

But as a rule, when you dealing with something that could get all the resources of a system is better to set limits before running, if you know that a process could be a cpu hog you could start it with cpulimit or run renice.

If you want to limit the processes that are already started, you will have to do it one by one by PID, but you can have a batch script to do that like the one below:

#!/bin/bash
LIMIT_PIDS=$(pgrep tesseract)   # PIDs in queue replace tesseract with your name
echo $LIMIT_PIDS
for i in $LIMIT_PIDS
do
    cpulimit -p $i -l 10 -z &   # to 10 percent processes
done

In my case pypdfocr launches the greedy tesseract.

Also in some cases were your CPU is pretty good you can just use a renice like this:

watch -n5 'pidof tesseract | xargs -L1 sudo renice +19'

Best Answer

Related Solutions

Ubuntu – How to open tty or execute some script when the shortcut Ctrl+Alt+F[1-6] doesn’t work and X11 is partially frozen but system is still working

Bash – Interrupt shell command line expansion

Related Question