On Wed, 2024-02-14 at 12:30 -0800, Suren Baghdasaryan wrote:

On Wed, Feb 14, 2024 at 12:17 PM Yosry Ahmed [off-list ref] wrote:

quoted

quoted

quoted

quoted

Performance overhead:
To evaluate performance we implemented an in-kernel test executing
multiple get_free_page/free_page and kmalloc/kfree calls with allocation
sizes growing from 8 to 240 bytes with CPU frequency set to max and CPU
affinity set to a specific CPU to minimize the noise. Below are results
from running the test on Ubuntu 22.04.2 LTS with 6.8.0-rc1 kernel on
56 core Intel Xeon:

                        kmalloc                 pgalloc
(1 baseline)            6.764s                  16.902s
(2 default disabled)    6.793s (+0.43%)         17.007s (+0.62%)
(3 default enabled)     7.197s (+6.40%)         23.666s (+40.02%)
(4 runtime enabled)     7.405s (+9.48%)         23.901s (+41.41%)
(5 memcg)               13.388s (+97.94%)       48.460s (+186.71%)

(6 default disabled+memcg)    13.332s (+97.10%)         48.105s (+184.61%)
(7 default enabled+memcg)     13.446s (+98.78%)       54.963s (+225.18%)

I think these numbers are very interesting for folks that already use
memcg. Specifically, the difference between 6 & 7, which seems to be
~0.85% and ~14.25%. IIUC, this means that the extra overhead is
relatively much lower if someone is already using memcgs.

Well, yes, percentage-wise it's much lower. If you look at the
absolute difference between 6 & 7 vs 2 & 3, it's quite close.

quoted

quoted

(6) shows a bit better performance than (5) but it's probably noise. I
would expect them to be roughly the same. Hope this helps.

quoted

Thanks for the data.  It does show that turning on memcg does not cost
extra overhead percentage wise.

Tim