On 6/9/23 5:07 PM, Eric Dumazet wrote:
On Fri, Jun 9, 2023 at 10:28 AM Abel Wu <wuyun.abel@xxxxxxxxxxxxx> wrote:
This is just a PoC patch intended to resume the discussion about
tcpmem isolation opened by Google in LPC'22 [1].
We are facing the same problem that the global shared threshold can
cause isolation issues. Low priority jobs can hog TCP memory and
adversely impact higher priority jobs. What's worse is that these
low priority jobs usually have smaller cpu weights leading to poor
ability to consume rx data.
To tackle this problem, an interface for non-root cgroup memory
controller named 'socket.urgent' is proposed. It determines whether
the sockets of this cgroup and its descendants can escape from the
constrains or not under global socket memory pressure.
The 'urgent' semantics will not take effect under memcg pressure in
order to protect against worse memstalls, thus will be the same as
before without this patch.
This proposal doesn't remove protocal's threshold as we found it
useful in restraining memory defragment. As aforementioned the low
priority jobs can hog lots of memory, which is unreclaimable and
unmovable, for some time due to small cpu weight.
So in practice we allow high priority jobs with net-memcg accounting
enabled to escape the global constrains if the net-memcg itselt is
not under pressure. While for lower priority jobs, the budget will
be tightened as the memory usage of 'urgent' jobs increases. In this
way we can finally achieve:
- Important jobs won't be priority inversed by the background
jobs in terms of socket memory pressure/limit.
- Global constrains are still effective, but only on non-urgent
jobs, useful for admins on policy decision on defrag.
Comments/Ideas are welcomed, thanks!
This seems to go in a complete opposite direction than memcg promises.
Can we fix memcg, so that :
Each group can use the memory it was provisioned (this includes TCP buffers)
Yes, but might not be easy once memory gets over-committed (which is
common in modern data-centers). So as a tradeoff, we intend to put
harder constraint on memory allocation for low priority jobs. Or else
if every job can use its provisioned memory, than there will be more
memstalls blocking random jobs which could be the important ones.
Either way hurts performance, but the difference is whose performance
gets hurt.
Memory protection (memory.{min,low}) helps the important jobs less
affected by memstalls. But once low priority jobs use lots of kernel
memory like sockmem, the protection might become much less efficient.
Global tcp_memory can disappear (set tcp_mem to infinity)
