lkml.org 
[lkml]   [2022]   [May]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 3/4] selftests: memcg: Adjust expected reclaim values of protected cgroups
On Fri, May 13, 2022 at 07:18:10PM +0200, Michal Koutny wrote:
> The numbers are not easy to derive in a closed form (certainly mere
> protections ratios do not apply), therefore use a simulation to obtain
> expected numbers.
>
> The new values make the protection tests succeed more precisely.
>
> % run as: octave-cli script
> %
> % Input configurations
> % -------------------
> % E parent effective protection
> % n nominal protection of siblings set at the givel level
> % c current consumption -,,-
>
> % example from testcase (values in GB)
> E = 50 / 1024;
> n = [75 25 0 500 ] / 1024;
> c = [50 50 50 0] / 1024;
>
> % Reclaim parameters
> % ------------------
>
> % Minimal reclaim amount (GB)
> cluster = 32*4 / 2**20;
>
> % Reclaim coefficient (think as 0.5^sc->priority)
> alpha = .1
>
> % Simulation parameters
> % ---------------------
> epsilon = 1e-7;
> timeout = 1000;
>
> % Simulation loop
> % ---------------------
> % Simulation assumes siblings consumed the initial amount of memory (w/out
> % reclaim) and then the reclaim starts, all memory is reclaimable, i.e. treated
> % same. It simulates only non-low reclaim and assumes all memory.min = 0.
>
> ch = [];
> eh = [];
> rh = [];
>
> for t = 1:timeout
> % low_usage
> u = min(c, n);
> siblings = sum(u);
>
> % effective_protection()
> protected = min(n, c); % start with nominal
> e = protected * min(1, E / siblings); % normalize overcommit
>
> % recursive protection
> unclaimed = max(0, E - siblings);
> parent_overuse = sum(c) - siblings;
> if (unclaimed > 0 && parent_overuse > 0)
> overuse = max(0, c - protected);
> e += unclaimed * (overuse / parent_overuse);
> endif
>
> % get_scan_count()
> r = alpha * c; % assume all memory is in a single LRU list
>
> % commit 1bc63fb1272b ("mm, memcg: make scan aggression always exclude protection")
> sz = max(e, c);
> r .*= (1 - (e+epsilon) ./ (sz+epsilon));
>
> % uncomment to debug prints
> % e, c, r
>
> % nothing to reclaim, reached equilibrium
> if max(r) < epsilon
> break;
> endif
>
> % SWAP_CLUSTER_MAX
> r = max(r, (r > epsilon) .* cluster);
> % XXX here I do parallel reclaim of all siblings
> % in reality reclaim is serialized and each sibling recalculates own residual
> c = max(c - r, 0);
>
> ch = [ch ; c];
> eh = [eh ; e];
> rh = [rh ; r];
> endfor
>
> t
> c, e

This is a cool stuff!

How about to place it into a separate file and add a comment into the code
with a reference?

Thanks!

\
 
 \ /
  Last update: 2022-05-13 20:54    [W:0.140 / U:0.080 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site