Use statistical approach to approximate the max number of nodes per population

[userspace-rcu.git] / rcuja / design.txt

diff --git a/rcuja/design.txt b/rcuja/design.txt
index 624eb1872b78df8f616a1e77a8e52b0e75041b3c..89e66b935467f7c5853f25408cb4b897a6657899 100644 (file)
--- a/rcuja/design.txt
+++ b/rcuja/design.txt
@@ -151,8 +151,8 @@ difference -- for millions of runs).
 
 tot entries   unbalance       largest linear array (stat. approx.)
 ---------------------------------------------------------------------
-41 entries:       9                   20.5+4.5=25 (target ~50/2=25)
-47 entries:       9                   23.5+4.5=28 (target ~56/2=28)
+48 entries:      2 (98%)              24+1=25 (target ~50/2=25)
+54 entries:      2 (97%)              27+1=28 (target ~56/2=28)
 
 Note: there exists rare worse cases where the unbalance is larger, but
 it happens _very_ rarely. But need to provide a fallback if the subclass
@@ -164,8 +164,8 @@ we can get for choice of distributions grouped by pairs of bits.
 
 tot entries     unbalance     largest linear array (stat. approx.)
 ---------------------------------------------------------------------
-80 entries:        20              20+5=25  (target: ~100/4=25)
-90 entries:        22              22.5+5.5=28  (target: ~112/4=28)
+92 entries:      8 (99%)             23+2=25  (target: ~100/4=25)
+104 entries:     8 (99%)             26+2=28  (target: ~112/4=28)
 
 
 Note: there exists rare worse cases where the unbalance is larger, but