sumbur 0.0.2 → 0.1.2

data/ext/sumbur/sumbur.c

@@ -1,6 +1,22 @@
 #include "ruby.h"
 
 #define L 0xFFFFFFFF
+
+static unsigned int L27_38[] = {L / 27, L / 28, L / 29, L / 30, L / 31, L / 32,
+                               L / 33, L / 34, L / 35, L / 36, L / 37, L / 38,
+                               L / 39, L / 40, L / 41, L / 42, L / 43, L / 44,
+                               L / 45, L / 46, L / 47, L / 48, L / 49, L / 50,
+                               L / 51, L / 52, L / 53, L / 54, L / 55, L / 56,
+                               L / 57, L / 58, L / 59, L / 60, L / 61, L / 62
+                               };
+static unsigned int LL27_38[] = {L/(26*27), L/(27*28), L/(28*29), L/(29*30), L/(30*31), L/(31*32),
+                                 L/(32*33), L/(33*34), L/(34*35), L/(35*36), L/(36*37), L/(37*38),
+                                 L/(38*39), L/(39*40), L/(40*41), L/(41*42), L/(42*43), L/(43*44),
+                                 L/(44*45), L/(45*46), L/(46*47), L/(47*48), L/(48*49), L/(49*50),
+                                 L/(50*51), L/(51*52), L/(52*53), L/(53*54), L/(54*55), L/(55*56),
+                                 L/(56*57), L/(57*58), L/(58*59), L/(59*60), L/(60*61), L/(61*62)
+                               };
+
 static VALUE
 rb_sumbur(VALUE self, VALUE hashed_int, VALUE capacity)
 {
@@ -8,13 +24,13 @@ rb_sumbur(VALUE self, VALUE hashed_int, VALUE capacity)
     unsigned int capa = NUM2UINT(capacity);
     unsigned int part, n, i, c;
 
-    if (capacity == 0) {
+    if (capa == 0) {
         rb_raise(rb_eArgError, "Sumbur is not applicable to empty cluster");
     }
 
     part = L / capa;
 
-    if (L - h <= part) return INT2FIX(0);
+    if (L - h < part) return INT2FIX(0);
 
     n = 1;
 
@@ -22,9 +38,9 @@ rb_sumbur(VALUE self, VALUE hashed_int, VALUE capacity)
         if (h >= L / 2) h -= L / 2;
         else {
             n = 2;
-            if (L / 2 - h < part) break;
+            if (L / 2 - h < part) return INT2FIX(1);
         }
-        if (capa == 2) break;
+        if (capa == 2) return INT2FIX(1);
 
 #define curslice(i) (L / (i * (i - 1)))
 #define unroll(i) \
@@ -32,17 +48,32 @@ rb_sumbur(VALUE self, VALUE hashed_int, VALUE capacity)
         else {                                  \
             h += curslice(i) * (i - n - 1);     \
             n = i;                              \
-            if (L / i - h < part) break;        \
+            if (L / i - h < part) return INT2FIX(n-1);        \
         }                                       \
-        if (capa == i) break
+        if (capa == i) return INT2FIX(n-1)
+
+        unroll(3); unroll(4); unroll(5);
+        unroll(6); unroll(7); unroll(8);
+        unroll(9); unroll(10); unroll(11);
+        unroll(12); unroll(13); unroll(14);
+        unroll(15); unroll(16); unroll(17);
+        unroll(18); unroll(19); unroll(20);
+        unroll(21); unroll(22); unroll(23);
+        unroll(24); unroll(25); unroll(26);
 
-        unroll(3); unroll(4); unroll(5); unroll(6); unroll(7);
-        unroll(8); unroll(9); unroll(10); unroll(11); unroll(12);
-        unroll(13); unroll(14); unroll(15); unroll(16); unroll(17);
-        unroll(18); unroll(19); unroll(20); unroll(21); unroll(22);
-        unroll(23); unroll(24);
+        for (i = 27; i <= capa && i <= 62; i++) {
+            c = LL27_38[i-27];
+            if (c <= h) {
+                h -= c;
+            }
+            else {
+                h += c * (i - n - 1);
+                n = i;
+                if (L27_38[i-27] - h < part) return INT2FIX(n-1);
+            }
+        }
 
-        for(i = 25; i <= capa; i++) {
+        for(i = 63; i <= capa; i++) {
             c = L / (i * (i - 1));
             if (c <= h) {
                 h -= c;
@@ -50,7 +81,7 @@ rb_sumbur(VALUE self, VALUE hashed_int, VALUE capacity)
             else {
                 h += c * (i - n - 1);
                 n = i;
-                if (L / i - h < part) break;
+                if (L / i - h < part) return INT2FIX(n - 1);
             }
         }
     } while(0);

data/lib/sumbur/pure_ruby.rb

@@ -9,7 +9,7 @@ module Sumbur
       l = 0xFFFFFFFF
       part = l / cluster_capacity
 
-      return 0  if l - hashed_integer <= part
+      return 0  if l - hashed_integer < part
 
       h = hashed_integer
       n = 1

data/lib/sumbur/version.rb

@@ -1,3 +1,3 @@
 module Sumbur
-  VERSION = "0.0.2"
+  VERSION = "0.1.2"
 end

data/test/test_sumbur.rb

@@ -3,11 +3,18 @@ require 'minitest/autorun'
 
 $spread_cache = {}
 
-HASHED = (1..1_000_000).map{|i| [i.hash & 0xFFFFFFFF, i]}
+HASHED = (1..1_000_000).map{|i|
+  h = i ^ (i >> 16)
+  h = (h * 0x85ebca6b) & 0xFFffFFff
+  h ^= h >> 13
+  h = (h * 0xc2b2ae35) & 0xFFffFFff
+  [h ^ (h >> 16), i]
+}
 
 def spread(num, capa, sumbur)
   start = Time.now
-  v = $spread_cache[ [num, capa, sumbur] ] ||= HASHED[0, num].
+  #v = $spread_cache[ [num, capa, sumbur] ] ||=
+  v = HASHED[0, num].
       map{|hash, int| [sumbur.sumbur(hash, capa), int]}.
       group_by{|serv, int| serv}
   d = Time.now - start
@@ -17,17 +24,17 @@ end
 
 shared_example = proc do
   it "should spread capacity" do
-    for num in [100_000, 1_000_000]
+    for num, eps in [[100_000, 0.05], [1_000_000, 0.011]]
       for capa in [2,3,7,8,17,18]
         spread(num, capa, sumbur).each{|serv, ints|
-          ints.size.must_be_within_epsilon num/capa, 7000.0/num
+          ints.size.must_be_within_epsilon num/capa, eps
         }
       end
     end
   end
 
   it "should reshard values cleanly" do
-    for num in [100_000, 1_000_000]
+    for num, eps in [[100_000, 0.2], [1_000_000, 0.05]]
       for capa in [2,3,7,8,17,18]
         cur = spread(num, capa, sumbur)
         nxt = spread(num, capa+1, sumbur)
@@ -35,9 +42,9 @@ shared_example = proc do
         for i in 0...capa
           (nxt[i] - cur[i]).must_be_empty
           moved += mvd = (cur[i] - nxt[i]).size
-          mvd.must_be_within_epsilon (num/capa - num/(capa+1)), 40000.0/num
+          mvd.must_be_within_epsilon (num/capa - num/(capa+1)), eps
         end
-        moved.must_be_within_epsilon num/(capa+1), 7000.0/num
+        moved.must_be_within_epsilon num/(capa+1), eps
       end
     end
   end
@@ -45,20 +52,32 @@ end
 
 require 'sumbur/pure_ruby'
 describe "Pure Ruby" do
-  let(:sumbur){ Sumbur::PureRuby }
+  def sumbur
+    Sumbur::PureRuby
+  end
   class_exec &shared_example
 end
 
 begin
-  require 'sumbur/native_sumbur'
+  if RUBY_ENGINE == 'jruby'
+    require 'sumbur/sumbur.jar'
+  else
+    require 'sumbur/native_sumbur'
+  end
 
   describe "Native" do
-    let(:sumbur) { Sumbur::NativeSumbur }
+    def sumbur
+      if RUBY_ENGINE == 'jruby'
+        Sumbur::Java
+      else
+        Sumbur::NativeSumbur
+      end
+    end
     class_exec &shared_example
 
     it "should produce same spread as pure ruby version" do
-      for capa in [2,3,4,7,8,9,17,18,19]
-        spread(1_000_000, capa, Sumbur::NativeSumbur).must_equal spread(1_000_000, capa, Sumbur::PureRuby)
+      for capa in [2,3,4,7,8,9,17,18,19,100]
+        spread(1_000_000, capa, sumbur).must_equal spread(1_000_000, capa, Sumbur::PureRuby)
       end
     end
   end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: sumbur
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.1.2
   prerelease: 
 platform: ruby
 authors:
@@ -10,9 +10,9 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-08-03 00:00:00.000000000 Z
+date: 2012-08-10 00:00:00.000000000 Z
 dependencies: []
-description: Sumbur - consistent spreading
+description: Sumbur - consistent spreading for server balancing
 email:
 - funny.falcon@gmail.com
 - uint32@mail.ru
@@ -51,7 +51,6 @@ rubyforge_project:
 rubygems_version: 1.8.24
 signing_key: 
 specification_version: 3
-summary: Sumbur - consistent spreading
+summary: Sumbur - consistent spreading for server balancing
 test_files:
 - test/test_sumbur.rb
-has_rdoc: