1<html> 2<!-- 3 Licensed to the Apache Software Foundation (ASF) under one or more 4 contributor license agreements. See the NOTICE file distributed with 5 this work for additional information regarding copyright ownership. 6 The ASF licenses this file to You under the Apache License, Version 2.0 7 (the "License"); you may not use this file except in compliance with 8 the License. You may obtain a copy of the License at 9 10 http://www.apache.org/licenses/LICENSE-2.0 11 12 Unless required by applicable law or agreed to in writing, software 13 distributed under the License is distributed on an "AS IS" BASIS, 14 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 15 See the License for the specific language governing permissions and 16 limitations under the License. 17 --> 18 <!-- $Revision: 1054186 $ $Date: 2011-01-01 03:28:46 +0100 (sam. 01 janv. 2011) $ --> 19 <body> 20 <p>Random number and random data generators.</p> 21 <p>Commons-math provides a few pseudo random number generators. The top level interface is RandomGenerator. 22 It is implemented by three classes: 23 <ul> 24 <li>{@link org.apache.commons.math.random.JDKRandomGenerator JDKRandomGenerator} 25 that extends the JDK provided generator</li> 26 <li>AbstractRandomGenerator as a helper for users generators</li> 27 <li>BitStreamGenerator which is an abstract class for several generators and 28 which in turn is extended by: 29 <ul> 30 <li>{@link org.apache.commons.math.random.MersenneTwister MersenneTwister}</li> 31 <li>{@link org.apache.commons.math.random.Well512a Well512a}</li> 32 <li>{@link org.apache.commons.math.random.Well1024a Well1024a}</li> 33 <li>{@link org.apache.commons.math.random.Well19937a Well19937a}</li> 34 <li>{@link org.apache.commons.math.random.Well19937c Well19937c}</li> 35 <li>{@link org.apache.commons.math.random.Well44497a Well44497a}</li> 36 <li>{@link org.apache.commons.math.random.Well44497b Well44497b}</li> 37 </ul> 38 </li> 39 </ul> 40 </p> 41 42 <p> 43 The JDK provided generator is a simple one that can be used only for very simple needs. 44 The Mersenne Twister is a fast generator with very good properties well suited for 45 Monte-Carlo simulation. It is equidistributed for generating vectors up to dimension 623 46 and has a huge period: 2<sup>19937</sup> - 1 (which is a Mersenne prime). This generator 47 is described in a paper by Makoto Matsumoto and Takuji Nishimura in 1998: <a 48 href="http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/ARTICLES/mt.pdf">Mersenne Twister: 49 A 623-Dimensionally Equidistributed Uniform Pseudo-Random Number Generator</a>, ACM 50 Transactions on Modeling and Computer Simulation, Vol. 8, No. 1, January 1998, pp 3--30. 51 The WELL generators are a family of generators with period ranging from 2<sup>512</sup> - 1 52 to 2<sup>44497</sup> - 1 (this last one is also a Mersenne prime) with even better properties 53 than Mersenne Twister. These generators are described in a paper by François Panneton, 54 Pierre L'Ecuyer and Makoto Matsumoto <a 55 href="http://www.iro.umontreal.ca/~lecuyer/myftp/papers/wellrng.pdf">Improved Long-Period 56 Generators Based on Linear Recurrences Modulo 2</a> ACM Transactions on Mathematical Software, 57 32, 1 (2006). The errata for the paper are in <a 58 href="http://www.iro.umontreal.ca/~lecuyer/myftp/papers/wellrng-errata.txt">wellrng-errata.txt</a>. 59 </p> 60 61 <p> 62 For simple sampling, any of these generators is sufficient. For Monte-Carlo simulations the 63 JDK generator does not have any of the good mathematical properties of the other generators, 64 so it should be avoided. The Mersenne twister and WELL generators have equidistribution properties 65 proven according to their bits pool size which is directly linked to their period (all of them 66 have maximal period, i.e. a generator with size n pool has a period 2<sup>n</sup>-1). They also 67 have equidistribution properties for 32 bits blocks up to s/32 dimension where s is their pool size. 68 So WELL19937c for exemple is equidistributed up to dimension 623 (19937/32). This means a Monte-Carlo 69 simulation generating a vector of n variables at each iteration has some guarantees on the properties 70 of the vector as long as its dimension does not exceed the limit. However, since we use bits from two 71 successive 32 bits generated integers to create one double, this limit is smaller when the variables are 72 of type double. so for Monte-Carlo simulation where less the 16 doubles are generated at each round, 73 WELL1024 may be sufficient. If a larger number of doubles are needed a generator with a larger pool 74 would be useful. 75 </p> 76 77 <p> 78 The WELL generators are more modern then MersenneTwister (the paper describing than has been published 79 in 2006 instead of 1998) and fix some of its (few) drawbacks. If initialization array contains many 80 zero bits, MersenneTwister may take a very long time (several hundreds of thousands of iterations to 81 reach a steady state with a balanced number of zero and one in its bits pool). So the WELL generators 82 are better to <i>escape zeroland</i> as explained by the WELL generators creators. The Well19937a and 83 Well44497a generator are not maximally equidistributed (i.e. there are some dimensions or bits blocks 84 size for which they are not equidistributed). The Well512a, Well1024a, Well19937c and Well44497b are 85 maximally equidistributed for blocks size up to 32 bits (they should behave correctly also for double 86 based on more than 32 bits blocks, but equidistribution is not proven at these blocks sizes). 87 </p> 88 89 <p> 90 The MersenneTwister generator uses a 624 elements integer array, so it consumes less than 2.5 kilobytes. 91 The WELL generators use 6 integer arrays with a size equal to the pool size, so for example the 92 WELL44497b generator uses about 33 kilobytes. This may be important if a very large number of 93 generator instances were used at the same time. 94 </p> 95 96 <p> 97 All generators are quite fast. As an example, here are some comparisons, obtained on a 64 bits JVM on a 98 linux computer with a 2008 processor (AMD phenom Quad 9550 at 2.2 GHz). The generation rate for 99 MersenneTwister was about 27 millions doubles per second (remember we generate two 32 bits integers for 100 each double). Generation rates for other PRNG, relative to MersenneTwister: 101 </p> 102 103 <p> 104 <table border="1" align="center"> 105 <tr BGCOLOR="#CCCCFF"><td colspan="2"><font size="+2">Example of performances</font></td></tr> 106 <tr BGCOLOR="#EEEEFF"><font size="+1"><td>Name</td><td>generation rate (relative to MersenneTwister)</td></font></tr> 107 <tr><td>{@link org.apache.commons.math.random.MersenneTwister MersenneTwister}</td><td>1</td></tr> 108 <tr><td>{@link org.apache.commons.math.random.JDKRandomGenerator JDKRandomGenerator}</td><td>between 0.96 and 1.16</td></tr> 109 <tr><td>{@link org.apache.commons.math.random.Well512a Well512a}</td><td>between 0.85 and 0.88</td></tr> 110 <tr><td>{@link org.apache.commons.math.random.Well1024a Well1024a}</td><td>between 0.63 and 0.73</td></tr> 111 <tr><td>{@link org.apache.commons.math.random.Well19937a Well19937a}</td><td>between 0.70 and 0.71</td></tr> 112 <tr><td>{@link org.apache.commons.math.random.Well19937c Well19937c}</td><td>between 0.57 and 0.71</td></tr> 113 <tr><td>{@link org.apache.commons.math.random.Well44497a Well44497a}</td><td>between 0.69 and 0.71</td></tr> 114 <tr><td>{@link org.apache.commons.math.random.Well44497b Well44497b}</td><td>between 0.65 and 0.71</td></tr> 115 </table> 116 </p> 117 118 <p> 119 So for most simulation problems, the better generators like {@link 120 org.apache.commons.math.random.Well19937c Well19937c} and {@link 121 org.apache.commons.math.random.Well44497b Well44497b} are probably very good choices. 122 </p> 123 124 <p> 125 Note that <em>none</em> of these generators are suitable for cryptography. They are devoted 126 to simulation, and to generate very long series with strong properties on the series as a whole 127 (equidistribution, no correlation ...). They do not attempt to create small series but with 128 very strong properties of unpredictability as needed in cryptography. 129 </p> 130 131 </body> 132</html> 133