System.Random.SplitMix

SplitMix is a splittable pseudorandom number generator (PRNG) that is quite fast.

Guy L. Steele, Jr., Doug Lea, and Christine H. Flood. 2014. Fast splittable pseudorandom number generators. In Proceedings of the 2014 ACM International Conference on Object Oriented Programming Systems Languages & Applications (OOPSLA '13). ACM, New York, NY, USA, 453-472. DOI: https://doi.org/10.1145/2660193.2660195

Guy L. Steele, Jr., Doug Lea, and Christine H. Flood. 2014. Fast splittable pseudorandom number generators. In Proceedings of the 2014 ACM International Conference on Object Oriented Programming Systems Languages & Applications (OOPSLA '14). ACM, New York, NY, USA, 453-472. DOI: https://doi.org/10.1145/2660193.2660195

The paper describes a new algorithm SplitMix for splittable pseudorandom number generator that is quite fast: 9 64 bit arithmetic/logical operations per 64 bits generated.

SplitMix is tested with two standard statistical test suites (DieHarder and TestU01, this implementation only using the former) and it appears to be adequate for "everyday" use, such as Monte Carlo algorithms and randomized data structures where speed is important.

In particular, it should not be used for cryptographic or security applications, because generated sequences of pseudorandom values are too predictable (the mixing functions are easily inverted, and two successive outputs suffice to reconstruct the internal state).

Note: This module supports all GHCs since GHC-7.0.4, but GHC-7.0 and GHC-7.2 have slow implementation, as there are no native popCount.

Documentation

data SMGen

SplitMix generator state.

Instances

Read SMGen

>>> readMaybe "SMGen 1 1" :: Maybe SMGen
Just (SMGen 1 1)

>>> readMaybe "SMGen 1 2" :: Maybe SMGen
Nothing

>>> readMaybe (show (mkSMGen 42)) :: Maybe SMGen
Just (SMGen 9297814886316923340 13679457532755275413)

Defined in System.Random.SplitMix

Methods

readsPrec :: Int -> ReadS SMGen

readList :: ReadS [SMGen]

readPrec :: ReadPrec SMGen

readListPrec :: ReadPrec [SMGen]

Show SMGen

Defined in System.Random.SplitMix

Methods

showsPrec :: Int -> SMGen -> ShowS

show :: SMGen -> String

showList :: [SMGen] -> ShowS

NFData SMGen

Defined in System.Random.SplitMix

Methods

rnf :: SMGen -> ()

RandomGen SMGen

Defined in System.Random.SplitMix

Methods

next :: SMGen -> (Int, SMGen)

genRange :: SMGen -> (Int, Int)

split :: SMGen -> (SMGen, SMGen)

nextWord64 :: SMGen -> (Word64, SMGen)

Generate a Word64.

>>> take 3 $ map (printf "%x") $ unfoldr (Just . nextWord64) (mkSMGen 1337) :: [String]
["b5c19e300e8b07b3","d600e0e216c0ac76","c54efc3b3cc5af29"]

nextInt :: SMGen -> (Int, SMGen)

Generate an Int.

nextDouble :: SMGen -> (Double, SMGen)

Generate a Double in [0, 1) range.

>>> take 8 $ map (printf "%0.3f") $ unfoldr (Just . nextDouble) (mkSMGen 1337) :: [String]
["0.710","0.836","0.771","0.409","0.297","0.527","0.589","0.067"]

splitSMGen :: SMGen -> (SMGen, SMGen)

Split a generator into a two uncorrelated generators.

Initialisation

mkSMGen :: Word64 -> SMGen

Preferred way to deterministically construct SMGen.

>>> mkSMGen 42
SMGen 9297814886316923340 13679457532755275413

initSMGen :: IO SMGen

Initialize SMGen using system time.

newSMGen :: IO SMGen

Derive a new generator instance from the global SMGen using splitSMGen.

seedSMGen

Arguments

:: Word64

seed

-> Word64

gamma

-> SMGen

Create SMGen using seed and gamma.

>>> seedSMGen 2 2
SMGen 2 3

seedSMGen' :: (Word64, Word64) -> SMGen

Like seedSMGen but takes a pair.

unseedSMGen :: SMGen -> (Word64, Word64)

Extract current state of SMGen.