Class CudaRandHost

Wrapper for a CUrand generator handle in host mode

Inheritance
System.Object
CudaRandHost
Inherited Members
System.Object.Equals(System.Object)
System.Object.Equals(System.Object, System.Object)
System.Object.GetHashCode()
System.Object.GetType()
System.Object.MemberwiseClone()
System.Object.ReferenceEquals(System.Object, System.Object)
System.Object.ToString()
Namespace: ManagedCuda.CudaRand
Assembly: CudaRand.dll
Syntax
public class CudaRandHost

Constructors

| Improve this Doc View Source

CudaRandHost(GeneratorType)

Creates a new random number generator of type Type

Declaration
public CudaRandHost(GeneratorType Type)
Parameters
Type Name Description
GeneratorType Type

Generator type

Properties

| Improve this Doc View Source

Generator

Returns the wrapped curand generator handle

Declaration
public CurandGenerator Generator { get; }
Property Value
Type Description
CurandGenerator

Methods

| Improve this Doc View Source

Dispose()

Dispose

Declaration
public void Dispose()
| Improve this Doc View Source

Dispose(Boolean)

For IDisposable

Declaration
protected virtual void Dispose(bool fDisposing)
Parameters
Type Name Description
System.Boolean fDisposing
| Improve this Doc View Source

Finalize()

For dispose

Declaration
protected void Finalize()
| Improve this Doc View Source

Generate(UInt32[])

Use generator to generate num 32-bit results into the device memory at output. The device memory must have been previously allocated and be large enough to hold all the results.

Results are 32-bit values with every bit random.

Declaration
public void Generate(uint[] output)
Parameters
Type Name Description
System.UInt32[] output

CudaDeviceVariable
| Improve this Doc View Source

Generate(UInt64[])

Use generator to generate num 64-bit results into the device memory at output. The device memory must have been previously allocated and be large enough to hold all the results.

Results are 64-bit values with every bit random.

Declaration
public void Generate(ulong[] output)
Parameters
Type Name Description
System.UInt64[] output

CudaDeviceVariable
| Improve this Doc View Source

GenerateLogNormal(Double[], Double, Double)

Use generator to generate num double results into the device memory at outputPtr. The device memory must have been previously allocated and be large enough to hold all the results. Launches are done with the stream set using curandSetStream(), or the null stream if no stream has been set.

Results are 64-bit floating point values with log-normal distribution based on an associated normal distribution with mean mean and standard deviation stddev.

Normally distributed results are generated from pseudorandom generators with a Box-Muller transform, and so require num to be even. Quasirandom generators use an inverse cumulative distribution function to preserve dimensionality. The normally distributed results are transformed into log-normal distribution.

There may be slight numerical differences between results generated on the GPU with generators created with ::curandCreateGenerator() and results calculated on the CPU with generators created with ::curandCreateGeneratorHost(). These differences arise because of differences in results for transcendental functions. In addition, future versions of CURAND may use newer versions of the CUDA math library, so different versions of CURAND may give slightly different numerical values.

Declaration
public void GenerateLogNormal(double[] output, double mean, double stddev)
Parameters
Type Name Description
System.Double[] output

CudaDeviceVariable
System.Double mean
System.Double stddev
| Improve this Doc View Source

GenerateLogNormal(Single[], Single, Single)

Use generator to generate num float results into the device memory at outputPtr. The device memory must have been previously allocated and be large enough to hold all the results. Launches are done with the stream set using ::curandSetStream(), or the null stream if no stream has been set.

Results are 32-bit floating point values with log-normal distribution based on an associated normal distribution with mean mean and standard deviation stddev.

Normally distributed results are generated from pseudorandom generators with a Box-Muller transform, and so require num to be even. Quasirandom generators use an inverse cumulative distribution function to preserve dimensionality.

The normally distributed results are transformed into log-normal distribution.

There may be slight numerical differences between results generated on the GPU with generators created with ::curandCreateGenerator() and results calculated on the CPU with generators created with ::curandCreateGeneratorHost(). These differences arise because of differences in results for transcendental functions. In addition, future versions of CURAND may use newer versions of the CUDA math library, so different versions of CURAND may give slightly different numerical values.

Declaration
public void GenerateLogNormal(float[] output, float mean, float stddev)
Parameters
Type Name Description
System.Single[] output

CudaDeviceVariable
System.Single mean
System.Single stddev
| Improve this Doc View Source

GenerateNormal(Double[], Double, Double)

Use generator to generate num double results into the device memory at outputPtr. The device memory must have been previously allocated and be large enough to hold all the results. Launches are done with the stream set using curandSetStream(), or the null stream if no stream has been set.

Results are 64-bit floating point values with mean mean and standard deviation stddev.

Normally distributed results are generated from pseudorandom generators with a Box-Muller transform, and so require num to be even. Quasirandom generators use an inverse cumulative distribution function to preserve dimensionality.

There may be slight numerical differences between results generated on the GPU with generators created with curandCreateGenerator() and results calculated on the CPU with generators created with curandCreateGeneratorHost(). These differences arise because of differences in results for transcendental functions. In addition, future versions of CURAND may use newer versions of the CUDA math library, so different versions of CURAND may give slightly different numerical values.

Declaration
public void GenerateNormal(double[] output, double mean, double stddev)
Parameters
Type Name Description
System.Double[] output

CudaDeviceVariable
System.Double mean
System.Double stddev
| Improve this Doc View Source

GenerateNormal(Single[], Single, Single)

Use generator to generate num float results into the device memory at outputPtr. The device memory must have been previously allocated and be large enough to hold all the results. Launches are done with the stream set using curandSetStream(), or the null stream if no stream has been set.

Results are 32-bit floating point values with mean mean and standard deviation stddev.

Normally distributed results are generated from pseudorandom generators with a Box-Muller transform, and so require num to be even. Quasirandom generators use an inverse cumulative distribution function to preserve dimensionality.

There may be slight numerical differences between results generated on the GPU with generators created with curandCreateGenerator() and results calculated on the CPU with generators created with curandCreateGeneratorHost(). These differences arise because of differences in results for transcendental functions. In addition, future versions of CURAND may use newer versions of the CUDA math library, so different versions of CURAND may give slightly different numerical values.

Declaration
public void GenerateNormal(float[] output, float mean, float stddev)
Parameters
Type Name Description
System.Single[] output

CudaDeviceVariable
System.Single mean
System.Single stddev
| Improve this Doc View Source

GenerateSeeds()

Generate the starting state of the generator. This function is automatically called by generation functions such as Generate(CudaDeviceVariable) and GenerateUniform(CudaDeviceVariable). It can be called manually for performance testing reasons to separate timings for starting state generation and random number generation.

Declaration
public void GenerateSeeds()
| Improve this Doc View Source

GenerateUniform(Double[])

Use generator to generate num double results into the device memory at outputPtr. The device memory must have been previously allocated and be large enough to hold all the results. Launches are done with the stream set using curandSetStream(), or the null stream if no stream has been set.

Results are 64-bit double precision floating point values between 0.0 and 1.0, excluding 0.0 and including 1.0.

Declaration
public void GenerateUniform(double[] output)
Parameters
Type Name Description
System.Double[] output

CudaDeviceVariable
| Improve this Doc View Source

GenerateUniform(Single[])

Use generator to generate num float results into the device memory at outputPtr. The device memory must have been previously allocated and be large enough to hold all the results. Launches are done with the stream set using SetStream(), or the null stream if no stream has been set.

Results are 32-bit floating point values between 0.0f and 1.0f, excluding 0.0f and including 1.0f.

Declaration
public void GenerateUniform(float[] output)
Parameters
Type Name Description
System.Single[] output

CudaDeviceVariable
| Improve this Doc View Source

GetDirectionVectors32(DirectionVectorSet)

Get an array of direction vectors that can be used for quasirandom number generation.

The array contains vectors for many dimensions. Each dimension has 32 vectors. Each individual vector is an unsigned int.

Legal values for set are: - JoeKuo6_32 (20,000 dimensions) - ScrambledJoeKuo6_32 (20,000 dimensions)

Declaration
public static DirectionVectors32[] GetDirectionVectors32(DirectionVectorSet set)
Parameters
Type Name Description
DirectionVectorSet set
Returns
Type Description
DirectionVectors32[]
| Improve this Doc View Source

GetDirectionVectors64(DirectionVectorSet)

Get an array of direction vectors that can be used for quasirandom number generation.

The array contains vectors for many dimensions. Each dimension has 64 vectors. Each individual vector is an unsigned long long.

Legal values for set are: - JoeKuo6_64 (20,000 dimensions) - ScrambledJoeKuo6_64 (20,000 dimensions)

Declaration
public static DirectionVectors64[] GetDirectionVectors64(DirectionVectorSet set)
Parameters
Type Name Description
DirectionVectorSet set
Returns
Type Description
DirectionVectors64[]
| Improve this Doc View Source

GetScrambleConstants32()

Get scramble constants that can be used for quasirandom number generation.

The array contains constants for many dimensions. Each dimension has a single uint constant.

Declaration
public static uint[] GetScrambleConstants32()
Returns
Type Description
System.UInt32[]
| Improve this Doc View Source

GetScrambleConstants64()

Get scramble constants that can be used for quasirandom number generation.

The array contains constants for many dimensions. Each dimension has a single ulong constant.

Declaration
public static ulong[] GetScrambleConstants64()
Returns
Type Description
System.UInt64[]
| Improve this Doc View Source

GetVersion()

Returns the version number of the dynamically linked CURAND library.

Declaration
public static Version GetVersion()
Returns
Type Description
System.Version
| Improve this Doc View Source

SetGeneratorOrdering(Ordering)

Set the ordering of results of the pseudo or quasirandom number generator.

Legal values of order for pseudorandom generators are:

- PseudoDefault

- PseudoBest

- PseudoSeeded

Legal values of order for quasirandom generators are:

- QuasiDefault

Declaration
public void SetGeneratorOrdering(Ordering order)
Parameters
Type Name Description
Ordering order
| Improve this Doc View Source

SetOffset(UInt64)

Set the absolute offset of the pseudo or quasirandom number generator.

All values of offset are valid. The offset position is absolute, not relative to the current position in the sequence.

Declaration
public void SetOffset(ulong offset)
Parameters
Type Name Description
System.UInt64 offset

All values of offset are valid.
| Improve this Doc View Source

SetPseudoRandomGeneratorSeed(UInt64)

Set the seed value of the pseudorandom number generator.

All values of seed are valid. Different seeds will produce different sequences. Different seeds will often not be statistically correlated with each other, but some pairs of seed values may generate sequences which are statistically correlated.

Declaration
public void SetPseudoRandomGeneratorSeed(ulong seed)
Parameters
Type Name Description
System.UInt64 seed

All values of seed are valid.
| Improve this Doc View Source

SetQuasiRandomGeneratorDimensions(UInt32)

Set the number of dimensions to be generated by the quasirandom number generator.

Legal values for dimensions are 1 to 20000.

Declaration
public void SetQuasiRandomGeneratorDimensions(uint dimensions)
Parameters
Type Name Description
System.UInt32 dimensions

Legal values for dimensions are 1 to 20000.