parallel.gpu.RandStream.getGlobalStream

R2026b

Get current global GPU random number stream

Syntax

stream = parallel.gpu.RandStream.getGlobalStream

Description

stream = parallel.gpu.RandStream.getGlobalStream returns the current global random number stream on the GPU.

Note

The gpurng function is a more concise alternative for many uses of parallel.gpu.RandStream.setGlobalStream.

Examples

Save the Default Global Stream

Use parallel.gpu.RandStream.getGlobalStream to save the default stream settings.

defaultStr = parallel.gpu.RandStream.getGlobalStream

defaultStr =

Threefry4x64_20 random stream on the GPU (current global stream)
             Seed: 0
  NormalTransform: BoxMuller

If you change the global stream, you can use the stream defaultStr to restore the default settings. For example, suppose that you change the global stream to a different stream.

newStr = parallel.gpu.RandStream("CombRecursive",NormalTransform="Inversion");
defaultStr = parallel.gpu.RandStream.setGlobalStream(newStr)

defaultStr =

Threefry4x64_20 random stream on the GPU
             Seed: 0
  NormalTransform: BoxMuller

defaultStr is no longer the current global GPU stream. Once you finish your calculations using the new global stream settings, you can reset the stream to the default settings.

newStr = parallel.gpu.RandStream.setGlobalStream(defaultStr)

newStr =

MRG32K3A random stream on the GPU
             Seed: 0
  NormalTransform: Inversion

Display defaultStr.

defaultStr

defaultStr =

Threefry4x64_20 random stream on the GPU (current global stream)
             Seed: 0
  NormalTransform: BoxMuller

defaultStr is once again the current global stream.

Output Arguments

stream — Global GPU random number stream

parallel.gpu.RandStream object

Global random number stream for generating random numbers on a GPU, returned as a parallel.gpu.RandStream object.

Extended Capabilities

Thread-Based Environment

Run code in the background using MATLAB® backgroundPool or accelerate code with Parallel Computing Toolbox™ ThreadPool.

The parallel.gpu.RandStream.getGlobalStream function fully supports thread-based environments. For more information, see Run MATLAB Functions in Thread-Based Environment.

Version History

Introduced in R2011b