Running multiple cores simultaneously using CUDA for GPU

Question

Is it possible to simultaneously launch two cores that perform independent tasks. For example, if I have this Cuda code

// host and device initialization
.......
.......

// launch kernel1
myMethod1 <<<.... >>> (params);

// launch kernel2
myMethod2 <<<.....>>> (params);

Assuming that these cores are independent, is it possible to run them simultaneously with the distribution of several grids / blocks for each. Does CUDA / OpenCL have this provision.

Answer 1

Only devices with the ability to calculate CUDA 2.0 and higher (i.e. Fermi) can support multiple simultaneous sound executions. See Section 3.2.6.3 of the CUDA 3.0 Programming Guide for:

2.0 . cudaGetDeviceProperties() concurrentKernels.

.

CUDA CUDA.

, , .

Answer 2

SM 2.0 .

, , . , , ​​ , , , , , ..

, . , SDK.

Answer 3

CUDA 2.1 = 16