Constant Temperature Simulation
[Tools]

Functions
__global__ void	gpuRescaleTKernel (int N, int dim, float T, FLOAT_ARRAY_TYPE *dVel, float *vSumSq)
	Temperature rescale kernel function.
void	rescaleT (int N, int dim, int numInterData, int numThreads, float T, int maxThreads, int maxBlocks, FLOAT_ARRAY_TYPE *dVel, float *sumSq)
	Rescale the system temperature.

---

Detailed Description

This module contains all functions for perfoming simulations at constant temperature

---

Function Documentation

__global__ void gpuRescaleTKernel	(	int	N,
		int	dim,
		float	T,
		FLOAT_ARRAY_TYPE *	dVel,
		float *	vSumSq
	)

GPU kernel which calculates the rescaling factor and rescales all velocity to fit the desired temperature. The kernel is called via rescaleT .

Attention:

This rescaling assumes at the moment a 3D simulation!

Parameters:

[in]	N	Number of particles
[in]	dim	System dimension
[in]	T	Desired temperature
[in,out]	dVel	Velocities
[in]	vSumSq	Squared sum of all velocities

Returns:

Returns CUDA error information

Definition at line 198 of file gpu_tools.cu.

void rescaleT	(	int	N,
		int	dim,
		int	numInterData,
		int	numThreads,
		float	T,
		int	maxThreads,
		int	maxBlocks,
		FLOAT_ARRAY_TYPE *	dVel,
		float *	sumSq
	)

To confine the system to a given temperature this function rescales the temperature accordingly.

Attention:

This rescaling assumes at the moment a 3D simulation!

Parameters:

[in]	N	Number of particles
[in]	dim	System dimension
[in]	numInterData	Intermediate data size for the reduction
[in]	numThreads	Number of threads used for reduction
[in]	T	Desired temperature
[in]	maxThreads	Maximum number of allowed threads
[in]	maxBlocks	Maximum number of allowed blocks
[in,out]	dVel	Velcity array on device
[in]	sumSq	Helper variable, content will be overwritten

Returns:

No return value.

Definition at line 188 of file gpu_tools.cu.

References checkCudaError(), and SqVecSumReduction().

Referenced by main().