Files
LBPM/tests/TestColorSquareTube.cpp
2018-05-16 14:00:35 -04:00

499 lines
17 KiB
C++

//*************************************************************************
// Lattice Boltzmann Simulator for Single Phase Flow in Porous Media
// James E. McCLure
//*************************************************************************
#include <stdio.h>
#include <iostream>
#include <fstream>
#include "common/ScaLBL.h"
#include "common/MPI_Helpers.h"
std::shared_ptr<Database> loadInputs( int nprocs )
{
auto db = std::make_shared<Database>( "Domain.in" );
const int dim = 50;
db->putScalar<int>( "BC", 0 );
if ( nprocs == 1 ){
db->putVector<int>( "nproc", { 1, 1, 1 } );
db->putVector<int>( "n", { 3, 1, 1 } );
db->putScalar<int>( "nspheres", 0 );
db->putVector<double>( "L", { 1, 1, 1 } );
} else if ( nprocs == 2 ) {
db->putVector<int>( "nproc", { 2, 1, 1 } );
db->putVector<int>( "n", { dim, dim, dim } );
db->putScalar<int>( "nspheres", 0 );
db->putVector<double>( "L", { 1, 1, 1 } );
} else if ( nprocs == 4 ) {
db->putVector<int>( "nproc", { 2, 2, 1 } );
db->putVector<int>( "n", { dim, dim, dim } );
db->putScalar<int>( "nspheres", 0 );
db->putVector<double>( "L", { 1, 1, 1 } );
} else if (nprocs==8){
db->putVector<int>( "nproc", { 2, 2, 2 } );
db->putVector<int>( "n", { dim, dim, dim } );
db->putScalar<int>( "nspheres", 0 );
db->putVector<double>( "L", { 1, 1, 1 } );
}
return db;
}
//***************************************************************************************
int main(int argc, char **argv)
{
//*****************************************
// ***** MPI STUFF ****************
//*****************************************
// Initialize MPI
int rank,nprocs;
MPI_Init(&argc,&argv);
MPI_Comm comm = MPI_COMM_WORLD;
MPI_Comm_rank(comm,&rank);
MPI_Comm_size(comm,&nprocs);
int check;
{
if (rank == 0){
printf("********************************************************\n");
printf("Running Color Model: TestColor \n");
printf("********************************************************\n");
}
// BGK Model parameters
string FILENAME;
unsigned int nBlocks, nthreads;
int timestepMax, interval;
double Fx,Fy,Fz,tol;
// Domain variables
int i,j,k,n;
//if (rank == 0) printf("dim=%d\n",dim);
int timestep = 1;
int timesteps = 100;
int centralNode = 2;
double tauA = 1.0;
double tauB = 1.0;
double rhoA = 1.0;
double rhoB = 1.0;
double alpha = 0.001;
double beta = 0.95;
double tau = 1.0;
double mu=(tau-0.5)/3.0;
double rlx_setA=1.0/tau;
double rlx_setB = 8.f*(2.f-rlx_setA)/(8.f-rlx_setA);
Fx = Fy = 0.f;
Fz = 0.f;
int typeBC;
double din, dout, flux;
double inletA,inletB,outletA,outletB;
inletA=1.f;
inletB=0.f;
outletA=0.f;
outletB=1.f;
typeBC=4;
flux = 10.f;
dout=1.f;
// Load inputs
auto db = loadInputs( nprocs );
int Nx = db->getVector<int>( "n" )[0];
int Ny = db->getVector<int>( "n" )[1];
int Nz = db->getVector<int>( "n" )[2];
int nprocx = db->getVector<int>( "nproc" )[0];
int nprocy = db->getVector<int>( "nproc" )[1];
int nprocz = db->getVector<int>( "nproc" )[2];
if (rank==0){
printf("********************************************************\n");
printf("Sub-domain size = %i x %i x %i\n",Nx,Ny,Nz);
printf("********************************************************\n");
}
double iVol_global = 1.0/Nx/Ny/Nz/nprocx/nprocy/nprocz;
Domain Dm(db);
Nx += 2;
Ny += 2;
Nz += 2;
int N = Nx*Ny*Nz;
//.......................................................................
// Assign the phase ID field
//.......................................................................
char LocalRankString[8];
sprintf(LocalRankString,"%05d",rank);
char LocalRankFilename[40];
sprintf(LocalRankFilename,"ID.%05i",rank);
for (k=0;k<Nz;k++){
for (j=0;j<Ny;j++){
for (i=0;i<Nx;i++){
n = k*Nx*Ny + j*Nx + i;
Dm.id[n]=0;
}
}
}
int kproc = rank/(nprocx*nprocy);
int jproc = (rank-nprocx*nprocy*kproc)/nprocx;
int iproc = rank-nprocx*nprocy*kproc-nprocx*jproc;
printf("rank=%i, %i,%i,%i \n",rank,iproc,jproc,kproc);
// Initialize a square tube
for (k=1;k<Nz-1;k++){
for (j=1;j<Ny-1;j++){
for (i=1;i<Nx-1;i++){
n = k*Nx*Ny + j*Nx + i;
int iglobal= i+(Nx-2)*iproc;
int jglobal= j+(Ny-2)*jproc;
int kglobal= k+(Nz-2)*kproc;
// Initialize phase position field for parallel bubble test
if (iglobal < 2) Dm.id[n]=0;
else if (iglobal > (Nx-2)*nprocx-2) Dm.id[n]=0;
else if (jglobal < 2) Dm.id[n]=0;
else if (jglobal > (Ny-2)*nprocy-2) Dm.id[n]=0;
else if (kglobal < 20) Dm.id[n]=1;
else Dm.id[n]=2;
}
}
}
Dm.CommInit(comm);
//.......................................................................
// Compute the media porosity, assign phase labels and solid composition
//.......................................................................
double sum;
double sum_local=0.0, porosity;
int Np=0; // number of local pore nodes
double *PhaseLabel;
PhaseLabel = new double[N];
Dm.AssignComponentLabels(PhaseLabel);
//.......................................................................
for (k=1;k<Nz-1;k++){
for (j=1;j<Ny-1;j++){
for (i=1;i<Nx-1;i++){
n = k*Nx*Ny+j*Nx+i;
if (Dm.id[n] > 0){
sum_local+=1.0;
Np++;
}
}
}
}
MPI_Allreduce(&sum_local,&sum,1,MPI_DOUBLE,MPI_SUM,comm);
porosity = sum*iVol_global;
if (rank==0) printf("Media porosity = %f \n",porosity);
if (rank==0) printf ("Create ScaLBL_Communicator \n");
MPI_Barrier(comm);
// Create a communicator for the device (will use optimized layout)
ScaLBL_Communicator ScaLBL_Comm(Dm);
//Create a second communicator based on the regular data layout
ScaLBL_Communicator ScaLBL_Comm_Regular(Dm);
//...........device phase ID.................................................
if (rank==0) printf ("Copying phase ID to device \n");
char *ID;
ScaLBL_AllocateDeviceMemory((void **) &ID, N); // Allocate device memory
// Copy to the device
ScaLBL_CopyToDevice(ID, Dm.id, N);
//...........................................................................
if (rank==0){
printf("Total domain size = %i \n",N);
printf("Reduced domain size = %i \n",Np);
}
// LBM variables
if (rank==0) printf ("Set up the neighborlist \n");
int neighborSize=18*Np*sizeof(int);
int *neighborList;
IntArray Map(Nx,Ny,Nz);
neighborList= new int[18*Np];
ScaLBL_Comm.MemoryOptimizedLayoutAA(Map,neighborList,Dm.id,Np);
MPI_Barrier(comm);
//......................device distributions.................................
int dist_mem_size = Np*sizeof(double);
if (rank==0) printf ("Allocating distributions \n");
int *NeighborList;
int *dvcMap;
// double *f_even,*f_odd;
double *fq, *Aq, *Bq;
double *Den, *Phi;
double *ColorGrad;
double *Vel;
double *Pressure;
//...........................................................................
ScaLBL_AllocateDeviceMemory((void **) &NeighborList, neighborSize);
ScaLBL_AllocateDeviceMemory((void **) &dvcMap, sizeof(int)*Np);
ScaLBL_AllocateDeviceMemory((void **) &fq, 19*dist_mem_size);
ScaLBL_AllocateDeviceMemory((void **) &Aq, 7*dist_mem_size);
ScaLBL_AllocateDeviceMemory((void **) &Bq, 7*dist_mem_size);
ScaLBL_AllocateDeviceMemory((void **) &Den, 2*dist_mem_size);
ScaLBL_AllocateDeviceMemory((void **) &Phi, sizeof(double)*Nx*Ny*Nz);
ScaLBL_AllocateDeviceMemory((void **) &Pressure, sizeof(double)*Np);
ScaLBL_AllocateDeviceMemory((void **) &Vel, 3*sizeof(double)*Np);
ScaLBL_AllocateDeviceMemory((void **) &ColorGrad, 3*sizeof(double)*Np);
//...........................................................................
// Update GPU data structures
if (rank==0) printf ("Setting up device map and neighbor list \n");
int *TmpMap;
TmpMap=new int[Np];
for (k=1; k<Nz-1; k++){
for (j=1; j<Ny-1; j++){
for (i=1; i<Nx-1; i++){
int idx=Map(i,j,k);
if (!(idx < 0))
TmpMap[idx] = k*Nx*Ny+j*Nx+i;
}
}
}
//for (int idx=0; idx<Np; idx++) printf("Map=%i\n",TmpMap[idx]);
ScaLBL_CopyToDevice(dvcMap, TmpMap, sizeof(int)*Np);
ScaLBL_DeviceBarrier();
delete [] TmpMap;
// copy the neighbor list
ScaLBL_CopyToDevice(NeighborList, neighborList, neighborSize);
// initialize phi based on PhaseLabel (include solid component labels)
ScaLBL_CopyToDevice(Phi, PhaseLabel, N*sizeof(double));
//...........................................................................
if (rank==0) printf ("Initializing distributions \n");
// Initialize the phase field and variables
ScaLBL_PhaseField_Init(dvcMap, Phi, Den, Aq, Bq, 0, ScaLBL_Comm.last_interior, Np);
if (Dm.kproc()==0){
ScaLBL_SetSlice_z(Phi,1.0,Nx,Ny,Nz,0);
ScaLBL_SetSlice_z(Phi,1.0,Nx,Ny,Nz,1);
ScaLBL_SetSlice_z(Phi,1.0,Nx,Ny,Nz,2);
}
if (Dm.kproc() == nprocz-1){
ScaLBL_SetSlice_z(Phi,-1.0,Nx,Ny,Nz,Nz-1);
ScaLBL_SetSlice_z(Phi,-1.0,Nx,Ny,Nz,Nz-2);
ScaLBL_SetSlice_z(Phi,-1.0,Nx,Ny,Nz,Nz-3);
}
//************ MAIN ITERATION LOOP (timing communications)***************************************
if (rank==0) printf("Beginning AA timesteps...\n");
if (rank==0) printf("********************************************************\n");
if (rank==0) printf("No. of timesteps for timing: %i \n", timesteps);
//.......create and start timer............
double starttime,stoptime,cputime;
ScaLBL_DeviceBarrier(); MPI_Barrier(comm);
starttime = MPI_Wtime();
//timesteps=20;
//timestep=1;
while (timestep < timesteps) {
// *************ODD TIMESTEP*************
// Compute the Phase indicator field
// Read for Aq, Bq happens in this routine (requires communication)
ScaLBL_Comm.BiSendD3Q7AA(Aq,Bq); //READ FROM NORMAL
ScaLBL_D3Q7_AAodd_PhaseField(NeighborList, dvcMap, Aq, Bq, Den, Phi, ScaLBL_Comm.next, Np, Np);
ScaLBL_Comm.BiRecvD3Q7AA(Aq,Bq); //WRITE INTO OPPOSITE
ScaLBL_D3Q7_AAodd_PhaseField(NeighborList, dvcMap, Aq, Bq, Den, Phi, 0, ScaLBL_Comm.next, Np);
// Halo exchange for phase field
ScaLBL_Comm_Regular.SendHalo(Phi);
ScaLBL_Comm_Regular.RecvHalo(Phi);
// Perform the collision operation
ScaLBL_Comm.SendD3Q19AA(fq); //READ FROM NORMAL
ScaLBL_D3Q19_AAodd_Color(NeighborList, dvcMap, fq, Aq, Bq, Den, Phi, Vel, rhoA, rhoB, tauA, tauB,
alpha, beta, Fx, Fy, Fz, Nx, Nx*Ny, ScaLBL_Comm.next, Np, Np);
ScaLBL_Comm.RecvD3Q19AA(fq); //WRITE INTO OPPOSITE
// Set BCs
if (typeBC > 0){
ScaLBL_Comm.Color_BC_z(dvcMap, Phi, Den, inletA, inletB);
ScaLBL_Comm.Color_BC_Z(dvcMap, Phi, Den, outletA, outletB);
}
if (typeBC == 3){
ScaLBL_Comm.D3Q19_Pressure_BC_z(NeighborList, fq, din, timestep);
ScaLBL_Comm.D3Q19_Pressure_BC_Z(NeighborList, fq, dout, timestep);
}
if (typeBC == 4){
din = ScaLBL_Comm.D3Q19_Flux_BC_z(NeighborList, fq, flux, timestep);
ScaLBL_Comm.D3Q19_Pressure_BC_Z(NeighborList, fq, dout, timestep);
}
ScaLBL_D3Q19_AAodd_Color(NeighborList, dvcMap, fq, Aq, Bq, Den, Phi, Vel, rhoA, rhoB, tauA, tauB,
alpha, beta, Fx, Fy, Fz, Nx, Nx*Ny, 0, ScaLBL_Comm.next, Np);
ScaLBL_DeviceBarrier(); MPI_Barrier(comm);
timestep++;
// *************EVEN TIMESTEP*************
// Compute the Phase indicator field
ScaLBL_Comm.BiSendD3Q7AA(Aq,Bq); //READ FROM NORMAL
ScaLBL_D3Q7_AAeven_PhaseField(dvcMap, Aq, Bq, Den, Phi, ScaLBL_Comm.next, Np, Np);
ScaLBL_Comm.BiRecvD3Q7AA(Aq,Bq); //WRITE INTO OPPOSITE
ScaLBL_D3Q7_AAeven_PhaseField(dvcMap, Aq, Bq, Den, Phi, 0, ScaLBL_Comm.next, Np);
// Halo exchange for phase field
ScaLBL_Comm_Regular.SendHalo(Phi);
ScaLBL_Comm_Regular.RecvHalo(Phi);
// Perform the collision operation
ScaLBL_Comm.SendD3Q19AA(fq); //READ FORM NORMAL
ScaLBL_D3Q19_AAeven_Color(dvcMap, fq, Aq, Bq, Den, Phi, Vel, rhoA, rhoB, tauA, tauB,
alpha, beta, Fx, Fy, Fz, Nx, Nx*Ny, ScaLBL_Comm.next, Np, Np);
ScaLBL_Comm.RecvD3Q19AA(fq); //WRITE INTO OPPOSITE
// Set boundary conditions
if (typeBC > 0){
ScaLBL_Comm.Color_BC_z(dvcMap, Phi, Den, inletA, inletB);
ScaLBL_Comm.Color_BC_Z(dvcMap, Phi, Den, outletA, outletB);
}
if (typeBC == 3){
ScaLBL_Comm.D3Q19_Pressure_BC_z(NeighborList, fq, din, timestep);
ScaLBL_Comm.D3Q19_Pressure_BC_Z(NeighborList, fq, dout, timestep);
}
else if (typeBC == 4){
din = ScaLBL_Comm.D3Q19_Flux_BC_z(NeighborList, fq, flux, timestep);
ScaLBL_Comm.D3Q19_Pressure_BC_Z(NeighborList, fq, dout, timestep);
}
ScaLBL_D3Q19_AAeven_Color(dvcMap, fq, Aq, Bq, Den, Phi, Vel, rhoA, rhoB, tauA, tauB,
alpha, beta, Fx, Fy, Fz, Nx, Nx*Ny, 0, ScaLBL_Comm.next, Np);
ScaLBL_DeviceBarrier(); MPI_Barrier(comm);
timestep++;
//************************************************************************
}
//************************************************************************
stoptime = MPI_Wtime();
// cout << "CPU time: " << (stoptime - starttime) << " seconds" << endl;
cputime = stoptime - starttime;
// cout << "Lattice update rate: "<< double(Nx*Ny*Nz*timestep)/cputime/1000000 << " MLUPS" << endl;
double MLUPS = double(Np*timestep)/cputime/1000000;
if (rank==0) printf("********************************************************\n");
if (rank==0) printf("CPU time = %f \n", cputime);
if (rank==0) printf("Lattice update rate (per process)= %f MLUPS \n", MLUPS);
MLUPS *= nprocs;
if (rank==0) printf("Lattice update rate (process)= %f MLUPS \n", MLUPS);
if (rank==0) printf("********************************************************\n");
// Number of memory references for color model
double MemoryRefs = double(Np)*(77*8+(9+7+7)*4); // extra memory refs to read from neighborlist (every other timestep)
// number of memory references for the swap algorithm - GigaBytes / second
if (rank==0) printf("DRAM bandwidth (per process)= %f GB/sec \n",MemoryRefs*timestep/1e9/cputime);
// Report bandwidth in Gigabits per second
// communication bandwidth includes both send and recieve
if (rank==0) printf("Communication bandwidth (per process)= %f Gbit/sec \n",ScaLBL_Comm.CommunicationCount*64*timestep/1e9/cputime);
if (rank==0) printf("Aggregated communication bandwidth = %f Gbit/sec \n",nprocs*ScaLBL_Comm.CommunicationCount*64*timestep/1e9/cputime);
double *VEL;
VEL= new double [3*Np];
int SIZE=3*Np*sizeof(double);
ScaLBL_D3Q19_Momentum(fq,Vel,Np);
ScaLBL_DeviceBarrier(); MPI_Barrier(comm);
ScaLBL_CopyToHost(&VEL[0],&Vel[0],SIZE);
sum_local=0.f;
sum = 0.f;
for (k=1;k<Nz-1;k++){
for (j=1;j<Ny-1;j++){
for (i=1;i<Nx-1;i++){
n = k*Nx*Ny+j*Nx+i;
if (Dm.id[n] > 0){
int idx = Map(i,j,k);
sum_local+=VEL[2*Np+idx];
}
}
}
}
MPI_Allreduce(&sum_local,&sum,1,MPI_DOUBLE,MPI_SUM,comm);
double PoreVel = sum*iVol_global;
if (rank==0) printf("Average velocity = %f \n",PoreVel);
if (rank==0){
printf("Printing inlet velocity for rank=0 \n");
k=1;
for (j=1;j<Ny-1;j++){
for (i=1;i<Nx-1;i++){
n = k*Nx*Ny+j*Nx+i;
if (Dm.id[n] > 0){
int idx = Map(i,j,k);
double vz = VEL[2*Np+idx];
printf("%f ",vz);
}
}
printf("\n");
}
}
/*
double *PHASE;
PHASE= new double [Nx*Ny*Nz];
SIZE=Nx*Ny*Nz*sizeof(double);
ScaLBL_CopyToHost(&PHASE[0],&Phi[0],SIZE);
FILE *OUTFILE;
sprintf(LocalRankFilename,"Phase.%05i.raw",rank);
OUTFILE = fopen(LocalRankFilename,"wb");
fwrite(PHASE,8,N,OUTFILE);
fclose(OUTFILE);
double *DENA, *DENB, *TMPDAT;
SIZE=Np*sizeof(double);
TMPDAT = new double [Np];
DENA= new double [Nx*Ny*Nz];
DENB= new double [Nx*Ny*Nz];
ScaLBL_CopyToHost(&TMPDAT[0],&Den[0],SIZE);
ScaLBL_Comm.RegularLayout(Map,TMPDAT,DENA);
ScaLBL_CopyToHost(&TMPDAT[0],&Den[Np],SIZE);
ScaLBL_Comm.RegularLayout(Map,TMPDAT,DENB);
FILE *AFILE;
sprintf(LocalRankFilename,"na.%05i.raw",rank);
AFILE = fopen(LocalRankFilename,"wb");
fwrite(DENA,8,N,AFILE);
fclose(AFILE);
FILE *BFILE;
sprintf(LocalRankFilename,"nb.%05i.raw",rank);
BFILE = fopen(LocalRankFilename,"wb");
fwrite(DENB,8,N,BFILE);
fclose(BFILE);
double *CG;
CG= new double [3*Np];
ScaLBL_D3Q19_Gradient(dvcMap, Phi, ColorGrad, 0, Np, Np, Nx, Ny, Nz);
ScaLBL_CopyToHost(&CG[0],&ColorGrad[0],3*SIZE);
for (int idx=0; idx<Np; idx++){
double C=CG[idx]*CG[idx]+CG[Np+idx]*CG[Np+idx]+CG[2*Np+idx]*CG[2*Np+idx];
TMPDAT[idx]=C;
}
ScaLBL_Comm.RegularLayout(Map,TMPDAT,DENB);
FILE *CGFILE;
sprintf(LocalRankFilename,"cgrad.%05i.raw",rank);
CGFILE = fopen(LocalRankFilename,"wb");
fwrite(DENB,8,N,CGFILE);
fclose(CGFILE);
*/
}
// ****************************************************
MPI_Barrier(comm);
MPI_Finalize();
// ****************************************************
return check;
}