moebiusband/NuSiF-Solver
2
0
Fork 2
Code Issues Pull Requests 1 Packages Projects Releases Wiki Activity

2D MPI enhanced solver

Browse Source
This commit is contained in:
Aditya Ujeniya
2024-07-24 12:04:14 +02:00
parent f48927ce6f
commit 312af6f663
52 changed files with 4246 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
BasicSolver/2D-mpi/src/comm-v2.c
View File

@@ -288,8 +288,6 @@ void commPartition(Comm* c, int jmax, int imax)
void commUpdateDatatypes(Comm* oldcomm, Comm* newcomm, int imaxLocal, int jmaxLocal)
{
Comm* newcomm;
#if defined _MPI
newcomm->comm = MPI_COMM_NULL;
int result = MPI_Comm_dup(oldcomm->comm, &newcomm->comm);

2
BasicSolver/2D-mpi/src/solver-mg.c
View File

@@ -13,7 +13,7 @@
#define FINEST_LEVEL 0
#define COARSEST_LEVEL (s->levels - 1)
#define S(i, j) s[(j) * (imaxLocal + 2) + (i)]
// #define S(i, j) s[(j) * (imaxLocal + 2) + (i)]
#define E(i, j) e[(j) * (imaxLocal + 2) + (i)]
#define R(i, j) r[(j) * (imaxLocal + 2) + (i)]
#define OLD(i, j) old[(j) * (imaxLocal + 2) + (i)]

87
EnhancedSolver/2D-mpi/Makefile Normal file
View File

@@ -0,0 +1,87 @@
#=======================================================================================
# Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
# All rights reserved.
# Use of this source code is governed by a MIT-style
# license that can be found in the LICENSE file.
#=======================================================================================
#CONFIGURE BUILD SYSTEM
TARGET = exe-$(TAG)
BUILD_DIR = ./$(TAG)
SRC_DIR = ./src
MAKE_DIR = ./
Q ?= @
#DO NOT EDIT BELOW
include $(MAKE_DIR)/config.mk
include $(MAKE_DIR)/include_$(TAG).mk
INCLUDES += -I$(SRC_DIR) -I$(BUILD_DIR)
VPATH = $(SRC_DIR)
SRC = $(filter-out $(wildcard $(SRC_DIR)/*-*.c),$(wildcard $(SRC_DIR)/*.c))
ASM = $(patsubst $(SRC_DIR)/%.c, $(BUILD_DIR)/%.s, $(SRC))
OBJ = $(patsubst $(SRC_DIR)/%.c, $(BUILD_DIR)/%.o, $(SRC))
OBJ += $(BUILD_DIR)/comm-$(COMM_TYPE).o
OBJ += $(BUILD_DIR)/solver-$(SOLVER).o
SOURCES = $(SRC) $(wildcard $(SRC_DIR)/*.h)
CPPFLAGS := $(CPPFLAGS) $(DEFINES) $(OPTIONS) $(INCLUDES)
${TARGET}: $(BUILD_DIR) $(OBJ)
$(info ===> LINKING $(TARGET))
$(Q)${LINKER} ${LFLAGS} -o $(TARGET) $(OBJ) $(LIBS)
$(BUILD_DIR)/%.o: %.c $(MAKE_DIR)/include_$(TAG).mk $(MAKE_DIR)/config.mk
$(info ===> COMPILE $@)
$(CC) -c $(CPPFLAGS) $(CFLAGS) $< -o $@
$(Q)$(GCC) $(CPPFLAGS) -MT $(@:.d=.o) -MM $< > $(BUILD_DIR)/$*.d
$(BUILD_DIR)/%.s: %.c
$(info ===> GENERATE ASM $@)
$(CC) -S $(CPPFLAGS) $(CFLAGS) $< -o $@
.PHONY: clean distclean vis_clean vis tags info asm format
vis:
$(info ===> GENERATE VISUALIZATION)
@gnuplot -e "filename='pressure.dat'" ./surface.plot
@gnuplot -e "filename='velocity.dat'" ./vector.plot
vis_clean:
$(info ===> CLEAN VISUALIZATION)
@rm -f *.dat
@rm -f *.png
@rm -f ./vis_files/*.dat
@rm -f ./vis_files/*.gif
clean: vis_clean
$(info ===> CLEAN)
@rm -rf $(BUILD_DIR)
@rm -f tags
distclean: clean
$(info ===> DIST CLEAN)
@rm -f $(TARGET)
@rm -f *.dat
@rm -f *.png
info:
$(info $(CFLAGS))
$(Q)$(CC) $(VERSION)
asm: $(BUILD_DIR) $(ASM)
tags:
$(info ===> GENERATE TAGS)
$(Q)ctags -R
format:
@for src in $(SOURCES) ; do \
echo "Formatting $$src" ; \
clang-format -i $$src ; \
done
@echo "Done"
$(BUILD_DIR):
@mkdir $(BUILD_DIR)
-include $(OBJ:.o=.d)

48
EnhancedSolver/2D-mpi/README.md Normal file
View File

@@ -0,0 +1,48 @@
# C source skeleton
## Build
1. Configure the toolchain and additional options in `config.mk`:
```
# Supported: GCC, CLANG, ICC
TAG ?= GCC
ENABLE_OPENMP ?= false
OPTIONS += -DARRAY_ALIGNMENT=64
#OPTIONS += -DVERBOSE_AFFINITY
#OPTIONS += -DVERBOSE_DATASIZE
#OPTIONS += -DVERBOSE_TIMER
```
The verbosity options enable detailed output about affinity settings, allocation sizes and timer resolution.
2. Build with:
```
make
```
You can build multiple toolchains in the same directory, but notice that the Makefile is only acting on the one currently set.
Intermediate build results are located in the `<TOOLCHAIN>` directory.
To output the executed commands use:
```
make Q=
```
3. Clean up with:
```
make clean
```
to clean intermediate build results.
```
make distclean
```
to clean intermediate build results and binary.
4. (Optional) Generate assembler:
```
make asm
```
The assembler files will also be located in the `<TOOLCHAIN>` directory.

79
EnhancedSolver/2D-mpi/backstep.par Normal file
View File

@@ -0,0 +1,79 @@
#==============================================================================
# Laminar Canal Flow
#==============================================================================
# Problem specific Data:
# ---------------------
name backstep # name of flow setup
bcTop 1 # flags for boundary conditions
bcBottom 1 # 1 = no-slip 3 = outflow
bcLeft 3 # 2 = free-slip 4 = periodic
bcRight 3 #
gx 0.0 # Body forces (e.g. gravity)
gy 0.0 #
re 36000.0 # Reynolds number
u_init 1.0 # initial value for velocity in x-direction
v_init 0.0 # initial value for velocity in y-direction
p_init 1.0 # initial value for pressure
# Geometry Data:
# -------------
xlength 7.0 # domain size in x-direction
ylength 1.5 # domain size in y-direction
imax 210 # number of interior cells in x-direction
jmax 45 # number of interior cells in y-direction
# Time Data:
# ---------
te 60.0 # final time
dt 0.02 # time stepsize
tau 0.5 # safety factor for time stepsize control (<0 constant delt)
# Pressure Iteration Data:
# -----------------------
itermax 500 # maximal number of pressure iteration in one time step
eps 0.0001 # stopping tolerance for pressure iteration
rho 0.52
omg 1.8 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 15 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------
numberOfParticles 200
startTime 0 #if you want to see particles trapped in recirculation zone, startTime should be set to 0
injectTimePeriod 1.0
writeTimePeriod 0.5
x1 0.0
y1 0.5
x2 0.0
y2 1.5
# Obstacle Geometry Data:
# -----------------------
# Shape 0 disable, 1 Rectangle/Square, 2 Circle
shape 1
xCenter 0.0
yCenter 0.0
xRectLength 2.0
yRectLength 1.0
circleRadius 1.0
#===============================================================================

78
EnhancedSolver/2D-mpi/canal.par Normal file
View File

@@ -0,0 +1,78 @@
#==============================================================================
# Laminar Canal Flow
#==============================================================================
# Problem specific Data:
# ---------------------
name canal # name of flow setup
bcTop 1 # flags for boundary conditions
bcBottom 1 # 1 = no-slip 3 = outflow
bcLeft 3 # 2 = free-slip 4 = periodic
bcRight 3 #
gx 0.0 # Body forces (e.g. gravity)
gy 0.0 #
re 100.0 # Reynolds number
u_init 1.0 # initial value for velocity in x-direction
v_init 0.0 # initial value for velocity in y-direction
p_init 0.0 # initial value for pressure
# Geometry Data:
# -------------
xlength 30.0 # domain size in x-direction
ylength 4.0 # domain size in y-direction
imax 200 # number of interior cells in x-direction
jmax 40 # number of interior cells in y-direction
# Time Data:
# ---------
te 60.0 # final time
dt 0.02 # time stepsize
tau 0.5 # safety factor for time stepsize control (<0 constant delt)
# Multigrid data:
# ---------
levels 2 # Multigrid levels
presmooth 5 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Pressure Iteration Data:
# -----------------------
itermax 500 # maximal number of pressure iteration in one time step
eps 0.00001 # stopping tolerance for pressure iteration
omg 1.8 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
# Particle Tracing Data:
# -----------------------
numberOfParticles 60
startTime 10.0
injectTimePeriod 4.0
writeTimePeriod 1.0
x1 1.0
y1 0.0
x2 1.0
y2 4.0
# Obstacle Geometry Data:
# -----------------------
# Shape 0 disable, 1 Rectangle/Square, 2 Circle
shape 0
xCenter 10.0
yCenter 2
xRectLength 6.0
yRectLength 1.0
circleRadius 1.0
#===============================================================================

17
EnhancedSolver/2D-mpi/config.mk Normal file
View File

@@ -0,0 +1,17 @@
# Supported: GCC, CLANG, ICC
TAG ?= ICC
# Supported: true, false
ENABLE_MPI ?= true
ENABLE_OPENMP ?= false
# Supported: rb, mg
SOLVER ?= mg
# Run in debug settings ?= mg
COMM_TYPE ?= v3
#Feature options
OPTIONS += -DARRAY_ALIGNMENT=64
OPTIONS += -DVERBOSE
# OPTIONS += -DTEST
#OPTIONS += -DVERBOSE_AFFINITY
#OPTIONS += -DVERBOSE_DATASIZE
#OPTIONS += -DVERBOSE_TIMER

79
EnhancedSolver/2D-mpi/dcavity.par Normal file
View File

@@ -0,0 +1,79 @@
#==============================================================================
# Driven Cavity
#==============================================================================
# Problem specific Data:
# ---------------------
name dcavity # name of flow setup
bcTop 1 # flags for boundary conditions
bcBottom 1 # 1 = no-slip 3 = outflow
bcLeft 1 # 2 = free-slip 4 = periodic
bcRight 1 #
gx 0.0 # Body forces (e.g. gravity)
gy 0.0 #
re 10.0 # Reynolds number
u_init 1.0 # initial value for velocity in x-direction
v_init 0.0 # initial value for velocity in y-direction
p_init 0.0 # initial value for pressure
# Geometry Data:
# -------------
xlength 1.0 # domain size in x-direction
ylength 1.0 # domain size in y-direction
imax 128 # number of interior cells in x-direction
jmax 128 # number of interior cells in y-direction
# Time Data:
# ---------
te 10.0 # final time
dt 0.02 # time stepsize
tau 0.5 # safety factor for time stepsize control (<0 constant delt)
# Pressure Iteration Data:
# -----------------------
itermax 1000 # maximal number of pressure iteration in one time step
eps 0.001 # stopping tolerance for pressure iteration
rho 0.5
omg 1.8 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 10 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------
numberOfParticles 200
startTime 2.0
injectTimePeriod 0.5
writeTimePeriod 0.2
x1 0.1
y1 0.9
x2 0.9
y2 0.9
# Obstacle Geometry Data:
# -----------------------
# Shape 0 disable, 1 Rectangle/Square, 2 Circle
shape 0
xCenter 0.5
yCenter 0.5
xRectLength 0.5
yRectLength 0.5
circleRadius 0.5
#===============================================================================

21
EnhancedSolver/2D-mpi/include_CLANG.mk Normal file
View File

@@ -0,0 +1,21 @@
ifeq ($(ENABLE_MPI),true)
CC = mpicc
DEFINES = -D_MPI
else
CC = cc
endif
GCC = cc
LINKER = $(CC)
ifeq ($(ENABLE_OPENMP),true)
OPENMP = -fopenmp
#OPENMP = -Xpreprocessor -fopenmp #required on Macos with homebrew libomp
LIBS = # -lomp
endif
VERSION = --version
CFLAGS = -Ofast -std=c17
LFLAGS = $(OPENMP) -lm
DEFINES += -D_GNU_SOURCE# -DDEBUG
INCLUDES = -I/opt/homebrew/include

20
EnhancedSolver/2D-mpi/include_GCC.mk Normal file
View File

@@ -0,0 +1,20 @@
ifeq ($(ENABLE_MPI),true)
CC = mpicc
DEFINES = -D_MPI
else
CC = gcc
endif
GCC = gcc
LINKER = $(CC)
ifeq ($(ENABLE_OPENMP),true)
OPENMP = -fopenmp
endif
VERSION = --version
CFLAGS = -Ofast -ffreestanding -std=c99 $(OPENMP)
LFLAGS = $(OPENMP)
DEFINES += -D_GNU_SOURCE
INCLUDES =
LIBS =

20
EnhancedSolver/2D-mpi/include_ICC.mk Normal file
View File

@@ -0,0 +1,20 @@
ifeq ($(ENABLE_MPI),true)
CC = mpiicc
DEFINES = -D_MPI
else
CC = icc
endif
GCC = gcc
LINKER = $(CC)
ifeq ($(ENABLE_OPENMP),true)
OPENMP = -qopenmp
endif
VERSION = --version
CFLAGS = -O3 -xHost -qopt-zmm-usage=high -std=c99 $(OPENMP)
LFLAGS = $(OPENMP)
DEFINES += -D_GNU_SOURCE# -DDEBUG
INCLUDES =
LIBS =

79
EnhancedSolver/2D-mpi/karman.par Normal file
View File

@@ -0,0 +1,79 @@
#==============================================================================
# Laminar Canal Flow
#==============================================================================
# Problem specific Data:
# ---------------------
name karman # name of flow setup
bcTop 1 # flags for boundary conditions
bcBottom 1 # 1 = no-slip 3 = outflow
bcLeft 3 # 2 = free-slip 4 = periodic
bcRight 3 #
gx 0.0 # Body forces (e.g. gravity)
gy 0.0 #
re 5050.0 # Reynolds number
u_init 1.0 # initial value for velocity in x-direction
v_init 0.0 # initial value for velocity in y-direction
p_init 0.0 # initial value for pressure
# Geometry Data:
# -------------
xlength 30.0 # domain size in x-direction
ylength 8.0 # domain size in y-direction
imax 400 # number of interior cells in x-direction
jmax 200 # number of interior cells in y-direction
# Time Data:
# ---------
te 150.0 # final time
dt 0.02 # time stepsize
tau 0.5 # safety factor for time stepsize control (<0 constant delt)
# Pressure Iteration Data:
# -----------------------
itermax 200 # maximal number of pressure iteration in one time step
eps 0.001 # stopping tolerance for pressure iteration
rho 0.52
omg 1.75 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 15 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------
numberOfParticles 200
startTime 50
injectTimePeriod 1.0
writeTimePeriod 0.5
x1 0.0
y1 3.8
x2 0.0
y2 4.1
# Obstacle Geometry Data:
# -----------------------
# Shape 0 disable, 1 Rectangle/Square, 2 Circle
shape 2
xCenter 5.0
yCenter 4.0
xRectLength 2.0
yRectLength 1.0
circleRadius 1.0
#===============================================================================

61
EnhancedSolver/2D-mpi/src/affinity.c Normal file
View File

@@ -0,0 +1,61 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#ifdef __linux__
#ifdef _OPENMP
#include <pthread.h>
#include <sched.h>
#include <stdio.h>
#include <stdlib.h>
#include <sys/syscall.h>
#include <sys/types.h>
#include <unistd.h>
#define MAX_NUM_THREADS 128
#define gettid() syscall(SYS_gettid)
static int getProcessorID(cpu_set_t* cpu_set)
{
int processorId;
for (processorId = 0; processorId < MAX_NUM_THREADS; processorId++) {
if (CPU_ISSET(processorId, cpu_set)) {
break;
}
}
return processorId;
}
int affinity_getProcessorId()
{
cpu_set_t cpu_set;
CPU_ZERO(&cpu_set);
sched_getaffinity(gettid(), sizeof(cpu_set_t), &cpu_set);
return getProcessorID(&cpu_set);
}
void affinity_pinThread(int processorId)
{
cpu_set_t cpuset;
pthread_t thread;
thread = pthread_self();
CPU_ZERO(&cpuset);
CPU_SET(processorId, &cpuset);
pthread_setaffinity_np(thread, sizeof(cpu_set_t), &cpuset);
}
void affinity_pinProcess(int processorId)
{
cpu_set_t cpuset;
CPU_ZERO(&cpuset);
CPU_SET(processorId, &cpuset);
sched_setaffinity(0, sizeof(cpu_set_t), &cpuset);
}
#endif /*_OPENMP*/
#endif /*__linux__*/

14
EnhancedSolver/2D-mpi/src/affinity.h Normal file
View File

@@ -0,0 +1,14 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#ifndef AFFINITY_H
#define AFFINITY_H
extern int affinity_getProcessorId();
extern void affinity_pinProcess(int);
extern void affinity_pinThread(int);
#endif /*AFFINITY_H*/

38
EnhancedSolver/2D-mpi/src/allocate.c Normal file
View File

@@ -0,0 +1,38 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#include <errno.h>
#include <stddef.h>
#include <stdio.h>
#include <stdlib.h>
#include "allocate.h"
void* allocate(size_t alignment, size_t bytesize)
{
int errorCode;
void* ptr;
errorCode = posix_memalign(&ptr, alignment, bytesize);
if (errorCode) {
if (errorCode == EINVAL) {
fprintf(stderr, "Error: Alignment parameter is not a power of two\n");
exit(EXIT_FAILURE);
}
if (errorCode == ENOMEM) {
fprintf(stderr, "Error: Insufficient memory to fulfill the request\n");
exit(EXIT_FAILURE);
}
}
if (ptr == NULL) {
fprintf(stderr, "Error: posix_memalign failed!\n");
exit(EXIT_FAILURE);
}
return ptr;
}

13
EnhancedSolver/2D-mpi/src/allocate.h Normal file
View File

@@ -0,0 +1,13 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#ifndef __ALLOCATE_H_
#define __ALLOCATE_H_
#include <stdlib.h>
extern void* allocate(size_t alignment, size_t bytesize);
#endif

204
EnhancedSolver/2D-mpi/src/comm-v1.c Normal file
View File

@@ -0,0 +1,204 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <stdlib.h>
#include "comm.h"
#ifdef _MPI
// subroutines local to this module
static int sum(int* sizes, int position)
{
int sum = 0;
for (int i = 0; i < position; i++) {
sum += sizes[i];
}
return sum;
}
static void gatherArray(
Comm* c, int cnt, int* rcvCounts, int* displs, double* src, double* dst)
{
double* sendbuffer = src + (c->imaxLocal + 2);
if (c->rank == 0) {
sendbuffer = src;
}
MPI_Gatherv(sendbuffer,
cnt,
MPI_DOUBLE,
dst,
rcvCounts,
displs,
MPI_DOUBLE,
0,
MPI_COMM_WORLD);
}
#endif // defined _MPI
// exported subroutines
int commIsBoundary(Comm* c, int direction)
{
#ifdef _MPI
switch (direction) {
case L:
return 1;
break;
case R:
return 1;
break;
case B:
return c->rank == 0;
break;
case T:
return c->rank == (c->size - 1);
break;
}
#endif
return 1;
}
void commExchange(Comm* c, double* grid)
{
#ifdef _MPI
MPI_Request requests[4] = { MPI_REQUEST_NULL,
MPI_REQUEST_NULL,
MPI_REQUEST_NULL,
MPI_REQUEST_NULL };
/* exchange ghost cells with top neighbor */
if (c->rank + 1 < c->size) {
int top = c->rank + 1;
double* src = grid + (c->jmaxLocal) * (c->imaxLocal + 2) + 1;
double* dst = grid + (c->jmaxLocal + 1) * (c->imaxLocal + 2) + 1;
MPI_Isend(src, c->imaxLocal, MPI_DOUBLE, top, 1, MPI_COMM_WORLD, &requests[0]);
MPI_Irecv(dst, c->imaxLocal, MPI_DOUBLE, top, 2, MPI_COMM_WORLD, &requests[1]);
}
/* exchange ghost cells with bottom neighbor */
if (c->rank > 0) {
int bottom = c->rank - 1;
double* src = grid + (c->imaxLocal + 2) + 1;
double* dst = grid + 1;
MPI_Isend(src, c->imaxLocal, MPI_DOUBLE, bottom, 2, MPI_COMM_WORLD, &requests[2]);
MPI_Irecv(dst, c->imaxLocal, MPI_DOUBLE, bottom, 1, MPI_COMM_WORLD, &requests[3]);
}
MPI_Waitall(4, requests, MPI_STATUSES_IGNORE);
#endif
}
void commShift(Comm* c, double* f, double* g)
{
#ifdef _MPI
MPI_Request requests[2] = { MPI_REQUEST_NULL, MPI_REQUEST_NULL };
/* shift G */
/* receive ghost cells from bottom neighbor */
if (c->rank > 0) {
int bottom = c->rank - 1;
MPI_Irecv(g + 1,
c->imaxLocal,
MPI_DOUBLE,
bottom,
0,
MPI_COMM_WORLD,
&requests[0]);
}
if (c->rank + 1 < c->size) {
int top = c->rank + 1;
double* buf = g + (c->jmaxLocal) * (c->imaxLocal + 2) + 1;
/* send ghost cells to top neighbor */
MPI_Isend(buf, c->imaxLocal, MPI_DOUBLE, top, 0, MPI_COMM_WORLD, &requests[1]);
}
MPI_Waitall(2, requests, MPI_STATUSES_IGNORE);
#endif
}
void commCollectResult(Comm* c,
double* ug,
double* vg,
double* pg,
double* u,
double* v,
double* p,
int jmax,
int imax)
{
#ifdef _MPI
int *rcvCounts, *displs;
int cnt = c->jmaxLocal * (imax + 2);
if (c->rank == 0) {
rcvCounts = (int*)malloc(c->size * sizeof(int));
displs = (int*)malloc(c->size * sizeof(int));
}
if (c->rank == 0 && c->size == 1) {
cnt = (c->jmaxLocal + 2) * (imax + 2);
} else if (c->rank == 0 || c->rank == (c->size - 1)) {
cnt = (c->jmaxLocal + 1) * (imax + 2);
}
MPI_Gather(&cnt, 1, MPI_INTEGER, rcvCounts, 1, MPI_INTEGER, 0, MPI_COMM_WORLD);
if (c->rank == 0) {
displs[0] = 0;
int cursor = rcvCounts[0];
for (int i = 1; i < c->size; i++) {
displs[i] = cursor;
cursor += rcvCounts[i];
}
}
gatherArray(c, cnt, rcvCounts, displs, p, pg);
gatherArray(c, cnt, rcvCounts, displs, u, ug);
gatherArray(c, cnt, rcvCounts, displs, v, vg);
#endif
}
void commPartition(Comm* c, int jmax, int imax)
{
#ifdef _MPI
c->imaxLocal = imax;
c->jmaxLocal = sizeOfRank(c->coords[JDIM], c->size, jmax);
#else
c->imaxLocal = imax;
c->jmaxLocal = jmax;
#endif
}
void commUpdateDatatypes(Comm* oldcomm, Comm* newcomm, int imaxLocal, int jmaxLocal)
{
Comm* newcomm;
#if defined _MPI
newcomm->comm = MPI_COMM_NULL;
int result = MPI_Comm_dup(oldcomm->comm, &newcomm->comm);
if (result == MPI_ERR_COMM) {
printf("\nNull communicator. Duplication failed !!\n");
}
#endif
newcomm->imaxLocal = imaxLocal;
newcomm->jmaxLocal = jmaxLocal;
}
void commFreeCommunicator(Comm* comm)
{
#ifdef _MPI
MPI_Comm_free(&comm->comm);
#endif
}

335
EnhancedSolver/2D-mpi/src/comm-v2.c Normal file
View File

@@ -0,0 +1,335 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <stdio.h>
#include <stdlib.h>
#include "comm.h"
#ifdef _MPI
// subroutines local to this module
static int sum(int* sizes, int position)
{
int sum = 0;
for (int i = 0; i < position; i++) {
sum += sizes[i];
}
return sum;
}
static void assembleResult(Comm* c, double* src, double* dst, int imax, int jmax)
{
MPI_Request* requests;
int numRequests = 1;
if (c->rank == 0) {
numRequests = c->size + 1;
} else {
numRequests = 1;
}
requests = (MPI_Request*)malloc(numRequests * sizeof(MPI_Request));
/* all ranks send their bulk array, including the external boundary layer */
MPI_Datatype bulkType;
int oldSizes[NDIMS] = { c->jmaxLocal + 2, c->imaxLocal + 2 };
int newSizes[NDIMS] = { c->jmaxLocal, c->imaxLocal };
int starts[NDIMS] = { 1, 1 };
if (commIsBoundary(c, L)) {
newSizes[CIDIM] += 1;
starts[CIDIM] = 0;
}
if (commIsBoundary(c, R)) {
newSizes[CIDIM] += 1;
}
if (commIsBoundary(c, B)) {
newSizes[CJDIM] += 1;
starts[CJDIM] = 0;
}
if (commIsBoundary(c, T)) {
newSizes[CJDIM] += 1;
}
MPI_Type_create_subarray(NDIMS,
oldSizes,
newSizes,
starts,
MPI_ORDER_C,
MPI_DOUBLE,
&bulkType);
MPI_Type_commit(&bulkType);
MPI_Isend(src, 1, bulkType, 0, 0, c->comm, &requests[0]);
int newSizesI[c->size];
int newSizesJ[c->size];
MPI_Gather(&newSizes[CIDIM], 1, MPI_INT, newSizesI, 1, MPI_INT, 0, MPI_COMM_WORLD);
MPI_Gather(&newSizes[CJDIM], 1, MPI_INT, newSizesJ, 1, MPI_INT, 0, MPI_COMM_WORLD);
/* rank 0 assembles the subdomains */
if (c->rank == 0) {
for (int i = 0; i < c->size; i++) {
MPI_Datatype domainType;
int oldSizes[NDIMS] = { jmax + 2, imax + 2 };
int newSizes[NDIMS] = { newSizesJ[i], newSizesI[i] };
int coords[NDIMS];
MPI_Cart_coords(c->comm, i, NDIMS, coords);
int starts[NDIMS] = { sum(newSizesJ, coords[JDIM]),
sum(newSizesI, coords[IDIM]) };
printf(
"Rank: %d, Coords(i,j): %d %d, Size(i,j): %d %d, Target Size(i,j): %d %d "
"Starts(i,j): %d %d\n",
i,
coords[IDIM],
coords[JDIM],
oldSizes[CIDIM],
oldSizes[CJDIM],
newSizes[CIDIM],
newSizes[CJDIM],
starts[CIDIM],
starts[CJDIM]);
MPI_Type_create_subarray(NDIMS,
oldSizes,
newSizes,
starts,
MPI_ORDER_C,
MPI_DOUBLE,
&domainType);
MPI_Type_commit(&domainType);
MPI_Irecv(dst, 1, domainType, i, 0, c->comm, &requests[i + 1]);
MPI_Type_free(&domainType);
}
}
MPI_Waitall(numRequests, requests, MPI_STATUSES_IGNORE);
}
#endif // defined _MPI
// exported subroutines
int commIsBoundary(Comm* c, int direction)
{
#ifdef _MPI
switch (direction) {
case L:
return c->coords[IDIM] == 0;
break;
case R:
return c->coords[IDIM] == (c->dims[IDIM] - 1);
break;
case B:
return c->coords[JDIM] == 0;
break;
case T:
return c->coords[JDIM] == (c->dims[JDIM] - 1);
break;
}
#endif
return 1;
}
void commExchange(Comm* c, double* grid)
{
#ifdef _MPI
MPI_Request requests[8];
for (int i = 0; i < 8; i++)
requests[i] = MPI_REQUEST_NULL;
for (int i = 0; i < NDIRS; i++) {
double* sbuf = grid + c->sdispls[i];
double* rbuf = grid + c->rdispls[i];
int tag = 0;
if (c->neighbours[i] != MPI_PROC_NULL) {
// printf("DEBUG: Rank %d - SendRecv with %d\n", c->rank, c->neighbours[i]);
tag = c->neighbours[i];
}
MPI_Irecv(rbuf,
1,
c->bufferTypes[i],
c->neighbours[i],
tag,
c->comm,
&requests[i * 2]);
MPI_Isend(sbuf,
1,
c->bufferTypes[i],
c->neighbours[i],
c->rank,
c->comm,
&requests[i * 2 + 1]);
}
MPI_Waitall(8, requests, MPI_STATUSES_IGNORE);
#endif
}
void commShift(Comm* c, double* f, double* g)
{
#ifdef _MPI
MPI_Request requests[4] = { MPI_REQUEST_NULL,
MPI_REQUEST_NULL,
MPI_REQUEST_NULL,
MPI_REQUEST_NULL };
/* shift G */
/* receive ghost cells from bottom neighbor */
double* buf = g + 1;
MPI_Irecv(buf,
1,
c->bufferTypes[B],
c->neighbours[B],
0,
c->comm,
&requests[0]);
/* send ghost cells to top neighbor */
buf = g + (c->jmaxLocal) * (c->imaxLocal + 2) + 1;
MPI_Isend(buf, 1, c->bufferTypes[T], c->neighbours[T], 0, c->comm, &requests[1]);
/* shift F */
/* receive ghost cells from left neighbor */
buf = f + (c->imaxLocal + 2);
MPI_Irecv(buf,
1,
c->bufferTypes[L],
c->neighbours[L],
1,
c->comm,
&requests[2]);
/* send ghost cells to right neighbor */
buf = f + (c->imaxLocal + 2) + (c->imaxLocal);
MPI_Isend(buf,
1,
c->bufferTypes[R],
c->neighbours[R],
1,
c->comm,
&requests[3]);
MPI_Waitall(4, requests, MPI_STATUSES_IGNORE);
#endif
}
void commCollectResult(Comm* c,
double* ug,
double* vg,
double* pg,
double* u,
double* v,
double* p,
int imax,
int jmax)
{
#ifdef _MPI
/* collect P */
assembleResult(c, p, pg, imax, jmax);
/* collect U */
assembleResult(c, u, ug, imax, jmax);
/* collect V */
assembleResult(c, v, vg, imax, jmax);
#endif
}
void commPartition(Comm* c, int jmax, int imax)
{
#ifdef _MPI
int dims[NDIMS] = { 0, 0 };
int periods[NDIMS] = { 0, 0 };
MPI_Dims_create(c->size, NDIMS, dims);
MPI_Cart_create(MPI_COMM_WORLD, NDIMS, dims, periods, 0, &c->comm);
MPI_Cart_shift(c->comm, IDIM, 1, &c->neighbours[L], &c->neighbours[R]);
MPI_Cart_shift(c->comm, JDIM, 1, &c->neighbours[B], &c->neighbours[T]);
MPI_Cart_get(c->comm, NDIMS, c->dims, periods, c->coords);
int imaxLocal = sizeOfRank(c->coords[IDIM], dims[IDIM], imax);
int jmaxLocal = sizeOfRank(c->coords[JDIM], dims[JDIM], jmax);
c->imaxLocal = imaxLocal;
c->jmaxLocal = jmaxLocal;
MPI_Datatype jBufferType;
MPI_Type_contiguous(imaxLocal, MPI_DOUBLE, &jBufferType);
MPI_Type_commit(&jBufferType);
MPI_Datatype iBufferType;
MPI_Type_vector(jmaxLocal, 1, imaxLocal + 2, MPI_DOUBLE, &iBufferType);
MPI_Type_commit(&iBufferType);
c->bufferTypes[L] = iBufferType;
c->bufferTypes[R] = iBufferType;
c->bufferTypes[B] = jBufferType;
c->bufferTypes[T] = jBufferType;
c->sdispls[L] = (imaxLocal + 2) + 1;
c->sdispls[R] = (imaxLocal + 2) + imaxLocal;
c->sdispls[B] = (imaxLocal + 2) + 1;
c->sdispls[T] = jmaxLocal * (imaxLocal + 2) + 1;
c->rdispls[L] = (imaxLocal + 2);
c->rdispls[R] = (imaxLocal + 2) + (imaxLocal + 1);
c->rdispls[B] = 1;
c->rdispls[T] = (jmaxLocal + 1) * (imaxLocal + 2) + 1;
#else
c->imaxLocal = imax;
c->jmaxLocal = jmax;
#endif
}
void commUpdateDatatypes(Comm* oldcomm, Comm* newcomm, int imaxLocal, int jmaxLocal)
{
#if defined _MPI
newcomm->comm = MPI_COMM_NULL;
int result = MPI_Comm_dup(oldcomm->comm, &newcomm->comm);
if (result == MPI_ERR_COMM) {
printf("\nNull communicator. Duplication failed !!\n");
}
newcomm->imaxLocal = imaxLocal;
newcomm->jmaxLocal = jmaxLocal;
MPI_Datatype jBufferType;
MPI_Type_contiguous(imaxLocal, MPI_DOUBLE, &jBufferType);
MPI_Type_commit(&jBufferType);
MPI_Datatype iBufferType;
MPI_Type_vector(jmaxLocal, 1, imaxLocal + 2, MPI_DOUBLE, &iBufferType);
MPI_Type_commit(&iBufferType);
newcomm->bufferTypes[L] = iBufferType;
newcomm->bufferTypes[R] = iBufferType;
newcomm->bufferTypes[B] = jBufferType;
newcomm->bufferTypes[T] = jBufferType;
newcomm->sdispls[L] = (imaxLocal + 2) + 1;
newcomm->sdispls[R] = (imaxLocal + 2) + imaxLocal;
newcomm->sdispls[B] = (imaxLocal + 2) + 1;
newcomm->sdispls[T] = jmaxLocal * (imaxLocal + 2) + 1;
newcomm->rdispls[L] = (imaxLocal + 2);
newcomm->rdispls[R] = (imaxLocal + 2) + (imaxLocal + 1);
newcomm->rdispls[B] = 1;
newcomm->rdispls[T] = (jmaxLocal + 1) * (imaxLocal + 2) + 1;
#else
newcomm->imaxLocal = imaxLocal;
newcomm->jmaxLocal = jmaxLocal;
#endif
}
void commFreeCommunicator(Comm* comm)
{
#ifdef _MPI
MPI_Comm_free(&comm->comm);
#endif
}

316
EnhancedSolver/2D-mpi/src/comm-v3.c Normal file
View File

@@ -0,0 +1,316 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <stdio.h>
#include <stdlib.h>
#include "comm.h"
#ifdef _MPI
// subroutines local to this module
static int sum(int* sizes, int position)
{
int sum = 0;
for (int i = 0; i < position; i++) {
sum += sizes[i];
}
return sum;
}
static void assembleResult(Comm* c, double* src, double* dst, int imax, int jmax)
{
MPI_Request* requests;
int numRequests = 1;
if (c->rank == 0) {
numRequests = c->size + 1;
} else {
numRequests = 1;
}
requests = (MPI_Request*)malloc(numRequests * sizeof(MPI_Request));
/* all ranks send their bulk array, including the external boundary layer */
MPI_Datatype bulkType;
int oldSizes[NDIMS] = { c->jmaxLocal + 2, c->imaxLocal + 2 };
int newSizes[NDIMS] = { c->jmaxLocal, c->imaxLocal };
int starts[NDIMS] = { 1, 1 };
if (commIsBoundary(c, L)) {
newSizes[CIDIM] += 1;
starts[CIDIM] = 0;
}
if (commIsBoundary(c, R)) {
newSizes[CIDIM] += 1;
}
if (commIsBoundary(c, B)) {
newSizes[CJDIM] += 1;
starts[CJDIM] = 0;
}
if (commIsBoundary(c, T)) {
newSizes[CJDIM] += 1;
}
MPI_Type_create_subarray(NDIMS,
oldSizes,
newSizes,
starts,
MPI_ORDER_C,
MPI_DOUBLE,
&bulkType);
MPI_Type_commit(&bulkType);
MPI_Isend(src, 1, bulkType, 0, 0, c->comm, &requests[0]);
int newSizesI[c->size];
int newSizesJ[c->size];
MPI_Gather(&newSizes[CIDIM], 1, MPI_INT, newSizesI, 1, MPI_INT, 0, MPI_COMM_WORLD);
MPI_Gather(&newSizes[CJDIM], 1, MPI_INT, newSizesJ, 1, MPI_INT, 0, MPI_COMM_WORLD);
/* rank 0 assembles the subdomains */
if (c->rank == 0) {
for (int i = 0; i < c->size; i++) {
MPI_Datatype domainType;
int oldSizes[NDIMS] = { jmax + 2, imax + 2 };
int newSizes[NDIMS] = { newSizesJ[i], newSizesI[i] };
int coords[NDIMS];
MPI_Cart_coords(c->comm, i, NDIMS, coords);
int starts[NDIMS] = { sum(newSizesJ, coords[JDIM]),
sum(newSizesI, coords[IDIM]) };
printf(
"Rank: %d, Coords(i,j): %d %d, Size(i,j): %d %d, Target Size(i,j): %d %d "
"Starts(i,j): %d %d\n",
i,
coords[IDIM],
coords[JDIM],
oldSizes[CIDIM],
oldSizes[CJDIM],
newSizes[CIDIM],
newSizes[CJDIM],
starts[CIDIM],
starts[CJDIM]);
MPI_Type_create_subarray(NDIMS,
oldSizes,
newSizes,
starts,
MPI_ORDER_C,
MPI_DOUBLE,
&domainType);
MPI_Type_commit(&domainType);
MPI_Irecv(dst, 1, domainType, i, 0, c->comm, &requests[i + 1]);
MPI_Type_free(&domainType);
}
}
MPI_Waitall(numRequests, requests, MPI_STATUSES_IGNORE);
}
#endif // defined _MPI
// exported subroutines
int commIsBoundary(Comm* c, int direction)
{
#ifdef _MPI
switch (direction) {
case L:
return c->coords[IDIM] == 0;
break;
case R:
return c->coords[IDIM] == (c->dims[IDIM] - 1);
break;
case B:
return c->coords[JDIM] == 0;
break;
case T:
return c->coords[JDIM] == (c->dims[JDIM] - 1);
break;
}
#endif
return 1;
}
void commExchange(Comm* c, double* grid)
{
#ifdef _MPI
int counts[NDIRS] = { 1, 1, 1, 1 };
MPI_Neighbor_alltoallw(grid,
counts,
c->sdispls,
c->bufferTypes,
grid,
counts,
c->rdispls,
c->bufferTypes,
c->comm);
#endif
}
void commShift(Comm* c, double* f, double* g)
{
#ifdef _MPI
MPI_Request requests[4] = { MPI_REQUEST_NULL,
MPI_REQUEST_NULL,
MPI_REQUEST_NULL,
MPI_REQUEST_NULL };
/* shift G */
/* receive ghost cells from bottom neighbor */
double* buf = g + 1;
MPI_Irecv(buf,
1,
c->bufferTypes[B],
c->neighbours[B],
0,
c->comm,
&requests[0]);
/* send ghost cells to top neighbor */
buf = g + (c->jmaxLocal) * (c->imaxLocal + 2) + 1;
MPI_Isend(buf, 1, c->bufferTypes[T], c->neighbours[T], 0, c->comm, &requests[1]);
/* shift F */
/* receive ghost cells from left neighbor */
buf = f + (c->imaxLocal + 2);
MPI_Irecv(buf,
1,
c->bufferTypes[L],
c->neighbours[L],
1,
c->comm,
&requests[2]);
/* send ghost cells to right neighbor */
buf = f + (c->imaxLocal + 2) + (c->imaxLocal);
MPI_Isend(buf,
1,
c->bufferTypes[R],
c->neighbours[R],
1,
c->comm,
&requests[3]);
MPI_Waitall(4, requests, MPI_STATUSES_IGNORE);
#endif
}
void commCollectResult(Comm* c,
double* ug,
double* vg,
double* pg,
double* u,
double* v,
double* p,
int imax,
int jmax)
{
#ifdef _MPI
/* collect P */
assembleResult(c, p, pg, imax, jmax);
/* collect U */
assembleResult(c, u, ug, imax, jmax);
/* collect V */
assembleResult(c, v, vg, imax, jmax);
#endif
}
void commPartition(Comm* c, int jmax, int imax)
{
#ifdef _MPI
int dims[NDIMS] = { 0, 0 };
int periods[NDIMS] = { 0, 0 };
MPI_Dims_create(c->size, NDIMS, dims);
MPI_Cart_create(MPI_COMM_WORLD, NDIMS, dims, periods, 0, &c->comm);
MPI_Cart_shift(c->comm, IDIM, 1, &c->neighbours[L], &c->neighbours[R]);
MPI_Cart_shift(c->comm, JDIM, 1, &c->neighbours[B], &c->neighbours[T]);
MPI_Cart_get(c->comm, NDIMS, c->dims, periods, c->coords);
int imaxLocal = sizeOfRank(c->coords[IDIM], dims[IDIM], imax);
int jmaxLocal = sizeOfRank(c->coords[JDIM], dims[JDIM], jmax);
c->imaxLocal = imaxLocal;
c->jmaxLocal = jmaxLocal;
MPI_Datatype jBufferType;
MPI_Type_contiguous(imaxLocal, MPI_DOUBLE, &jBufferType);
MPI_Type_commit(&jBufferType);
MPI_Datatype iBufferType;
MPI_Type_vector(jmaxLocal, 1, imaxLocal + 2, MPI_DOUBLE, &iBufferType);
MPI_Type_commit(&iBufferType);
c->bufferTypes[L] = iBufferType;
c->bufferTypes[R] = iBufferType;
c->bufferTypes[B] = jBufferType;
c->bufferTypes[T] = jBufferType;
size_t dblsize = sizeof(double);
c->sdispls[L] = ((imaxLocal + 2) + 1) * dblsize;
c->sdispls[R] = ((imaxLocal + 2) + imaxLocal) * dblsize;
c->sdispls[B] = ((imaxLocal + 2) + 1) * dblsize;
c->sdispls[T] = (jmaxLocal * (imaxLocal + 2) + 1) * dblsize;
c->rdispls[L] = (imaxLocal + 2) * dblsize;
c->rdispls[R] = ((imaxLocal + 2) + (imaxLocal + 1)) * dblsize;
c->rdispls[B] = 1 * dblsize;
c->rdispls[T] = ((jmaxLocal + 1) * (imaxLocal + 2) + 1) * dblsize;
#else
c->imaxLocal = imax;
c->jmaxLocal = jmax;
#endif
}
void commUpdateDatatypes(Comm* oldcomm, Comm* newcomm, int imaxLocal, int jmaxLocal)
{
#if defined _MPI
int result = MPI_Comm_dup(oldcomm->comm, &newcomm->comm);
if (result == MPI_ERR_COMM) {
printf("\nNull communicator. Duplication failed !!\n");
}
newcomm->imaxLocal = imaxLocal;
newcomm->jmaxLocal = jmaxLocal;
MPI_Datatype jBufferType;
MPI_Type_contiguous(imaxLocal, MPI_DOUBLE, &jBufferType);
MPI_Type_commit(&jBufferType);
MPI_Datatype iBufferType;
MPI_Type_vector(jmaxLocal, 1, imaxLocal + 2, MPI_DOUBLE, &iBufferType);
MPI_Type_commit(&iBufferType);
newcomm->bufferTypes[L] = iBufferType;
newcomm->bufferTypes[R] = iBufferType;
newcomm->bufferTypes[B] = jBufferType;
newcomm->bufferTypes[T] = jBufferType;
newcomm->sdispls[L] = (imaxLocal + 2) + 1;
newcomm->sdispls[R] = (imaxLocal + 2) + imaxLocal;
newcomm->sdispls[B] = (imaxLocal + 2) + 1;
newcomm->sdispls[T] = jmaxLocal * (imaxLocal + 2) + 1;
newcomm->rdispls[L] = (imaxLocal + 2);
newcomm->rdispls[R] = (imaxLocal + 2) + (imaxLocal + 1);
newcomm->rdispls[B] = 1;
newcomm->rdispls[T] = (jmaxLocal + 1) * (imaxLocal + 2) + 1;
#else
newcomm->imaxLocal = imaxLocal;
newcomm->jmaxLocal = jmaxLocal;
#endif
}
void commFreeCommunicator(Comm* comm)
{
#ifdef _MPI
MPI_Comm_free(&comm->comm);
#endif
}

129
EnhancedSolver/2D-mpi/src/comm.c Normal file
View File

@@ -0,0 +1,129 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <stdio.h>
#include <stdlib.h>
#include "comm.h"
// subroutines local to this module
int sizeOfRank(int rank, int size, int N)
{
return N / size + ((N % size > rank) ? 1 : 0);
}
void commReduction(double* v, int op)
{
#ifdef _MPI
if (op == MAX) {
MPI_Allreduce(MPI_IN_PLACE, v, 1, MPI_DOUBLE, MPI_MAX, MPI_COMM_WORLD);
} else if (op == SUM) {
MPI_Allreduce(MPI_IN_PLACE, v, 1, MPI_DOUBLE, MPI_SUM, MPI_COMM_WORLD);
}
#endif
}
void commPrintConfig(Comm* c)
{
#ifdef _MPI
fflush(stdout);
MPI_Barrier(MPI_COMM_WORLD);
if (commIsMaster(c)) {
printf("Communication setup:\n");
}
for (int i = 0; i < c->size; i++) {
if (i == c->rank) {
printf("\tRank %d of %d\n", c->rank, c->size);
printf("\tNeighbours (bottom, top, left, right): %d %d, %d, %d\n",
c->neighbours[B],
c->neighbours[T],
c->neighbours[L],
c->neighbours[R]);
printf("\tIs boundary:\n");
printf("\t\tLEFT: %d\n", commIsBoundary(c, L));
printf("\t\tRIGHT: %d\n", commIsBoundary(c, R));
printf("\t\tBOTTOM: %d\n", commIsBoundary(c, B));
printf("\t\tTOP: %d\n", commIsBoundary(c, T));
printf("\tCoordinates (i,j) %d %d\n", c->coords[IDIM], c->coords[JDIM]);
printf("\tDims (i,j) %d %d\n", c->dims[IDIM], c->dims[JDIM]);
printf("\tLocal domain size (i,j) %dx%d\n", c->imaxLocal, c->jmaxLocal);
fflush(stdout);
}
MPI_Barrier(MPI_COMM_WORLD);
}
#endif
}
void commInit(Comm* c, int argc, char** argv)
{
#ifdef _MPI
MPI_Init(&argc, &argv);
MPI_Comm_rank(MPI_COMM_WORLD, &(c->rank));
MPI_Comm_size(MPI_COMM_WORLD, &(c->size));
#else
c->rank = 0;
c->size = 1;
#endif
}
void commTestInit(Comm* c, double* p, double* f, double* g)
{
int imax = c->imaxLocal;
int jmax = c->jmaxLocal;
int rank = c->rank;
for (int j = 0; j < jmax + 2; j++) {
for (int i = 0; i < imax + 2; i++) {
p[j * (imax + 2) + i] = rank;
f[j * (imax + 2) + i] = rank;
g[j * (imax + 2) + i] = rank;
}
}
}
static void testWriteFile(char* filename, double* grid, int imax, int jmax)
{
FILE* fp = fopen(filename, "w");
if (fp == NULL) {
printf("Error!\n");
exit(EXIT_FAILURE);
}
for (int j = 0; j < jmax + 2; j++) {
for (int i = 0; i < imax + 2; i++) {
fprintf(fp, "%.2f ", grid[j * (imax + 2) + i]);
}
fprintf(fp, "\n");
}
fclose(fp);
}
void commTestWrite(Comm* c, double* p, double* f, double* g)
{
int imax = c->imaxLocal;
int jmax = c->jmaxLocal;
int rank = c->rank;
char filename[30];
snprintf(filename, 30, "ptest-%d.dat", rank);
testWriteFile(filename, p, imax, jmax);
snprintf(filename, 30, "ftest-%d.dat", rank);
testWriteFile(filename, f, imax, jmax);
snprintf(filename, 30, "gtest-%d.dat", rank);
testWriteFile(filename, g, imax, jmax);
}
void commFinalize(Comm* c)
{
#ifdef _MPI
MPI_Finalize();
#endif
}

57
EnhancedSolver/2D-mpi/src/comm.h Normal file
View File

@@ -0,0 +1,57 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#ifndef __COMM_H_
#define __COMM_H_
#if defined(_MPI)
#include <mpi.h>
#endif
enum direction { L = 0, R, B, T, NDIRS }; // L = Left, R = Right, B = Bottom, T =Top
enum dimension { IDIM = 0, JDIM, NDIMS };
enum cdimension { CJDIM = 0, CIDIM };
enum layer { HALO = 0, BULK };
enum op { MAX = 0, SUM };
typedef struct {
int rank;
int size;
#if defined(_MPI)
MPI_Comm comm;
MPI_Datatype bufferTypes[NDIRS];
MPI_Aint sdispls[NDIRS];
MPI_Aint rdispls[NDIRS];
#endif
int neighbours[NDIRS];
int coords[NDIMS], dims[NDIMS];
int imaxLocal, jmaxLocal;
} Comm;
extern int sizeOfRank(int rank, int size, int N);
extern void commInit(Comm* c, int argc, char** argv);
extern void commTestInit(Comm* c, double* p, double* f, double* g);
extern void commTestWrite(Comm* c, double* p, double* f, double* g);
extern void commFinalize(Comm* c);
extern void commPartition(Comm* c, int jmax, int imax);
extern void commPrintConfig(Comm*);
extern void commExchange(Comm*, double*);
extern void commShift(Comm* c, double* f, double* g);
extern void commReduction(double* v, int op);
extern int commIsBoundary(Comm* c, int direction);
extern void commUpdateDatatypes(Comm*, Comm*, int, int);
extern void commFreeCommunicator(Comm*);
extern void commCollectResult(Comm* c,
double* ug,
double* vg,
double* pg,
double* u,
double* v,
double* p,
int jmax,
int imax);
static inline int commIsMaster(Comm* c) { return c->rank == 0; }
#endif // __COMM_H_

718
EnhancedSolver/2D-mpi/src/discretization.c Normal file
View File

@@ -0,0 +1,718 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <float.h>
#include <math.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "allocate.h"
#include "comm.h"
#include "discretization.h"
#include "parameter.h"
#include "util.h"
static double distance(double i, double j, double iCenter, double jCenter)
{
return sqrt(pow(iCenter - i, 2) + pow(jCenter - j, 2) * 1.0);
}
double sumOffset(double* sizes, int init, int offset, int coord)
{
double sum = 0;
for (int i = init - offset; coord > 0; i -= offset, --coord) {
sum += sizes[i];
}
return sum;
}
void print(Discretization* d, double* grid)
{
int imaxLocal = d->comm.imaxLocal;
for (int i = 0; i < d->comm.size; i++) {
if (i == d->comm.rank) {
sleep(1 * d->comm.rank);
printf("### RANK %d LVL "
"###################################################### #\n ",
d->comm.rank);
for (int j = 0; j < d->comm.jmaxLocal + 2; j++) {
printf("%02d: ", j);
for (int i = 0; i < d->comm.imaxLocal + 2; i++) {
printf("%2.2f ", grid[j * (imaxLocal + 2) + i]);
}
printf("\n");
}
fflush(stdout);
}
}
}
static void printConfig(Discretization* d)
{
if (commIsMaster(&d->comm)) {
printf("Parameters for #%s#\n", d->problem);
printf("BC Left:%d Right:%d Bottom:%d Top:%d\n",
d->bcLeft,
d->bcRight,
d->bcBottom,
d->bcTop);
printf("\tReynolds number: %.2f\n", d->re);
printf("\tGx Gy: %.2f %.2f\n", d->gx, d->gy);
printf("Geometry data:\n");
printf("\tDomain box size (x, y): %.2f, %.2f\n",
d->grid.xlength,
d->grid.ylength);
printf("\tCells (x, y): %d, %d\n", d->grid.imax, d->grid.jmax);
printf("\tCell size (dx, dy): %f, %f\n", d->grid.dx, d->grid.dy);
printf("Timestep parameters:\n");
printf("\tDefault stepsize: %.2f, Final time %.2f\n", d->dt, d->te);
printf("\tdt bound: %.6f\n", d->dtBound);
printf("\tTau factor: %.2f\n", d->tau);
printf("Iterative s parameters:\n");
printf("\tgamma factor: %f\n", d->gamma);
}
commPrintConfig(&d->comm);
}
void initDiscretiztion(Discretization* d, Parameter* params)
{
d->problem = params->name;
d->bcLeft = params->bcLeft;
d->bcRight = params->bcRight;
d->bcBottom = params->bcBottom;
d->bcTop = params->bcTop;
d->grid.imax = params->imax;
d->grid.jmax = params->jmax;
d->grid.xlength = params->xlength;
d->grid.ylength = params->ylength;
d->grid.dx = params->xlength / params->imax;
d->grid.dy = params->ylength / params->jmax;
d->re = params->re;
d->gx = params->gx;
d->gy = params->gy;
d->dt = params->dt;
d->te = params->te;
d->tau = params->tau;
d->gamma = params->gamma;
/* allocate arrays */
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
size_t size = (imaxLocal + 2) * (jmaxLocal + 2);
d->u = allocate(64, size * sizeof(double));
d->v = allocate(64, size * sizeof(double));
d->p = allocate(64, size * sizeof(double));
d->rhs = allocate(64, size * sizeof(double));
d->f = allocate(64, size * sizeof(double));
d->g = allocate(64, size * sizeof(double));
d->grid.s = allocate(64, size * sizeof(double));
for (int i = 0; i < size; i++) {
d->u[i] = params->u_init;
d->v[i] = params->v_init;
d->p[i] = params->p_init;
d->rhs[i] = 0.0;
d->f[i] = 0.0;
d->g[i] = 0.0;
d->grid.s[i] = FLUID;
}
double dx = d->grid.dx;
double dy = d->grid.dy;
double invSqrSum = 1.0 / (dx * dx) + 1.0 / (dy * dy);
d->dtBound = 0.5 * d->re * 1.0 / invSqrSum;
d->xLocal = d->comm.imaxLocal * d->grid.dx;
d->yLocal = d->comm.jmaxLocal * d->grid.dy;
double xLocal[d->comm.size];
double yLocal[d->comm.size];
#ifdef _MPI
MPI_Allgather(&d->xLocal, 1, MPI_DOUBLE, xLocal, 1, MPI_DOUBLE, d->comm.comm);
MPI_Allgather(&d->yLocal, 1, MPI_DOUBLE, yLocal, 1, MPI_DOUBLE, d->comm.comm);
#endif
d->xOffset = sumOffset(xLocal,
d->comm.rank,
d->comm.dims[JDIM],
d->comm.coords[IDIM]);
d->yOffset = sumOffset(yLocal, d->comm.rank, 1, d->comm.coords[JDIM]);
d->xOffsetEnd = d->xOffset + d->xLocal;
d->yOffsetEnd = d->yOffset + d->yLocal;
printf("Rank : %d, xOffset : %.2f, yOffset : %.2f, xOffsetEnd : %.2f, yOffsetEnd : "
"%.2f\n",
d->comm.rank,
d->xOffset,
d->yOffset,
d->xOffsetEnd,
d->yOffsetEnd);
double* s = d->grid.s;
int iOffset = 0, jOffset = 0;
double xCenter = 0, yCenter = 0, radius = 0;
double x1 = 0, x2 = 0, y1 = 0, y2 = 0;
switch (params->shape) {
case NOSHAPE:
break;
case RECT:
x1 = params->xCenter - params->xRectLength / 2;
x2 = params->xCenter + params->xRectLength / 2;
y1 = params->yCenter - params->yRectLength / 2;
y2 = params->yCenter + params->yRectLength / 2;
iOffset = d->xOffset / dx;
jOffset = d->yOffset / dy;
for (int j = 1; j < jmaxLocal + 1; ++j) {
for (int i = 1; i < imaxLocal + 1; ++i) {
if ((x1 <= ((i + iOffset) * dx)) && (((i + iOffset) * dx) <= x2) &&
(y1 <= ((j + jOffset) * dy)) && (((j + jOffset) * dy) <= y2)) {
S(i, j) = OBSTACLE;
}
}
}
break;
case CIRCLE:
xCenter = params->xCenter;
yCenter = params->yCenter;
radius = params->circleRadius;
iOffset = d->xOffset / dx;
jOffset = d->yOffset / dy;
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = 1; i < imaxLocal + 1; i++) {
if (distance(((i + iOffset) * dx),
((j + jOffset) * dy),
xCenter,
yCenter) <= radius) {
S(i, j) = OBSTACLE;
}
}
}
break;
default:
break;
}
#ifdef _MPI
commExchange(&d->comm, s);
#endif
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = 1; i < imaxLocal + 1; i++) {
if (S(i, j - 1) == FLUID && S(i, j + 1) == OBSTACLE && S(i, j) == OBSTACLE)
S(i, j) = BOTTOM; // BOTTOM
if (S(i - 1, j) == FLUID && S(i + 1, j) == OBSTACLE && S(i, j) == OBSTACLE)
S(i, j) = LEFT; // LEFT
if (S(i + 1, j) == FLUID && S(i - 1, j) == OBSTACLE && S(i, j) == OBSTACLE)
S(i, j) = RIGHT; // RIGHT
if (S(i, j + 1) == FLUID && S(i, j - 1) == OBSTACLE && S(i, j) == OBSTACLE)
S(i, j) = TOP; // TOP
if (S(i - 1, j - 1) == FLUID && S(i, j - 1) == FLUID &&
S(i - 1, j) == FLUID && S(i + 1, j + 1) == OBSTACLE &&
(S(i, j) == OBSTACLE || S(i, j) == LEFT || S(i, j) == BOTTOM))
S(i, j) = BOTTOMLEFT; // BOTTOMLEFT
if (S(i + 1, j - 1) == FLUID && S(i, j - 1) == FLUID &&
S(i + 1, j) == FLUID && S(i - 1, j + 1) == OBSTACLE &&
(S(i, j) == OBSTACLE || S(i, j) == RIGHT || S(i, j) == BOTTOM))
S(i, j) = BOTTOMRIGHT; // BOTTOMRIGHT
if (S(i - 1, j + 1) == FLUID && S(i - 1, j) == FLUID &&
S(i, j + 1) == FLUID && S(i + 1, j - 1) == OBSTACLE &&
(S(i, j) == OBSTACLE || S(i, j) == LEFT || S(i, j) == TOP))
S(i, j) = TOPLEFT; // TOPLEFT
if (S(i + 1, j + 1) == FLUID && S(i + 1, j) == FLUID &&
S(i, j + 1) == FLUID && S(i - 1, j - 1) == OBSTACLE &&
(S(i, j) == OBSTACLE || S(i, j) == RIGHT || S(i, j) == TOP))
S(i, j) = TOPRIGHT; // TOPRIGHT
}
}
#ifdef VERBOSE
printConfig(d);
#endif
}
void computeRHS(Discretization* d)
{
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
double idx = 1.0 / d->grid.dx;
double idy = 1.0 / d->grid.dy;
double idt = 1.0 / d->dt;
double* rhs = d->rhs;
double* f = d->f;
double* g = d->g;
commShift(&d->comm, f, g);
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = 1; i < imaxLocal + 1; i++) {
RHS(i, j) = ((F(i, j) - F(i - 1, j)) * idx + (G(i, j) - G(i, j - 1)) * idy) *
idt;
}
}
}
static double maxElement(Discretization* d, double* m)
{
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
int size = (imaxLocal + 2) * (jmaxLocal + 2);
double maxval = DBL_MIN;
for (int i = 0; i < size; i++) {
maxval = MAX(maxval, fabs(m[i]));
}
commReduction(&maxval, MAX);
return maxval;
}
void computeTimestep(Discretization* d)
{
double dt = d->dtBound;
double dx = d->grid.dx;
double dy = d->grid.dy;
double umax = maxElement(d, d->u);
double vmax = maxElement(d, d->v);
if (umax > 0) {
dt = (dt > dx / umax) ? dx / umax : dt;
}
if (vmax > 0) {
dt = (dt > dy / vmax) ? dy / vmax : dt;
}
d->dt = dt * d->tau;
}
void setBoundaryConditions(Discretization* d)
{
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
double* u = d->u;
double* v = d->v;
if (commIsBoundary(&d->comm, T)) {
switch (d->bcTop) {
case NOSLIP:
for (int i = 1; i < imaxLocal + 1; i++) {
V(i, jmaxLocal) = 0.0;
U(i, jmaxLocal + 1) = -U(i, jmaxLocal);
}
break;
case SLIP:
for (int i = 1; i < imaxLocal + 1; i++) {
V(i, jmaxLocal) = 0.0;
U(i, jmaxLocal + 1) = U(i, jmaxLocal);
}
break;
case OUTFLOW:
for (int i = 1; i < imaxLocal + 1; i++) {
U(i, jmaxLocal + 1) = U(i, jmaxLocal);
V(i, jmaxLocal) = V(i, jmaxLocal - 1);
}
break;
case PERIODIC:
break;
}
}
if (commIsBoundary(&d->comm, B)) {
switch (d->bcBottom) {
case NOSLIP:
for (int i = 1; i < imaxLocal + 1; i++) {
V(i, 0) = 0.0;
U(i, 0) = -U(i, 1);
}
break;
case SLIP:
for (int i = 1; i < imaxLocal + 1; i++) {
V(i, 0) = 0.0;
U(i, 0) = U(i, 1);
}
break;
case OUTFLOW:
for (int i = 1; i < imaxLocal + 1; i++) {
U(i, 0) = U(i, 1);
V(i, 0) = V(i, 1);
}
break;
case PERIODIC:
break;
}
}
if (commIsBoundary(&d->comm, R)) {
switch (d->bcRight) {
case NOSLIP:
for (int j = 1; j < jmaxLocal + 1; j++) {
U(imaxLocal, j) = 0.0;
V(imaxLocal + 1, j) = -V(imaxLocal, j);
}
break;
case SLIP:
for (int j = 1; j < jmaxLocal + 1; j++) {
U(imaxLocal, j) = 0.0;
V(imaxLocal + 1, j) = V(imaxLocal, j);
}
break;
case OUTFLOW:
for (int j = 1; j < jmaxLocal + 1; j++) {
U(imaxLocal, j) = U(imaxLocal - 1, j);
V(imaxLocal + 1, j) = V(imaxLocal, j);
}
break;
case PERIODIC:
break;
}
}
if (commIsBoundary(&d->comm, L)) {
switch (d->bcLeft) {
case NOSLIP:
for (int j = 1; j < jmaxLocal + 1; j++) {
U(0, j) = 0.0;
V(0, j) = -V(1, j);
}
break;
case SLIP:
for (int j = 1; j < jmaxLocal + 1; j++) {
U(0, j) = 0.0;
V(0, j) = V(1, j);
}
break;
case OUTFLOW:
for (int j = 1; j < jmaxLocal + 1; j++) {
U(0, j) = U(1, j);
V(0, j) = V(1, j);
}
break;
case PERIODIC:
break;
}
}
}
void setSpecialBoundaryCondition(Discretization* d)
{
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
double* u = d->u;
double* s = d->grid.s;
if (strcmp(d->problem, "dcavity") == 0) {
if (commIsBoundary(&d->comm, T)) {
for (int i = 1; i < imaxLocal + 1; i++) {
U(i, jmaxLocal + 1) = 2.0 - U(i, jmaxLocal);
}
}
} else if (strcmp(d->problem, "canal") == 0) {
if (commIsBoundary(&d->comm, L)) {
double ylength = d->grid.ylength;
double dy = d->grid.dy;
int rest = d->grid.jmax % d->comm.dims[JDIM];
int yc = d->comm.rank * (d->grid.jmax / d->comm.dims[JDIM]) +
MIN(rest, d->comm.rank);
double ys = dy * (yc + 0.5);
double y;
// printf("RANK %d yc: %d ys: %f\n",d->comm.rank, yc, ys);
for (int j = 1; j < jmaxLocal + 1; j++) {
y = ys + dy * (j - 0.5);
U(0, j) = y * (ylength - y) * 4.0 / (ylength * ylength);
}
}
} else if (strcmp(d->problem, "backstep") == 0) {
for (int j = 1; j < jmaxLocal + 1; j++) {
if (S(0, j) == FLUID) U(0, j) = 1.0;
}
} else if (strcmp(d->problem, "karman") == 0) {
for (int j = 1; j < jmaxLocal + 1; j++) {
U(0, j) = 1.0;
}
}
/* print(solver, solver->u); */
}
void setObjectBoundaryCondition(Discretization* d)
{
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
double* u = d->u;
double* v = d->v;
double* s = d->grid.s;
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = 1; i < imaxLocal + 1; i++) {
switch ((int)S(i, j)) {
case TOP:
U(i, j) = -U(i, j + 1);
U(i - 1, j) = -U(i - 1, j + 1);
V(i, j) = 0.0;
break;
case BOTTOM:
U(i, j) = -U(i, j - 1);
U(i - 1, j) = -U(i - 1, j - 1);
V(i, j) = 0.0;
break;
case LEFT:
U(i - 1, j) = 0.0;
V(i, j) = -V(i - 1, j);
V(i, j - 1) = -V(i - 1, j - 1);
break;
case RIGHT:
U(i, j) = 0.0;
V(i, j) = -V(i + 1, j);
V(i, j - 1) = -V(i + 1, j - 1);
break;
case TOPLEFT:
U(i, j) = -U(i, j + 1);
U(i - 1, j) = 0.0;
V(i, j) = 0.0;
V(i, j - 1) = -V(i - 1, j - 1);
break;
case TOPRIGHT:
U(i, j) = 0.0;
U(i - 1, j) = -U(i - 1, j + 1);
V(i, j) = 0.0;
V(i, j - 1) = -V(i + 1, j - 1);
break;
case BOTTOMLEFT:
U(i, j) = -U(i, j - 1);
U(i - 1, j) = 0.0;
V(i, j) = -V(i - 1, j);
V(i, j - 1) = 0.0;
break;
case BOTTOMRIGHT:
U(i, j) = 0.0;
U(i - 1, j) = -U(i - 1, j - 1);
V(i, j) = -V(i, j + 1);
V(i, j - 1) = 0.0;
break;
}
}
}
}
void computeFG(Discretization* d)
{
double* u = d->u;
double* v = d->v;
double* f = d->f;
double* g = d->g;
double* s = d->grid.s;
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
double gx = d->gx;
double gy = d->gy;
double gamma = d->gamma;
double dt = d->dt;
double inverseRe = 1.0 / d->re;
double inverseDx = 1.0 / d->grid.dx;
double inverseDy = 1.0 / d->grid.dy;
double du2dx, dv2dy, duvdx, duvdy;
double du2dx2, du2dy2, dv2dx2, dv2dy2;
commExchange(&d->comm, u);
commExchange(&d->comm, v);
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = 1; i < imaxLocal + 1; i++) {
if (S(i, j) == FLUID) {
du2dx = inverseDx * 0.25 *
((U(i, j) + U(i + 1, j)) * (U(i, j) + U(i + 1, j)) -
(U(i, j) + U(i - 1, j)) * (U(i, j) + U(i - 1, j))) +
gamma * inverseDx * 0.25 *
(fabs(U(i, j) + U(i + 1, j)) * (U(i, j) - U(i + 1, j)) +
fabs(U(i, j) + U(i - 1, j)) * (U(i, j) - U(i - 1, j)));
duvdy = inverseDy * 0.25 *
((V(i, j) + V(i + 1, j)) * (U(i, j) + U(i, j + 1)) -
(V(i, j - 1) + V(i + 1, j - 1)) *
(U(i, j) + U(i, j - 1))) +
gamma * inverseDy * 0.25 *
(fabs(V(i, j) + V(i + 1, j)) * (U(i, j) - U(i, j + 1)) +
fabs(V(i, j - 1) + V(i + 1, j - 1)) *
(U(i, j) - U(i, j - 1)));
du2dx2 = inverseDx * inverseDx *
(U(i + 1, j) - 2.0 * U(i, j) + U(i - 1, j));
du2dy2 = inverseDy * inverseDy *
(U(i, j + 1) - 2.0 * U(i, j) + U(i, j - 1));
F(i, j) = U(i, j) +
dt * (inverseRe * (du2dx2 + du2dy2) - du2dx - duvdy + gx);
duvdx = inverseDx * 0.25 *
((U(i, j) + U(i, j + 1)) * (V(i, j) + V(i + 1, j)) -
(U(i - 1, j) + U(i - 1, j + 1)) *
(V(i, j) + V(i - 1, j))) +
gamma * inverseDx * 0.25 *
(fabs(U(i, j) + U(i, j + 1)) * (V(i, j) - V(i + 1, j)) +
fabs(U(i - 1, j) + U(i - 1, j + 1)) *
(V(i, j) - V(i - 1, j)));
dv2dy = inverseDy * 0.25 *
((V(i, j) + V(i, j + 1)) * (V(i, j) + V(i, j + 1)) -
(V(i, j) + V(i, j - 1)) * (V(i, j) + V(i, j - 1))) +
gamma * inverseDy * 0.25 *
(fabs(V(i, j) + V(i, j + 1)) * (V(i, j) - V(i, j + 1)) +
fabs(V(i, j) + V(i, j - 1)) * (V(i, j) - V(i, j - 1)));
dv2dx2 = inverseDx * inverseDx *
(V(i + 1, j) - 2.0 * V(i, j) + V(i - 1, j));
dv2dy2 = inverseDy * inverseDy *
(V(i, j + 1) - 2.0 * V(i, j) + V(i, j - 1));
G(i, j) = V(i, j) +
dt * (inverseRe * (dv2dx2 + dv2dy2) - duvdx - dv2dy + gy);
} else {
switch ((int)S(i, j)) {
case TOP:
G(i, j) = V(i, j);
break;
case BOTTOM:
G(i, j - 1) = V(i, j - 1);
break;
case LEFT:
F(i - 1, j) = U(i - 1, j);
break;
case RIGHT:
F(i, j) = U(i, j);
break;
case TOPLEFT:
F(i - 1, j) = U(i - 1, j);
G(i, j) = V(i, j);
break;
case TOPRIGHT:
F(i, j) = U(i, j);
G(i, j) = V(i, j);
break;
case BOTTOMLEFT:
F(i - 1, j) = U(i - 1, j);
G(i, j - 1) = V(i, j - 1);
break;
case BOTTOMRIGHT:
F(i, j) = U(i, j);
G(i, j - 1) = V(i, j - 1);
break;
}
}
}
}
/* ----------------------------- boundary of F --------------------------- */
if (commIsBoundary(&d->comm, L)) {
for (int j = 1; j < jmaxLocal + 1; j++) {
F(0, j) = U(0, j);
}
}
if (commIsBoundary(&d->comm, R)) {
for (int j = 1; j < jmaxLocal + 1; j++) {
F(imaxLocal, j) = U(imaxLocal, j);
}
}
/* ----------------------------- boundary of G --------------------------- */
if (commIsBoundary(&d->comm, B)) {
for (int i = 1; i < imaxLocal + 1; i++) {
G(i, 0) = V(i, 0);
}
}
if (commIsBoundary(&d->comm, T)) {
for (int i = 1; i < imaxLocal + 1; i++) {
G(i, jmaxLocal) = V(i, jmaxLocal);
}
}
}
void adaptUV(Discretization* d)
{
int imaxLocal = d->comm.imaxLocal;
int jmaxLocal = d->comm.jmaxLocal;
double* p = d->p;
double* u = d->u;
double* v = d->v;
double* f = d->f;
double* g = d->g;
double factorX = d->dt / d->grid.dx;
double factorY = d->dt / d->grid.dy;
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = 1; i < imaxLocal + 1; i++) {
U(i, j) = F(i, j) - (P(i + 1, j) - P(i, j)) * factorX;
V(i, j) = G(i, j) - (P(i, j + 1) - P(i, j)) * factorY;
}
}
}
void writeResult(Discretization* d, double* u, double* v, double* p)
{
int imax = d->grid.imax;
int jmax = d->grid.jmax;
double dx = d->grid.dx;
double dy = d->grid.dy;
double x = 0.0, y = 0.0;
FILE* fp;
fp = fopen("pressure.dat", "w");
if (fp == NULL) {
printf("Error!\n");
exit(EXIT_FAILURE);
}
for (int j = 1; j <= jmax; j++) {
y = (double)(j - 0.5) * dy;
for (int i = 1; i <= imax; i++) {
x = (double)(i - 0.5) * dx;
fprintf(fp, "%.2f %.2f %f\n", x, y, p[j * (imax + 2) + i]);
}
fprintf(fp, "\n");
}
fclose(fp);
fp = fopen("velocity.dat", "w");
if (fp == NULL) {
printf("Error!\n");
exit(EXIT_FAILURE);
}
for (int j = 1; j <= jmax; j++) {
y = dy * (j - 0.5);
for (int i = 1; i <= imax; i++) {
x = dx * (i - 0.5);
double velU = (u[j * (imax + 2) + i] + u[j * (imax + 2) + (i - 1)]) / 2.0;
double velV = (v[j * (imax + 2) + i] + v[(j - 1) * (imax + 2) + i]) / 2.0;
double len = sqrt((velU * velU) + (velV * velV));
fprintf(fp, "%.2f %.2f %f %f %f\n", x, y, velU, velV, len);
}
}
fclose(fp);
}

66
EnhancedSolver/2D-mpi/src/discretization.h Normal file
View File

@@ -0,0 +1,66 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#ifndef __DISCRETIZATION_H_
#define __DISCRETIZATION_H_
#include "comm.h"
#include "grid.h"
#include "parameter.h"
#include<unistd.h>
enum BC { NOSLIP = 1, SLIP, OUTFLOW, PERIODIC };
enum OBJECTBOUNDARY {
FLUID = 0,
TOP,
BOTTOM,
LEFT,
RIGHT,
TOPLEFT,
BOTTOMLEFT,
TOPRIGHT,
BOTTOMRIGHT,
OBSTACLE
};
enum SHAPE { NOSHAPE = 0, RECT, CIRCLE };
typedef struct {
/* geometry and grid information */
Grid grid;
/* arrays */
double *p, *rhs;
double *f, *g;
double *u, *v;
/* parameters */
double re, tau, gamma;
double gx, gy;
/* time stepping */
double dt, te;
double dtBound;
char* problem;
double xLocal, yLocal, xOffset, yOffset, xOffsetEnd, yOffsetEnd;
int bcLeft, bcRight, bcBottom, bcTop;
/* communication */
Comm comm;
} Discretization;
extern void initDiscretiztion(Discretization*, Parameter*);
extern void computeRHS(Discretization*);
extern void normalizePressure(Discretization*);
extern void computeTimestep(Discretization*);
extern void setBoundaryConditions(Discretization*);
extern void setSpecialBoundaryCondition(Discretization*);
extern void setObjectBoundaryCondition(Discretization*);
extern void computeFG(Discretization*);
extern void adaptUV(Discretization*);
extern void writeResult(Discretization* s, double* u, double* v, double* p);
extern double sumOffset(double* , int , int , int );
extern void print(Discretization* , double* );
#endif

17
EnhancedSolver/2D-mpi/src/grid.h Normal file
View File

@@ -0,0 +1,17 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#ifndef __GRID_H_
#define __GRID_H_
typedef struct {
double dx, dy;
int imax, jmax;
double xlength, ylength;
double* s;
} Grid;
#endif // __GRID_H_

54
EnhancedSolver/2D-mpi/src/likwid-marker.h Normal file
View File

@@ -0,0 +1,54 @@
/*
* =======================================================================================
*
* Author: Jan Eitzinger (je), jan.eitzinger@fau.de
* Copyright (c) 2020 RRZE, University Erlangen-Nuremberg
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*
* =======================================================================================
*/
#ifndef LIKWID_MARKERS_H
#define LIKWID_MARKERS_H
#ifdef LIKWID_PERFMON
#include <likwid.h>
#define LIKWID_MARKER_INIT likwid_markerInit()
#define LIKWID_MARKER_THREADINIT likwid_markerThreadInit()
#define LIKWID_MARKER_SWITCH likwid_markerNextGroup()
#define LIKWID_MARKER_REGISTER(regionTag) likwid_markerRegisterRegion(regionTag)
#define LIKWID_MARKER_START(regionTag) likwid_markerStartRegion(regionTag)
#define LIKWID_MARKER_STOP(regionTag) likwid_markerStopRegion(regionTag)
#define LIKWID_MARKER_CLOSE likwid_markerClose()
#define LIKWID_MARKER_RESET(regionTag) likwid_markerResetRegion(regionTag)
#define LIKWID_MARKER_GET(regionTag, nevents, events, time, count) \
likwid_markerGetRegion(regionTag, nevents, events, time, count)
#else /* LIKWID_PERFMON */
#define LIKWID_MARKER_INIT
#define LIKWID_MARKER_THREADINIT
#define LIKWID_MARKER_SWITCH
#define LIKWID_MARKER_REGISTER(regionTag)
#define LIKWID_MARKER_START(regionTag)
#define LIKWID_MARKER_STOP(regionTag)
#define LIKWID_MARKER_CLOSE
#define LIKWID_MARKER_GET(regionTag, nevents, events, time, count)
#define LIKWID_MARKER_RESET(regionTag)
#endif /* LIKWID_PERFMON */
#endif /*LIKWID_MARKERS_H*/

120
EnhancedSolver/2D-mpi/src/main.c Normal file
View File

@@ -0,0 +1,120 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include "allocate.h"
#include "comm.h"
#include "discretization.h"
#include "grid.h"
#include "parameter.h"
#include "particletracing.h"
#include "progress.h"
#include "timing.h"
static void writeResults(Discretization* s)
{
#ifdef _MPI
size_t bytesize = (s->grid.imax + 2) * (s->grid.jmax + 2) * sizeof(double);
double* ug = allocate(64, bytesize);
double* vg = allocate(64, bytesize);
double* pg = allocate(64, bytesize);
commCollectResult(&s->comm, ug, vg, pg, s->u, s->v, s->p, s->grid.imax, s->grid.jmax);
if (commIsMaster(&s->comm)) {
writeResult(s, ug, vg, pg);
}
free(ug);
free(vg);
free(pg);
#else
writeResult(s, s->u, s->v, s->p);
#endif
}
int main(int argc, char** argv)
{
int rank;
double timeStart, timeStop;
Parameter p;
Discretization d;
Solver s;
ParticleTracer particletracer;
commInit(&d.comm, argc, argv);
initParameter(&p);
if (argc != 2) {
printf("Usage: %s <configFile>\n", argv[0]);
exit(EXIT_SUCCESS);
}
readParameter(&p, argv[1]);
commPartition(&d.comm, p.jmax, p.imax);
if (commIsMaster(&d.comm)) {
printParameter(&p);
}
initDiscretiztion(&d, &p);
initSolver(&s, &d, &p);
initParticleTracer(&particletracer, &d, &p);
#ifdef TEST
commPrintConfig(&d.comm);
commTestInit(&d.comm, d.p, d.f, d.g);
commExchange(&d.comm, d.p);
commShift(&d.comm, d.f, d.g);
commTestWrite(&d.comm, d.p, d.f, d.g);
writeResults(&d);
commFinalize(&d.comm);
exit(EXIT_SUCCESS);
#endif
#ifndef VERBOSE
initProgress(d.te);
#endif
double tau = d.tau;
double te = d.te;
double t = 0.0;
timeStart = getTimeStamp();
while (t <= te) {
if (tau > 0.0) computeTimestep(&d);
setBoundaryConditions(&d);
setSpecialBoundaryCondition(&d);
setObjectBoundaryCondition(&d);
computeFG(&d);
computeRHS(&d);
solve(&s, d.p, d.rhs);
adaptUV(&d);
trace(&particletracer, &d, t);
t += d.dt;
#ifdef VERBOSE
if (commIsMaster(s.comm)) {
printf("TIME %f , TIMESTEP %f\n", t, d.dt);
}
#else
printProgress(t);
#endif
}
timeStop = getTimeStamp();
#ifndef VERBOSE
stopProgress();
#endif
if (commIsMaster(s.comm)) {
printf("Solution took %.2fs\n", timeStop - timeStart);
}
freeParticles(&particletracer);
writeResults(&d);
commFinalize(s.comm);
return EXIT_SUCCESS;
}

143
EnhancedSolver/2D-mpi/src/parameter.c Normal file
View File

@@ -0,0 +1,143 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "parameter.h"
#include "util.h"
#define MAXLINE 4096
void initParameter(Parameter* param)
{
param->xlength = 1.0;
param->ylength = 1.0;
param->imax = 100;
param->jmax = 100;
param->itermax = 1000;
param->eps = 0.0001;
param->omg = 1.8;
param->levels = 5;
param->presmooth = 5;
param->postsmooth = 5;
}
void readParameter(Parameter* param, const char* filename)
{
FILE* fp = fopen(filename, "r");
char line[MAXLINE];
int i;
if (!fp) {
fprintf(stderr, "Could not open parameter file: %s\n", filename);
exit(EXIT_FAILURE);
}
while (!feof(fp)) {
line[0] = '\0';
fgets(line, MAXLINE, fp);
for (i = 0; line[i] != '\0' && line[i] != '#'; i++)
;
line[i] = '\0';
char* tok = strtok(line, " ");
char* val = strtok(NULL, " ");
#define PARSE_PARAM(p, f) \
if (strncmp(tok, #p, sizeof(#p) / sizeof(#p[0]) - 1) == 0) { \
param->p = f(val); \
}
#define PARSE_STRING(p) PARSE_PARAM(p, strdup)
#define PARSE_INT(p) PARSE_PARAM(p, atoi)
#define PARSE_REAL(p) PARSE_PARAM(p, atof)
if (tok != NULL && val != NULL) {
PARSE_REAL(xlength);
PARSE_REAL(ylength);
PARSE_INT(imax);
PARSE_INT(jmax);
PARSE_INT(itermax);
PARSE_REAL(eps);
PARSE_REAL(omg);
PARSE_REAL(re);
PARSE_REAL(tau);
PARSE_REAL(gamma);
PARSE_REAL(dt);
PARSE_REAL(te);
PARSE_REAL(gx);
PARSE_REAL(gy);
PARSE_STRING(name);
PARSE_INT(bcLeft);
PARSE_INT(bcRight);
PARSE_INT(bcBottom);
PARSE_INT(bcTop);
PARSE_INT(levels);
PARSE_INT(presmooth);
PARSE_INT(postsmooth);
PARSE_REAL(u_init);
PARSE_REAL(v_init);
PARSE_REAL(p_init);
/* Added new particle tracing parameters */
PARSE_INT(numberOfParticles);
PARSE_REAL(startTime);
PARSE_REAL(injectTimePeriod);
PARSE_REAL(writeTimePeriod);
PARSE_REAL(x1);
PARSE_REAL(y1);
PARSE_REAL(x2);
PARSE_REAL(y2);
/* Added obstacle geometry parameters */
PARSE_INT(shape);
PARSE_REAL(xCenter);
PARSE_REAL(yCenter);
PARSE_REAL(xRectLength);
PARSE_REAL(yRectLength);
PARSE_REAL(circleRadius);
}
}
fclose(fp);
}
void printParameter(Parameter* param)
{
printf("Parameters for %s\n", param->name);
printf("Boundary conditions Left:%d Right:%d Bottom:%d Top:%d\n",
param->bcLeft,
param->bcRight,
param->bcBottom,
param->bcTop);
printf("\tReynolds number: %.2f\n", param->re);
printf("\tInit arrays: U:%.2f V:%.2f P:%.2f\n",
param->u_init,
param->v_init,
param->p_init);
printf("Geometry data:\n");
printf("\tDomain box size (x, y): %.2f, %.2f\n", param->xlength, param->ylength);
printf("\tCells (x, y): %d, %d\n", param->imax, param->jmax);
printf("Timestep parameters:\n");
printf("\tDefault stepsize: %.2f, Final time %.2f\n", param->dt, param->te);
printf("\tTau factor: %.2f\n", param->tau);
printf("Iterative solver parameters:\n");
printf("\tMax iterations: %d\n", param->itermax);
printf("\tepsilon (stopping tolerance) : %f\n", param->eps);
printf("\tgamma (stopping tolerance) : %f\n", param->gamma);
printf("\tomega (SOR relaxation): %f\n", param->omg);
printf("Particle tracer parameters:\n");
printf("\tNumber of particles: %d\n", param->numberOfParticles);
printf("\tstartTime : %f\n", param->startTime);
printf("\tinjecTimePeriod : %f\n", param->injectTimePeriod);
printf("\twriteTimePeriod: %f\n", param->writeTimePeriod);
printf("\tx1 : %f, x2 : %f, y1 : %f, y2 : %f\n",
param->x1,
param->x2,
param->y1,
param->y2);
printf("\tShape : %d\n", param->shape);
}

38
EnhancedSolver/2D-mpi/src/parameter.h Normal file
View File

@@ -0,0 +1,38 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#ifndef __PARAMETER_H_
#define __PARAMETER_H_
typedef struct {
double xlength, ylength;
int imax, jmax;
int itermax;
double eps, omg;
double re, tau, gamma;
double te, dt;
double gx, gy;
char* name;
int bcLeft, bcRight, bcBottom, bcTop;
double u_init, v_init, p_init;
int levels, presmooth, postsmooth;
int numberOfParticles;
double startTime, injectTimePeriod, writeTimePeriod;
double x1, y1, x2, y2;
int shape;
double xCenter, yCenter, xRectLength, yRectLength, circleRadius;
} Parameter;
void initParameter(Parameter*);
void readParameter(Parameter*, const char*);
void printParameter(Parameter*);
#endif

612
EnhancedSolver/2D-mpi/src/particletracing.c Normal file
View File

@@ -0,0 +1,612 @@
/*
* Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <float.h>
#include <math.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "particletracing.h"
#define U(i, j) u[(j) * (imaxLocal + 2) + (i)]
#define V(i, j) v[(j) * (imaxLocal + 2) + (i)]
#define S(i, j) s[(j) * (imaxLocal + 2) + (i)]
static int ts = 0;
#define IDIM 0
#define JDIM 1
#define XOFFSET 0
#define YOFFSET 1
#define XOFFSETEND 2
#define YOFFSETEND 3
static double sum(int* sizes, int size)
{
double sum = 0;
for (int i = 0; i < size; ++i) {
sum += sizes[i];
}
return sum;
}
void printUV(ParticleTracer* particletracer, double* u, double* v)
{
int imaxLocal = particletracer->imaxLocal;
for (int i = 0; i < particletracer->size; i++) {
if (i == particletracer->rank) {
printf(
"\n### RANK %d #######################################################\n",
particletracer->rank);
printf("\nGrid U : \n");
for (int j = 0; j < particletracer->jmaxLocal + 2; j++) {
printf("%02d: ", j);
for (int i = 0; i < particletracer->imaxLocal + 2; i++) {
printf("%4.2f ", u[j * (imaxLocal + 2) + i]);
}
printf("\n");
}
fflush(stdout);
printf("\nGrid V : \n");
for (int j = 0; j < particletracer->jmaxLocal + 2; j++) {
printf("%02d: ", j);
for (int i = 0; i < particletracer->imaxLocal + 2; i++) {
printf("%4.2f ", v[j * (imaxLocal + 2) + i]);
}
printf("\n");
}
fflush(stdout);
}
#ifdef _MPI
MPI_Barrier(MPI_COMM_WORLD);
#endif
}
}
void initParticleTracer(
ParticleTracer* particletracer, Discretization* d, Parameter* params)
{
int dims[NDIMS] = { 0, 0 };
int periods[NDIMS] = { 0, 0 };
/* initializing local properties from params */
particletracer->rank = d->comm.rank;
particletracer->size = d->comm.size;
particletracer->numberOfParticles = params->numberOfParticles;
particletracer->startTime = params->startTime;
particletracer->injectTimePeriod = params->injectTimePeriod;
particletracer->writeTimePeriod = params->writeTimePeriod;
particletracer->dt = params->dt;
particletracer->dx = params->xlength / params->imax;
particletracer->dy = params->ylength / params->jmax;
particletracer->xlength = params->xlength;
particletracer->ylength = params->ylength;
particletracer->x1 = params->x1;
particletracer->y1 = params->y1;
particletracer->x2 = params->x2;
particletracer->y2 = params->y2;
particletracer->lastInjectTime = params->startTime;
particletracer->lastUpdateTime = params->startTime;
particletracer->lastWriteTime = params->startTime;
particletracer->pointer = 0;
particletracer->totalParticles = 0;
particletracer->imax = params->imax;
particletracer->jmax = params->jmax;
particletracer->imaxLocal = d->comm.imaxLocal;
particletracer->jmaxLocal = d->comm.jmaxLocal;
particletracer->removedParticles = 0;
// Estimating the number of particles over the number of timesteps that could be
// required.
particletracer->estimatedNumParticles = (particletracer->imaxLocal *
particletracer->jmaxLocal);
// Allocating memory for the estimated particles over the timesteps.
particletracer->particlePool = malloc(
sizeof(Particle) * particletracer->estimatedNumParticles);
// Initializing the number of particles to 0 and turning OFF all of the particles.
// Contain information in x and y length metric, not i and j discretization metrics.
for (int i = 0; i < particletracer->estimatedNumParticles; ++i) {
particletracer->particlePool[i].x = 0.0;
particletracer->particlePool[i].y = 0.0;
particletracer->particlePool[i].flag = false;
}
// Creating a linearly spaced particle line.
particletracer->linSpaceLine = malloc(
sizeof(Particle) * particletracer->numberOfParticles);
// Creating an array for each rank that will hold information about
// offsets from other ranks. Holds each ranks x and y length metrics.
particletracer->offset = (double*)malloc(sizeof(double) * 4 * particletracer->size);
// Calculating each ranks local x and y length metrics.
double offset[4][particletracer->size];
// Calculating each ranks x and y local lengths.
particletracer->xLocal = d->xLocal;
particletracer->yLocal = d->yLocal;
double xLocal[particletracer->size];
double yLocal[particletracer->size];
// Calculate own x and y length metric offset based on other ranks offset data.
particletracer->xOffset = d->xOffset;
particletracer->yOffset = d->yOffset;
particletracer->xOffsetEnd = d->xOffsetEnd;
particletracer->yOffsetEnd = d->yOffsetEnd;
printf("Rank : %d, xOffset : %.2f, yOffset : %.2f, xOffsetEnd : %.2f, yOffsetEnd : "
"%.2f\n",
particletracer->rank,
particletracer->xOffset,
particletracer->yOffset,
particletracer->xOffsetEnd,
particletracer->yOffsetEnd);
#ifdef _MPI
// Gather each ranks x and y length metric that marks each ranks own territory.
// Once the boundary leaves local domain, then it needs to know which ranks to send.
// And to know whos boundary it is, we need to know the rank.
MPI_Allgather(&particletracer->xOffset,
1,
MPI_DOUBLE,
offset[0],
1,
MPI_DOUBLE,
d->comm.comm);
MPI_Allgather(&particletracer->yOffset,
1,
MPI_DOUBLE,
offset[1],
1,
MPI_DOUBLE,
d->comm.comm);
MPI_Allgather(&particletracer->xOffsetEnd,
1,
MPI_DOUBLE,
offset[2],
1,
MPI_DOUBLE,
d->comm.comm);
MPI_Allgather(&particletracer->yOffsetEnd,
1,
MPI_DOUBLE,
offset[3],
1,
MPI_DOUBLE,
d->comm.comm);
#endif
memcpy(particletracer->offset, offset, sizeof(offset));
for (int i = 0; i < particletracer->numberOfParticles; ++i) {
double spacing = (double)i / (double)(particletracer->numberOfParticles - 1);
particletracer->linSpaceLine[i].x = spacing * particletracer->x1 +
(1.0 - spacing) * particletracer->x2;
particletracer->linSpaceLine[i].y = spacing * particletracer->y1 +
(1.0 - spacing) * particletracer->y2;
particletracer->linSpaceLine[i].flag = true;
}
#ifdef _MPI
// Create the mpi_particle datatype
MPI_Datatype mpi_particle;
int lengths[3] = { 1, 1, 1 };
MPI_Aint displacements[3];
Particle dummy_particle;
MPI_Aint base_address;
MPI_Get_address(&dummy_particle, &base_address);
MPI_Get_address(&dummy_particle.x, &displacements[0]);
MPI_Get_address(&dummy_particle.y, &displacements[1]);
MPI_Get_address(&dummy_particle.flag, &displacements[2]);
displacements[0] = MPI_Aint_diff(displacements[0], base_address);
displacements[1] = MPI_Aint_diff(displacements[1], base_address);
displacements[2] = MPI_Aint_diff(displacements[2], base_address);
MPI_Datatype types[3] = { MPI_DOUBLE, MPI_DOUBLE, MPI_C_BOOL };
MPI_Type_create_struct(3,
lengths,
displacements,
types,
&particletracer->mpi_particle);
MPI_Type_commit(&particletracer->mpi_particle);
#endif
}
void printParticles(ParticleTracer* particletracer)
{
for (int i = 0; i < particletracer->totalParticles; ++i) {
printf("Rank : %d Particle position X : %.2f, Y : %.2f, flag : %d, total pt : "
"%d, pointer : %d, xOffset : %.2f, yOffset : %.2f, xOffsetEnd : %.2f, "
"yOffsetEnd : %.2f\n",
particletracer->rank,
particletracer->particlePool[i].x,
particletracer->particlePool[i].y,
particletracer->particlePool[i].flag,
particletracer->totalParticles,
particletracer->pointer,
particletracer->xOffset,
particletracer->yOffset,
particletracer->xOffsetEnd,
particletracer->yOffsetEnd);
}
}
void injectParticles(ParticleTracer* particletracer)
{
double x, y;
for (int i = 0; i < particletracer->numberOfParticles; ++i) {
x = particletracer->linSpaceLine[i].x;
y = particletracer->linSpaceLine[i].y;
if (x >= particletracer->xOffset && y >= particletracer->yOffset &&
x <= particletracer->xOffsetEnd && y <= particletracer->yOffsetEnd) {
particletracer->particlePool[particletracer->pointer].x = x;
particletracer->particlePool[particletracer->pointer].y = y;
particletracer->particlePool[particletracer->pointer].flag = true;
++(particletracer->pointer);
++(particletracer->totalParticles);
}
}
}
void advanceParticles(ParticleTracer* particletracer,
double* restrict u,
double* restrict v,
double* restrict s,
Comm* comm,
double time)
{
int imax = particletracer->imax;
int jmax = particletracer->jmax;
int imaxLocal = particletracer->imaxLocal;
int jmaxLocal = particletracer->jmaxLocal;
double dx = particletracer->dx;
double dy = particletracer->dy;
double xlength = particletracer->xlength;
double ylength = particletracer->ylength;
Particle buff[particletracer->size][100];
for (int i = 0; i < particletracer->size; ++i) {
for (int j = 0; j < 100; ++j) {
buff[i][j].x = 0.0;
buff[i][j].y = 0.0;
buff[i][j].flag = false;
}
}
int particleBufIndex[particletracer->size];
memset(particleBufIndex, 0, sizeof(particleBufIndex));
for (int i = 0; i < particletracer->totalParticles; ++i) {
if (particletracer->particlePool[i].flag == true) {
double xTemp = particletracer->particlePool[i].x;
double yTemp = particletracer->particlePool[i].y;
double x = xTemp - particletracer->xOffset;
double y = yTemp - particletracer->yOffset;
int iCoord = (int)(x / dx) + 1;
int jCoord = (int)((y + 0.5 * dy) / dy) + 1;
double x1 = (double)(iCoord - 1) * dx;
double y1 = ((double)(jCoord - 1) - 0.5) * dy;
double x2 = (double)iCoord * dx;
double y2 = ((double)jCoord - 0.5) * dy;
double u_n = (1.0 / (dx * dy)) *
((x2 - x) * (y2 - y) * U(iCoord - 1, jCoord - 1) +
(x - x1) * (y2 - y) * U(iCoord, jCoord - 1) +
(x2 - x) * (y - y1) * U(iCoord - 1, jCoord) +
(x - x1) * (y - y1) * U(iCoord, jCoord));
double new_x = (x + particletracer->xOffset) + particletracer->dt * u_n;
particletracer->particlePool[i].x = new_x;
iCoord = (int)((x + 0.5 * dx) / dx) + 1;
jCoord = (int)(y / dy) + 1;
x1 = ((double)(iCoord - 1) - 0.5) * dx;
y1 = (double)(jCoord - 1) * dy;
x2 = ((double)iCoord - 0.5) * dx;
y2 = (double)jCoord * dy;
double v_n = (1.0 / (dx * dy)) *
((x2 - x) * (y2 - y) * V(iCoord - 1, jCoord - 1) +
(x - x1) * (y2 - y) * V(iCoord, jCoord - 1) +
(x2 - x) * (y - y1) * V(iCoord - 1, jCoord) +
(x - x1) * (y - y1) * V(iCoord, jCoord));
double new_y = (y + particletracer->yOffset) + particletracer->dt * v_n;
particletracer->particlePool[i].y = new_y;
if (((new_x < particletracer->xOffset) ||
(new_x >= particletracer->xOffsetEnd) ||
(new_y < particletracer->yOffset) ||
(new_y >= particletracer->yOffsetEnd))) {
// New logic to transfer particles to neighbouring ranks or discard the
// particle.
for (int i = 0; i < particletracer->size; ++i) {
if ((new_x >=
particletracer->offset[i + particletracer->size * XOFFSET]) &&
(new_x <= particletracer
->offset[i + particletracer->size * XOFFSETEND]) &&
(new_y >=
particletracer->offset[i + particletracer->size * YOFFSET]) &&
(new_y <= particletracer
->offset[i + particletracer->size * YOFFSETEND]) &&
i != particletracer->rank) {
buff[i][particleBufIndex[i]].x = new_x;
buff[i][particleBufIndex[i]].y = new_y;
buff[i][particleBufIndex[i]].flag = true;
++particleBufIndex[i];
}
}
particletracer->particlePool[i].flag = false;
particletracer->removedParticles++;
}
int i_new = new_x / dx, j_new = new_y / dy;
int iOffset = particletracer->xOffset / dx,
jOffset = particletracer->yOffset / dy;
if (S(i_new - iOffset, j_new - jOffset) != FLUID) {
particletracer->particlePool[i].flag = false;
particletracer->removedParticles++;
}
}
}
#ifdef _MPI
for (int i = 0; i < particletracer->size; ++i) {
if (i != particletracer->rank) {
MPI_Send(buff[i], 100, particletracer->mpi_particle, i, 0, comm->comm);
}
}
for (int i = 0; i < particletracer->size; ++i) {
if (i != particletracer->rank) {
MPI_Recv(buff[i],
100,
particletracer->mpi_particle,
i,
0,
comm->comm,
MPI_STATUS_IGNORE);
}
}
for (int i = 0; i < particletracer->size; ++i) {
if (i != particletracer->rank) {
for (int j = 0; j < 100; ++j) {
if (buff[i][j].flag == true) {
particletracer->particlePool[particletracer->pointer].x = buff[i][j]
.x;
particletracer->particlePool[particletracer->pointer].y = buff[i][j]
.y;
particletracer->particlePool[particletracer->pointer].flag = true;
++(particletracer->pointer);
++(particletracer->totalParticles);
}
}
}
}
#endif
}
void freeParticles(ParticleTracer* particletracer)
{
free(particletracer->particlePool);
free(particletracer->linSpaceLine);
free(particletracer->offset);
}
void writeParticles(ParticleTracer* particletracer, Comm* comm)
{
int collectedBuffIndex[particletracer->size];
#ifdef _MPI
MPI_Gather(&particletracer->totalParticles,
1,
MPI_INT,
collectedBuffIndex,
1,
MPI_INT,
0,
comm->comm);
if (particletracer->rank != 0) {
Particle buff[particletracer->totalParticles];
for (int i = 0; i < particletracer->totalParticles; ++i) {
buff[i].x = particletracer->particlePool[i].x;
buff[i].y = particletracer->particlePool[i].y;
buff[i].flag = particletracer->particlePool[i].flag;
// printf("Rank : %d sending to rank 0 X : %.2f, Y : %.2f with totalpt :
// %d\n", particletracer->rank, buff[i].x, buff[i].y,
// particletracer->totalParticles);
}
MPI_Send(buff,
particletracer->totalParticles,
particletracer->mpi_particle,
0,
1,
comm->comm);
}
#endif
if (particletracer->rank == 0) {
char filename[50];
FILE* fp;
snprintf(filename, 50, "vis_files/particles_%d.dat", ts);
fp = fopen(filename, "w");
if (fp == NULL) {
printf("Error!\n");
exit(EXIT_FAILURE);
}
// fprintf(fp, "# vtk DataFile Version 3.0\n");
// fprintf(fp, "PAMPI cfd solver particle tracing file\n");
// fprintf(fp, "ASCII\n");
// fprintf(fp, "DATASET UNSTRUCTURED_GRID\n");
// fprintf(fp, "FIELD FieldData 2\n");
// fprintf(fp, "TIME 1 1 double\n");
// fprintf(fp, "%d\n", ts);
// fprintf(fp, "CYCLE 1 1 int\n");
// fprintf(fp, "1\n");
int overallTotalParticles = sum(collectedBuffIndex, particletracer->size);
// fprintf(fp, "POINTS %d float\n", overallTotalParticles);
// printf("Total particles : %d\n", overallTotalParticles);
#ifdef _MPI
for (int i = 1; i < particletracer->size; ++i) {
Particle recvBuff[collectedBuffIndex[i]];
MPI_Recv(&recvBuff,
collectedBuffIndex[i],
particletracer->mpi_particle,
i,
1,
comm->comm,
MPI_STATUS_IGNORE);
for (int j = 0; j < collectedBuffIndex[i]; ++j) {
double x = recvBuff[j].x;
double y = recvBuff[j].y;
fprintf(fp, "%f %f\n", x, y);
// printf("Rank : 0 receiving from rank %d X : %.2f, Y : %.2f with totalpt
// : %d\n", i, x, y, particletracer->totalParticles);
}
}
#endif
for (int i = 0; i < particletracer->totalParticles; ++i) {
double x = particletracer->particlePool[i].x;
double y = particletracer->particlePool[i].y;
fprintf(fp, "%f %f\n", x, y);
}
// fprintf(fp, "CELLS %d %d\n", overallTotalParticles, 2 * overallTotalParticles);
// for (int i = 0; i < overallTotalParticles; ++i)
// {
// fprintf(fp, "1 %d\n", i);
// }
// fprintf(fp, "CELL_TYPES %d\n", overallTotalParticles);
// for (int i = 0; i < overallTotalParticles; ++i)
// {
// fprintf(fp, "1\n");
// }
fclose(fp);
}
++ts;
}
void printParticleTracerParameters(ParticleTracer* particletracer)
{
printf("Particle Tracing data:\n");
printf("Rank : %d\n", particletracer->rank);
printf("\tNumber of particles : %d being injected for every period of %.2f\n",
particletracer->numberOfParticles,
particletracer->injectTimePeriod);
printf("\tstartTime : %.2f\n", particletracer->startTime);
printf("\t(Line along which the particles are to be injected) \n\tx1 : %.2f, y1 : "
"%.2f, x2 : %.2f, y2 : %.2f\n",
particletracer->x1,
particletracer->y1,
particletracer->x2,
particletracer->y2);
printf("\tPointer : %d, TotalParticles : %d\n",
particletracer->pointer,
particletracer->totalParticles);
printf("\tdt : %.2f, dx : %.2f, dy : %.2f\n",
particletracer->dt,
particletracer->dx,
particletracer->dy);
printf("\txOffset : %.2f, yOffset : %.2f\n",
particletracer->xOffset,
particletracer->yOffset);
printf("\txOffsetEnd : %.2f, yOffsetEnd : %.2f\n",
particletracer->xOffsetEnd,
particletracer->yOffsetEnd);
printf("\txLocal : %.2f, yLocal : %.2f\n",
particletracer->xLocal,
particletracer->yLocal);
}
void trace(ParticleTracer* particletracer, Discretization* d, double time)
{
if (time >= particletracer->startTime) {
if ((time - particletracer->lastInjectTime) >= particletracer->injectTimePeriod) {
injectParticles(particletracer);
particletracer->lastInjectTime = time;
}
if ((time - particletracer->lastWriteTime) >= particletracer->writeTimePeriod) {
writeParticles(particletracer, &d->comm);
particletracer->lastWriteTime = time;
}
advanceParticles(particletracer, d->u, d->v, d->grid.s, &d->comm, time);
if (particletracer->removedParticles > (particletracer->totalParticles * 0.2)) {
compress(particletracer);
}
particletracer->lastUpdateTime = time;
}
}
void compress(ParticleTracer* particletracer)
{
Particle* memPool = particletracer->particlePool;
Particle tempPool[particletracer->totalParticles];
int totalParticles = 0;
printf("Performing compression ...");
for (int i = 0; i < particletracer->totalParticles; ++i) {
if (memPool[i].flag == true) {
tempPool[totalParticles].x = memPool[i].x;
tempPool[totalParticles].y = memPool[i].y;
tempPool[totalParticles].flag = memPool[i].flag;
++totalParticles;
}
}
printf(" remove %d particles\n", particletracer->totalParticles - totalParticles);
particletracer->totalParticles = totalParticles;
particletracer->removedParticles = 0;
particletracer->pointer = totalParticles + 1;
memcpy(particletracer->particlePool, tempPool, totalParticles * sizeof(Particle));
}

64
EnhancedSolver/2D-mpi/src/particletracing.h Normal file
View File

@@ -0,0 +1,64 @@
/*
* Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#ifndef __PARTICLETRACING_H_
#define __PARTICLETRACING_H_
#include "allocate.h"
#include "parameter.h"
#include "solver.h"
#include <mpi.h>
#include <stdbool.h>
#define NDIMS 2
typedef enum COORD { X = 0, Y, NCOORD } COORD;
typedef struct {
double x, y;
bool flag;
} Particle;
typedef struct {
int numberOfParticles, removedParticles, totalParticles;
double startTime, injectTimePeriod, writeTimePeriod, lastInjectTime, lastUpdateTime,
lastWriteTime;
int estimatedNumParticles;
double dx, dy, dt;
Particle* linSpaceLine;
Particle* particlePool;
int pointer;
double imax, jmax, xlength, ylength, imaxLocal, jmaxLocal;
double x1, y1, x2, y2;
int size, rank;
#ifdef _MPI
MPI_Datatype mpi_particle;
#endif
double xLocal, yLocal, xOffset, yOffset, xOffsetEnd, yOffsetEnd;
double* offset;
} ParticleTracer;
extern void initParticleTracer(ParticleTracer*, Discretization*, Parameter*);
extern void injectParticles(ParticleTracer*);
extern void advanceParticles(ParticleTracer*, double*, double*, double*, Comm*, double);
extern void freeParticles(ParticleTracer*);
extern void writeParticles(ParticleTracer*, Comm*);
extern void printParticleTracerParameters(ParticleTracer*);
extern void printParticles(ParticleTracer*);
extern void trace(ParticleTracer*, Discretization*, double);
extern void compress(ParticleTracer*);
#endif

51
EnhancedSolver/2D-mpi/src/progress.c Normal file
View File

@@ -0,0 +1,51 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <math.h>
#include <mpi.h>
#include <stdio.h>
#include <string.h>
#include "progress.h"
static double _end;
static int _current;
void initProgress(double end)
{
_end = end;
_current = 0;
printf("[ ]");
fflush(stdout);
}
void printProgress(double current)
{
int new = (int)rint((current / _end) * 10.0);
if (new > _current) {
char progress[11];
_current = new;
progress[0] = 0;
for (int i = 0; i < 10; i++) {
if (i < _current) {
sprintf(progress + strlen(progress), "#");
} else {
sprintf(progress + strlen(progress), " ");
}
}
printf("\r[%s]", progress);
}
fflush(stdout);
}
void stopProgress()
{
printf("\n");
fflush(stdout);
}

14
EnhancedSolver/2D-mpi/src/progress.h Normal file
View File

@@ -0,0 +1,14 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#ifndef __PROGRESS_H_
#define __PROGRESS_H_
extern void initProgress(double);
extern void printProgress(double);
extern void stopProgress();
#endif

324
EnhancedSolver/2D-mpi/src/solver-mg.c Normal file
View File

@@ -0,0 +1,324 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <stdio.h>
#include <stdlib.h>
#include "allocate.h"
#include "solver.h"
#include "util.h"
#define FINEST_LEVEL 0
#define COARSEST_LEVEL (s->levels - 1)
#define E(i, j) e[(j) * (imaxLocal + 2) + (i)]
#define R(i, j) r[(j) * (imaxLocal + 2) + (i)]
#define OLD(i, j) old[(j) * (imaxLocal + 2) + (i)]
void printSolver(Solver* s, double* grid, char* gridname)
{
if (0 == s->comm->rank) printf("Grid name : %s", gridname);
int imaxLocal = s->comm->imaxLocal;
int jmaxLocal = s->comm->jmaxLocal;
for (int i = 0; i < s->comm->size; i++) {
if (i == s->comm->rank) {
sleep(1 * s->comm->rank);
printf("### RANK %d LVL "
"###################################################### #\n ",
s->comm->rank);
for (int j = 0; j < jmaxLocal + 2; j++) {
printf("%02d: ", j);
for (int i = 0; i < imaxLocal + 2; i++) {
printf("%2.2f ", grid[j * (imaxLocal + 2) + i]);
}
printf("\n");
}
fflush(stdout);
}
}
}
static void restrictMG(Solver* s, int level, Comm* comm)
{
int imaxLocal = comm->imaxLocal;
int jmaxLocal = comm->jmaxLocal;
double* r = s->r[level + 1];
double* old = s->r[level];
#ifdef _MPI
commExchange(comm, old);
#endif
for (int j = 1; j < (jmaxLocal / 2) + 1; j++) {
for (int i = 1; i < (imaxLocal / 2) + 1; i++) {
R(i, j) = (OLD(2 * i - 1, 2 * j - 1) + OLD(2 * i, 2 * j - 1) * 2 +
OLD(2 * i + 1, 2 * j - 1) + OLD(2 * i - 1, 2 * j) * 2 +
OLD(2 * i, 2 * j) * 4 + OLD(2 * i + 1, 2 * j) * 2 +
OLD(2 * i - 1, 2 * j + 1) + OLD(2 * i, 2 * j + 1) * 2 +
OLD(2 * i + 1, 2 * j + 1)) /
16.0;
}
}
}
static void prolongate(Solver* s, int level, Comm* comm)
{
int imaxLocal = comm->imaxLocal;
int jmaxLocal = comm->jmaxLocal;
double* old = s->r[level + 1];
double* e = s->r[level];
for (int j = 2; j < jmaxLocal + 1; j += 2) {
for (int i = 2; i < imaxLocal + 1; i += 2) {
E(i, j) = OLD(i / 2, j / 2);
}
}
}
static void correct(Solver* s, double* p, int level, Comm* comm)
{
double* e = s->e[level];
int imaxLocal = comm->imaxLocal;
int jmaxLocal = comm->jmaxLocal;
for (int j = 1; j < jmaxLocal + 1; ++j) {
for (int i = 1; i < imaxLocal + 1; ++i) {
P(i, j) += E(i, j);
}
}
}
static void setBoundaryCondition(Solver* s, double* p, int imaxLocal, int jmaxLocal)
{
#ifdef _MPI
if (commIsBoundary(s->comm, B)) { // set bottom bc
for (int i = 1; i < imaxLocal + 1; i++) {
P(i, 0) = P(i, 1);
}
}
if (commIsBoundary(s->comm, T)) { // set top bc
for (int i = 1; i < imaxLocal + 1; i++) {
P(i, jmaxLocal + 1) = P(i, jmaxLocal);
}
}
if (commIsBoundary(s->comm, L)) { // set left bc
for (int j = 1; j < jmaxLocal + 1; j++) {
P(0, j) = P(1, j);
}
}
if (commIsBoundary(s->comm, R)) { // set right bc
for (int j = 1; j < jmaxLocal + 1; j++) {
P(imaxLocal + 1, j) = P(imaxLocal, j);
}
}
#else
for (int i = 1; i < imaxLocal + 1; i++) {
P(i, 0) = P(i, 1);
P(i, jmaxLocal + 1) = P(i, jmaxLocal);
}
for (int j = 1; j < jmaxLocal + 1; j++) {
P(0, j) = P(1, j);
P(imaxLocal + 1, j) = P(imaxLocal, j);
}
#endif
}
static double smooth(Solver* s, double* p, double* rhs, int level, Comm* comm)
{
int imaxLocal = comm->imaxLocal;
int jmaxLocal = comm->jmaxLocal;
int imax = s->grid->imax;
int jmax = s->grid->jmax;
double dx2 = s->grid->dx * s->grid->dx;
double dy2 = s->grid->dy * s->grid->dy;
double idx2 = 1.0 / dx2;
double idy2 = 1.0 / dy2;
double factor = s->omega * 0.5 * (dx2 * dy2) / (dx2 + dy2);
double* r = s->r[level];
double res = 1.0;
int pass, jsw, isw;
jsw = 1;
for (pass = 0; pass < 2; pass++) {
isw = jsw;
#ifdef _MPI
commExchange(comm, p);
#endif
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = isw; i < imaxLocal + 1; i += 2) {
P(i, j) -= factor *
(RHS(i, j) -
((P(i + 1, j) - 2.0 * P(i, j) + P(i - 1, j)) * idx2 +
(P(i, j + 1) - 2.0 * P(i, j) + P(i, j - 1)) * idy2));
}
isw = 3 - isw;
}
jsw = 3 - jsw;
}
}
static double calculateResidual(Solver* s, double* p, double* rhs, int level, Comm* comm)
{
int imax = s->grid->imax;
int jmax = s->grid->jmax;
int imaxLocal = comm->imaxLocal;
int jmaxLocal = comm->jmaxLocal;
double dx2 = s->grid->dx * s->grid->dx;
double dy2 = s->grid->dy * s->grid->dy;
double idx2 = 1.0 / dx2;
double idy2 = 1.0 / dy2;
double factor = s->omega * 0.5 * (dx2 * dy2) / (dx2 + dy2);
double* r = s->r[level];
double res = 1.0;
int pass, jsw, isw;
jsw = 1;
for (pass = 0; pass < 2; pass++) {
isw = jsw;
#ifdef _MPI
commExchange(comm, p);
#endif
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = isw; i < imaxLocal + 1; i += 2) {
R(i, j) = RHS(i, j) -
((P(i + 1, j) - 2.0 * P(i, j) + P(i - 1, j)) * idx2 +
(P(i, j + 1) - 2.0 * P(i, j) + P(i, j - 1)) * idy2);
res += (R(i, j) * R(i, j));
}
isw = 3 - isw;
}
jsw = 3 - jsw;
}
#ifdef _MPI
commReduction(&res, SUM);
#endif
res = res / (double)(imax * jmax);
#ifdef DEBUG
if (commIsMaster(s->comm)) {
printf("%d Residuum: %e\n", it, res);
}
#endif
return res;
}
static double multiGrid(Solver* s, double* p, double* rhs, int level, Comm* comm)
{
double res = 0.0;
// coarsest level
if (level == COARSEST_LEVEL) {
for (int i = 0; i < 5; i++) {
smooth(s, p, rhs, level, comm);
}
return res;
}
// pre-smoothing
for (int i = 0; i < s->presmooth; i++) {
smooth(s, p, rhs, level, comm);
if (level == FINEST_LEVEL)
setBoundaryCondition(s, p, comm->imaxLocal, comm->jmaxLocal);
}
// calculate residuals
res = calculateResidual(s, p, rhs, level, comm);
// restrict
restrictMG(s, level, comm);
Comm newcomm;
commUpdateDatatypes(s->comm, &newcomm, comm->imaxLocal / 2, comm->jmaxLocal / 2);
// MGSolver on residual and error.
multiGrid(s, s->e[level + 1], s->r[level + 1], level + 1, &newcomm);
commFreeCommunicator(&newcomm);
// prolongate
prolongate(s, level, comm);
// correct p on finer level using residual
correct(s, p, level, comm);
if (level == FINEST_LEVEL)
setBoundaryCondition(s, p, comm->imaxLocal, comm->jmaxLocal);
// post-smoothing
for (int i = 0; i < s->postsmooth; i++) {
smooth(s, p, rhs, level, comm);
if (level == FINEST_LEVEL)
setBoundaryCondition(s, p, comm->imaxLocal, comm->jmaxLocal);
}
return res;
}
void initSolver(Solver* s, Discretization* d, Parameter* p)
{
s->eps = p->eps;
s->omega = p->omg;
s->itermax = p->itermax;
s->levels = p->levels;
s->grid = &d->grid;
s->comm = &d->comm;
s->presmooth = p->presmooth;
s->postsmooth = p->postsmooth;
int imax = s->grid->imax;
int jmax = s->grid->jmax;
int levels = s->levels;
printf("Using Multigrid solver with %d levels\n", levels);
s->r = malloc(levels * sizeof(double*));
s->e = malloc(levels * sizeof(double*));
size_t size = (imax + 2) * (jmax + 2) * sizeof(double);
for (int j = 0; j < levels; j++) {
s->r[j] = allocate(64, size);
s->e[j] = allocate(64, size);
for (int i = 0; i < (imax + 2) * (jmax + 2); i++) {
s->r[j][i] = 0.0;
s->e[j][i] = 0.0;
}
}
}
void solve(Solver* s, double* p, double* rhs)
{
double res = multiGrid(s, p, rhs, 0, s->comm);
#ifdef VERBOSE
if (commIsMaster(s->comm)) {
printf("Residuum: %.6f\n", res);
}
#endif
}

129
EnhancedSolver/2D-mpi/src/solver-rb.c Normal file
View File

@@ -0,0 +1,129 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#include <math.h>
#include <stdio.h>
#include "allocate.h"
#include "comm.h"
#include "discretization.h"
#include "parameter.h"
#include "solver.h"
#include "util.h"
void printSolver(Solver* s, double* grid, char* gridname)
{
printf("Grid name : %s", gridname);
int imaxLocal = s->comm->imaxLocal;
int jmaxLocal = s->comm->jmaxLocal;
for (int i = 0; i < s->comm->size; i++) {
if (i == s->comm->rank) {
sleep(1 * s->comm->rank);
printf("### RANK %d LVL "
"###################################################### #\n ",
s->comm->rank);
for (int j = 0; j < jmaxLocal + 2; j++) {
printf("%02d: ", j);
for (int i = 0; i < imaxLocal + 2; i++) {
printf("%2.2f ", grid[j * (imaxLocal + 2) + i]);
}
printf("\n");
}
fflush(stdout);
}
}
}
void initSolver(Solver* s, Discretization* d, Parameter* p)
{
s->grid = &d->grid;
s->eps = p->eps;
s->omega = p->omg;
s->itermax = p->itermax;
s->comm = &d->comm;
}
void solve(Solver* s, double* p, double* rhs)
{
int imax = s->grid->imax;
int jmax = s->grid->jmax;
int imaxLocal = s->comm->imaxLocal;
int jmaxLocal = s->comm->jmaxLocal;
double eps = s->eps;
int itermax = s->itermax;
double dx2 = s->grid->dx * s->grid->dx;
double dy2 = s->grid->dy * s->grid->dy;
double idx2 = 1.0 / dx2;
double idy2 = 1.0 / dy2;
double factor = s->omega * 0.5 * (dx2 * dy2) / (dx2 + dy2);
double epssq = eps * eps;
int pass, jsw, isw;
int it = 0;
double res = 1.0;
while ((res >= epssq) && (it < itermax)) {
jsw = 1;
for (pass = 0; pass < 2; pass++) {
isw = jsw;
commExchange(s->comm, p);
for (int j = 1; j < jmaxLocal + 1; j++) {
for (int i = isw; i < imaxLocal + 1; i += 2) {
double r = RHS(i, j) -
((P(i + 1, j) - 2.0 * P(i, j) + P(i - 1, j)) * idx2 +
(P(i, j + 1) - 2.0 * P(i, j) + P(i, j - 1)) * idy2);
P(i, j) -= (factor * r);
res += (r * r);
}
isw = 3 - isw;
}
jsw = 3 - jsw;
}
if (commIsBoundary(s->comm, B)) { // set bottom bc
for (int i = 1; i < imaxLocal + 1; i++) {
P(i, 0) = P(i, 1);
}
}
if (commIsBoundary(s->comm, T)) { // set top bc
for (int i = 1; i < imaxLocal + 1; i++) {
P(i, jmaxLocal + 1) = P(i, jmaxLocal);
}
}
if (commIsBoundary(s->comm, L)) { // set left bc
for (int j = 1; j < jmaxLocal + 1; j++) {
P(0, j) = P(1, j);
}
}
if (commIsBoundary(s->comm, R)) { // set right bc
for (int j = 1; j < jmaxLocal + 1; j++) {
P(imaxLocal + 1, j) = P(imaxLocal, j);
}
}
commReduction(&res, SUM);
res = res / (double)(imax * jmax);
#ifdef DEBUG
if (commIsMaster(s->comm)) {
printf("%d Residuum: %e\n", it, res);
}
#endif
it++;
}
#ifdef VERBOSE
if (commIsMaster(s->comm)) {
printf("Solver took %d iterations to reach %f\n", it, sqrt(res));
}
#endif
}

30
EnhancedSolver/2D-mpi/src/solver.h Normal file
View File

@@ -0,0 +1,30 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved. This file is part of nusif-solver.
* Use of this source code is governed by a MIT style
* license that can be found in the LICENSE file.
*/
#ifndef __SOLVER_H_
#define __SOLVER_H_
#include "comm.h"
#include "discretization.h"
#include "grid.h"
#include "mpi.h"
#include "parameter.h"
typedef struct {
/* geometry and grid information */
Grid* grid;
/* parameters */
double eps, omega;
int itermax;
int levels, presmooth, postsmooth;
double **r, **e;
/* communication */
Comm* comm;
} Solver;
void initSolver(Solver*, Discretization*, Parameter*);
void solve(Solver*, double*, double*);
void printSolver(Solver* , double*, char*);
#endif

22
EnhancedSolver/2D-mpi/src/timing.c Normal file
View File

@@ -0,0 +1,22 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#include <stdlib.h>
#include <time.h>
double getTimeStamp(void)
{
struct timespec ts;
clock_gettime(CLOCK_MONOTONIC, &ts);
return (double)ts.tv_sec + (double)ts.tv_nsec * 1.e-9;
}
double getTimeResolution(void)
{
struct timespec ts;
clock_getres(CLOCK_MONOTONIC, &ts);
return (double)ts.tv_sec + (double)ts.tv_nsec * 1.e-9;
}

13
EnhancedSolver/2D-mpi/src/timing.h Normal file
View File

@@ -0,0 +1,13 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#ifndef __TIMING_H_
#define __TIMING_H_
extern double getTimeStamp(void);
extern double getTimeResolution(void);
#endif // __TIMING_H_

30
EnhancedSolver/2D-mpi/src/util.h Normal file
View File

@@ -0,0 +1,30 @@
/*
* Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
* All rights reserved.
* Use of this source code is governed by a MIT-style
* license that can be found in the LICENSE file.
*/
#ifndef __UTIL_H_
#define __UTIL_H_
#define HLINE \
"----------------------------------------------------------------------------\n"
#ifndef MIN
#define MIN(x, y) ((x) < (y) ? (x) : (y))
#endif
#ifndef MAX
#define MAX(x, y) ((x) > (y) ? (x) : (y))
#endif
#ifndef ABS
#define ABS(a) ((a) >= 0 ? (a) : -(a))
#endif
#define P(i, j) p[(j) * (imaxLocal + 2) + (i)]
#define F(i, j) f[(j) * (imaxLocal + 2) + (i)]
#define G(i, j) g[(j) * (imaxLocal + 2) + (i)]
#define U(i, j) u[(j) * (imaxLocal + 2) + (i)]
#define V(i, j) v[(j) * (imaxLocal + 2) + (i)]
#define S(i, j) s[(j) * (imaxLocal + 2) + (i)]
#define RHS(i, j) rhs[(j) * (imaxLocal + 2) + (i)]
#endif // __UTIL_H_

7
EnhancedSolver/2D-mpi/surface.plot Normal file
View File

@@ -0,0 +1,7 @@
set terminal png size 1024,768 enhanced font ,12
set output 'p.png'
set datafile separator whitespace
set grid
set hidden3d
splot 'pressure.dat' using 1:2:3 with lines

6
EnhancedSolver/2D-mpi/vector.plot Normal file
View File

@@ -0,0 +1,6 @@
set terminal png size 3600,1400 enhanced font ,12
set output 'velocity.png'
set datafile separator whitespace
set size ratio -1
plot 'velocity.dat' using 1:2:3:4:5 with vectors filled head size 0.01,20,60 lc palette

10
EnhancedSolver/2D-mpi/vis_files/animate.plot Normal file
View File

@@ -0,0 +1,10 @@
unset border; unset tics; unset key;
set term gif animate delay 30
set output "trace.gif"
set xrange [0:1]
set yrange [0:1]
do for [ts=0:120] {
plot "particles_".ts.".dat" with points pointtype 7
}
unset output

14
EnhancedSolver/2D-mpi/vis_files/backstep_animate.plot Normal file
View File

@@ -0,0 +1,14 @@
unset border; unset tics; unset key;
set term gif animate delay 10
set output "trace.gif"
set xrange [0:7]
set yrange [0:1.5]
set size ratio -1
set object 1 rect from 0.0,0.0 to 1.0,0.5 lw 5
do for [ts=0:300] {
plot "particles_".ts.".dat" with points pointtype 7
}
unset output

0
EnhancedSolver/2D-seq/vis_files/canal.plot → EnhancedSolver/2D-mpi/vis_files/canal_animate.plot
View File

13
EnhancedSolver/2D-mpi/vis_files/karman_animate.plot Normal file
View File

@@ -0,0 +1,13 @@
unset border; unset tics; unset key;
set term gif animate delay 10
set output "trace.gif"
set xrange [0:30]
set yrange [0:8]
set size ratio -1
set object 1 circle front at 5.0,4.0 size 1.0 fillcolor rgb "black" lw 2
do for [ts=0:500] {
plot "particles_".ts.".dat" with points pointtype 7 pointsize 0.3
}
unset output

12
EnhancedSolver/2D-seq/Makefile
View File

@@ -70,6 +70,18 @@ format:
done
@echo "Done"
vis:
$(info ===> GENERATE VISUALIZATION)
@gnuplot -e "filename='pressure.dat'" ./surface.plot
@gnuplot -e "filename='velocity.dat'" ./vector.plot
vis_clean:
$(info ===> CLEAN VISUALIZATION)
@rm -f *.dat
@rm -f *.png
@rm -f ./vis_files/*.dat
@rm -f ./vis_files/*.gif
$(BUILD_DIR):
@mkdir $(BUILD_DIR)

11
EnhancedSolver/2D-seq/backstep.par
View File

@@ -15,7 +15,7 @@ bcRight 3 #
gx 0.0 # Body forces (e.g. gravity)
gy 0.0 #
re 36500.0 # Reynolds number
re 20000.0 # Reynolds number
u_init 1.0 # initial value for velocity in x-direction
v_init 0.0 # initial value for velocity in y-direction
@@ -45,11 +45,18 @@ rho 0.52
omg 1.8 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 5 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------
numberOfParticles 200
startTime 100
startTime 10
injectTimePeriod 1.0
writeTimePeriod 0.5

8
EnhancedSolver/2D-seq/canal.par
View File

@@ -44,7 +44,13 @@ eps 0.0001 # stopping tolerance for pressure iteration
rho 0.52
omg 1.8 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
levels 5 # Multigrid levels
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 5 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------

2
EnhancedSolver/2D-seq/config.mk
View File

@@ -1,5 +1,5 @@
# Supported: GCC, CLANG, ICC
TAG ?= CLANG
TAG ?= ICC
ENABLE_OPENMP ?= false
# Supported: sor, rb, mg
SOLVER ?= mg

8
EnhancedSolver/2D-seq/dcavity.par
View File

@@ -44,7 +44,13 @@ eps 0.001 # stopping tolerance for pressure iteration
rho 0.5
omg 1.8 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
levels 5 # Multigrid levels
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 10 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------

10
EnhancedSolver/2D-seq/karman.par
View File

@@ -44,13 +44,19 @@ eps 0.001 # stopping tolerance for pressure iteration
rho 0.52
omg 1.75 # relaxation parameter for SOR iteration
gamma 0.9 # upwind differencing factor gamma
levels 5 # Multigrid levels
# Multigrid data:
# ---------
levels 3 # Multigrid levels
presmooth 5 # Pre-smoothning iterations
postsmooth 5 # Post-smoothning iterations
# Particle Tracing Data:
# -----------------------
numberOfParticles 200
startTime 201
startTime 50
injectTimePeriod 1.0
writeTimePeriod 0.5

2
EnhancedSolver/2D-seq/src/particletracing.c
View File

@@ -98,7 +98,7 @@ static void advanceParticles(
if (!gridIsFluid(p->grid, newI, newJ)) {
p->particlePool[i].flag = false;
p->removedParticles++;
printf("Forbidden movement of particle into obstacle!\n");
// printf("Forbidden movement of particle into obstacle!\n");
}
}
}

10
EnhancedSolver/2D-seq/vis_files/canal_animate.plot Normal file
View File

@@ -0,0 +1,10 @@
unset border; unset tics; unset key;
set term gif animate delay 30
set output "trace.gif"
set xrange [0:30]
set yrange [0:4]
do for [ts=0:120] {
plot "particles_".ts.".dat" with points pointtype 7
}
unset output