You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
psblas3/test/comm/spmv/psb_spmv_overlap_test.f90

803 lines
27 KiB
Fortran

!> Test program for overlapping communication and computation with psb_spmm.
!!
!! This benchmark compares two equivalent SpMV paths:
!! 1. Serialized halo exchange + compute
!! 2. Overlapped psb_spmm(..., doswap=.true.)
!!
module psb_spmv_overlap_test
use psb_base_mod
use psb_util_mod
use psb_comm_factory_mod, only: psb_comm_set
use psb_comm_schemes_mod, only: psb_comm_isend_irecv_, psb_comm_ineighbor_alltoallv_, &
& psb_comm_persistent_ineighbor_alltoallv_
implicit none
interface
function d_func_3d(x,y,z) result(val)
import :: psb_dpk_
real(psb_dpk_), intent(in) :: x,y,z
real(psb_dpk_) :: val
end function d_func_3d
end interface
contains
function d_null_func_3d(x,y,z) result(val)
real(psb_dpk_), intent(in) :: x,y,z
real(psb_dpk_) :: val
val = dzero
end function d_null_func_3d
!
! functions parametrizing the differential equation
!
!
! Note: b1, b2 and b3 are the coefficients of the first
! derivative of the unknown function. The default
! we apply here is to have them zero, so that the resulting
! matrix is symmetric/hermitian and suitable for
! testing with CG and FCG.
! When testing methods for non-hermitian matrices you can
! change the B1/B2/B3 functions to e.g. done/sqrt((3*done))
!
function b1(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: b1
real(psb_dpk_), intent(in) :: x,y,z
b1=dzero
end function b1
function b2(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: b2
real(psb_dpk_), intent(in) :: x,y,z
b2=dzero
end function b2
function b3(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: b3
real(psb_dpk_), intent(in) :: x,y,z
b3=dzero
end function b3
function c(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: c
real(psb_dpk_), intent(in) :: x,y,z
c=dzero
end function c
function a1(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: a1
real(psb_dpk_), intent(in) :: x,y,z
a1=done/80
end function a1
function a2(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: a2
real(psb_dpk_), intent(in) :: x,y,z
a2=done/80
end function a2
function a3(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: a3
real(psb_dpk_), intent(in) :: x,y,z
a3=done/80
end function a3
function g(x,y,z)
use psb_base_mod, only : psb_dpk_, done, dzero
implicit none
real(psb_dpk_) :: g
real(psb_dpk_), intent(in) :: x,y,z
g = dzero
if (x == done) then
g = done
else if (x == dzero) then
g = exp(y**2-z**2)
end if
end function g
!
! subroutine to allocate and fill in the coefficient matrix and
! the rhs.
!
subroutine psb_d_gen_pde3d(ctxt,idim,a,bv,xv,desc_a,afmt,info,&
& f,amold,vmold,imold,partition,nrl,iv,tnd)
use psb_base_mod
use psb_util_mod
!
! Discretizes the partial differential equation
!
! a1 dd(u) a2 dd(u) a3 dd(u) b1 d(u) b2 d(u) b3 d(u)
! - ------ - ------ - ------ + ----- + ------ + ------ + c u = f
! dxdx dydy dzdz dx dy dz
!
! with Dirichlet boundary conditions
! u = g
!
! on the unit cube 0<=x,y,z<=1.
!
!
! Note that if b1=b2=b3=c=0., the PDE is the Laplace equation.
!
implicit none
integer(psb_ipk_) :: idim
type(psb_dspmat_type) :: a
type(psb_d_vect_type) :: xv,bv
type(psb_desc_type) :: desc_a
type(psb_ctxt_type) :: ctxt
integer(psb_ipk_) :: info
character(len=*) :: afmt
procedure(d_func_3d), optional :: f
class(psb_d_base_sparse_mat), optional :: amold
class(psb_d_base_vect_type), optional :: vmold
class(psb_i_base_vect_type), optional :: imold
integer(psb_ipk_), optional :: partition, nrl,iv(:)
logical, optional :: tnd
! Local variables.
integer(psb_ipk_), parameter :: nb=20
type(psb_d_csc_sparse_mat) :: acsc
type(psb_d_coo_sparse_mat) :: acoo
type(psb_d_csr_sparse_mat) :: acsr
real(psb_dpk_) :: zt(nb),x,y,z
integer(psb_ipk_) :: nnz,nr,nlr,i,j,ii,ib,k, partition_
integer(psb_lpk_) :: m,n,glob_row,nt
integer(psb_ipk_) :: ix,iy,iz,ia,indx_owner
! For 3D partition
! Note: integer control variables going directly into an MPI call
! must be 4 bytes, i.e. psb_mpk_
integer(psb_mpk_) :: npdims(3), npp, minfo
integer(psb_ipk_) :: npx,npy,npz, iamx,iamy,iamz,mynx,myny,mynz
integer(psb_ipk_), allocatable :: bndx(:),bndy(:),bndz(:)
! Process grid
integer(psb_ipk_) :: np, iam, nth
integer(psb_ipk_) :: icoeff
integer(psb_lpk_), allocatable :: irow(:),icol(:),myidx(:)
real(psb_dpk_), allocatable :: val(:)
! deltah dimension of each grid cell
! deltat discretization time
real(psb_dpk_) :: deltah, sqdeltah, deltah2
real(psb_dpk_), parameter :: rhs=dzero,one=done,zero=dzero
real(psb_dpk_) :: t0, t1, t2, t3, tasb, talc, ttot, tgen, tcdasb
integer(psb_ipk_) :: err_act
procedure(d_func_3d), pointer :: f_
logical :: tnd_
character(len=20) :: name, ch_err,tmpfmt
info = psb_success_
name = 'create_matrix'
call psb_erractionsave(err_act)
call psb_info(ctxt, iam, np)
if (present(f)) then
f_ => f
else
f_ => d_null_func_3d
end if
deltah = done/(idim+1)
sqdeltah = deltah*deltah
deltah2 = (2*done)* deltah
if (present(partition)) then
if ((1<= partition).and.(partition <= 3)) then
partition_ = partition
else
write(*,*) 'Invalid partition choice ',partition,' defaulting to 3'
partition_ = 3
end if
else
partition_ = 3
end if
! initialize array descriptor and sparse matrix storage. provide an
! estimate of the number of non zeroes
m = (1_psb_lpk_*idim)*idim*idim
n = m
nnz = ((n*7)/(np))
if(iam == psb_root_) write(psb_out_unit,'("Generating Matrix (size=",i0,")...")')n
t0 = psb_wtime()
select case(partition_)
case(1)
! A BLOCK partition
if (present(nrl)) then
nr = nrl
else
!
! Using a simple BLOCK distribution.
!
nt = (m+np-1)/np
nr = max(0,min(nt,m-(iam*nt)))
end if
nt = nr
call psb_sum(ctxt,nt)
if (nt /= m) then
write(psb_err_unit,*) iam, 'Initialization error ',nr,nt,m
info = -1
call psb_barrier(ctxt)
call psb_abort(ctxt)
return
end if
!
! First example of use of CDALL: specify for each process a number of
! contiguous rows
!
call psb_cdall(ctxt,desc_a,info,nl=nr)
myidx = desc_a%get_global_indices()
nlr = size(myidx)
case(2)
! A partition defined by the user through IV
if (present(iv)) then
if (size(iv) /= m) then
write(psb_err_unit,*) iam, 'Initialization error: wrong IV size',size(iv),m
info = -1
call psb_barrier(ctxt)
call psb_abort(ctxt)
return
end if
else
write(psb_err_unit,*) iam, 'Initialization error: IV not present'
info = -1
call psb_barrier(ctxt)
call psb_abort(ctxt)
return
end if
!
! Second example of use of CDALL: specify for each row the
! process that owns it
!
call psb_cdall(ctxt,desc_a,info,vg=iv)
myidx = desc_a%get_global_indices()
nlr = size(myidx)
case(3)
! A 3-dimensional partition
! A nifty MPI function will split the process list
npdims = 0
call mpi_dims_create(np,3,npdims,info)
npx = npdims(1)
npy = npdims(2)
npz = npdims(3)
allocate(bndx(0:npx),bndy(0:npy),bndz(0:npz))
! We can reuse idx2ijk for process indices as well.
call idx2ijk(iamx,iamy,iamz,iam,npx,npy,npz,base=0)
! Now let's split the 3D cube in hexahedra
call dist1Didx(bndx,idim,npx)
mynx = bndx(iamx+1)-bndx(iamx)
call dist1Didx(bndy,idim,npy)
myny = bndy(iamy+1)-bndy(iamy)
call dist1Didx(bndz,idim,npz)
mynz = bndz(iamz+1)-bndz(iamz)
! How many indices do I own?
nlr = mynx*myny*mynz
allocate(myidx(nlr))
! Now, let's generate the list of indices I own
nr = 0
do i=bndx(iamx),bndx(iamx+1)-1
do j=bndy(iamy),bndy(iamy+1)-1
do k=bndz(iamz),bndz(iamz+1)-1
nr = nr + 1
call ijk2idx(myidx(nr),i,j,k,idim,idim,idim)
end do
end do
end do
if (nr /= nlr) then
write(psb_err_unit,*) iam,iamx,iamy,iamz, 'Initialization error: NR vs NLR ',&
& nr,nlr,mynx,myny,mynz
info = -1
call psb_barrier(ctxt)
call psb_abort(ctxt)
end if
!
! Third example of use of CDALL: specify for each process
! the set of global indices it owns.
!
call psb_cdall(ctxt,desc_a,info,vl=myidx)
case default
write(psb_err_unit,*) iam, 'Initialization error: should not get here'
info = -1
call psb_barrier(ctxt)
call psb_abort(ctxt)
return
end select
if (info == psb_success_) call psb_spall(a,desc_a,info,nnz=nnz)
! define rhs from boundary conditions; also build initial guess
if (info == psb_success_) call psb_geall(xv,desc_a,info)
if (info == psb_success_) call psb_geall(bv,desc_a,info)
call psb_barrier(ctxt)
talc = psb_wtime()-t0
if (info /= psb_success_) then
info=psb_err_from_subroutine_
ch_err='allocation rout.'
call psb_errpush(info,name,a_err=ch_err)
goto 9999
end if
! we build an auxiliary matrix consisting of one row at a
! time; just a small matrix. might be extended to generate
! a bunch of rows per call.
!
allocate(val(20*nb),irow(20*nb),&
&icol(20*nb),stat=info)
if (info /= psb_success_ ) then
info=psb_err_alloc_dealloc_
call psb_errpush(info,name)
goto 9999
endif
! loop over rows belonging to current process in a block
! distribution.
call psb_barrier(ctxt)
t1 = psb_wtime()
do ii=1, nlr,nb
ib = min(nb,nlr-ii+1)
icoeff = 1
do k=1,ib
i=ii+k-1
! local matrix pointer
glob_row=myidx(i)
! compute gridpoint coordinates
call idx2ijk(ix,iy,iz,glob_row,idim,idim,idim)
! x, y, z coordinates
x = (ix-1)*deltah
y = (iy-1)*deltah
z = (iz-1)*deltah
zt(k) = f_(x,y,z)
! internal point: build discretization
!
! term depending on (x-1,y,z)
!
val(icoeff) = -a1(x,y,z)/sqdeltah-b1(x,y,z)/deltah2
if (ix == 1) then
zt(k) = g(dzero,y,z)*(-val(icoeff)) + zt(k)
else
call ijk2idx(icol(icoeff),ix-1,iy,iz,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
endif
! term depending on (x,y-1,z)
val(icoeff) = -a2(x,y,z)/sqdeltah-b2(x,y,z)/deltah2
if (iy == 1) then
zt(k) = g(x,dzero,z)*(-val(icoeff)) + zt(k)
else
call ijk2idx(icol(icoeff),ix,iy-1,iz,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
endif
! term depending on (x,y,z-1)
val(icoeff)=-a3(x,y,z)/sqdeltah-b3(x,y,z)/deltah2
if (iz == 1) then
zt(k) = g(x,y,dzero)*(-val(icoeff)) + zt(k)
else
call ijk2idx(icol(icoeff),ix,iy,iz-1,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
endif
! term depending on (x,y,z)
val(icoeff)=(2*done)*(a1(x,y,z)+a2(x,y,z)+a3(x,y,z))/sqdeltah &
& + c(x,y,z)
call ijk2idx(icol(icoeff),ix,iy,iz,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
! term depending on (x,y,z+1)
val(icoeff)=-a3(x,y,z)/sqdeltah+b3(x,y,z)/deltah2
if (iz == idim) then
zt(k) = g(x,y,done)*(-val(icoeff)) + zt(k)
else
call ijk2idx(icol(icoeff),ix,iy,iz+1,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
endif
! term depending on (x,y+1,z)
val(icoeff)=-a2(x,y,z)/sqdeltah+b2(x,y,z)/deltah2
if (iy == idim) then
zt(k) = g(x,done,z)*(-val(icoeff)) + zt(k)
else
call ijk2idx(icol(icoeff),ix,iy+1,iz,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
endif
! term depending on (x+1,y,z)
val(icoeff)=-a1(x,y,z)/sqdeltah+b1(x,y,z)/deltah2
if (ix==idim) then
zt(k) = g(done,y,z)*(-val(icoeff)) + zt(k)
else
call ijk2idx(icol(icoeff),ix+1,iy,iz,idim,idim,idim)
irow(icoeff) = glob_row
icoeff = icoeff+1
endif
end do
call psb_spins(icoeff-1,irow,icol,val,a,desc_a,info)
if(info /= psb_success_) exit
call psb_geins(ib,myidx(ii:ii+ib-1),zt(1:ib),bv,desc_a,info)
if(info /= psb_success_) exit
zt(:)=dzero
call psb_geins(ib,myidx(ii:ii+ib-1),zt(1:ib),xv,desc_a,info)
if(info /= psb_success_) exit
end do
tgen = psb_wtime()-t1
if(info /= psb_success_) then
info=psb_err_from_subroutine_
ch_err='insert rout.'
call psb_errpush(info,name,a_err=ch_err)
goto 9999
end if
deallocate(val,irow,icol)
call psb_barrier(ctxt)
t1 = psb_wtime()
call psb_cdasb(desc_a,info,mold=imold)
tcdasb = psb_wtime()-t1
call psb_barrier(ctxt)
t1 = psb_wtime()
if (info == psb_success_) then
if (present(amold)) then
call psb_spasb(a,desc_a,info,mold=amold,bld_and=tnd)
else
call psb_spasb(a,desc_a,info,afmt=afmt,bld_and=tnd)
end if
end if
call psb_barrier(ctxt)
if(info /= psb_success_) then
info=psb_err_from_subroutine_
ch_err='asb rout.'
call psb_errpush(info,name,a_err=ch_err)
goto 9999
end if
if (info == psb_success_) call psb_geasb(xv,desc_a,info,mold=vmold)
if (info == psb_success_) call psb_geasb(bv,desc_a,info,mold=vmold)
if(info /= psb_success_) then
info=psb_err_from_subroutine_
ch_err='asb rout.'
call psb_errpush(info,name,a_err=ch_err)
goto 9999
end if
tasb = psb_wtime()-t1
call psb_barrier(ctxt)
ttot = psb_wtime() - t0
call psb_amx(ctxt,talc)
call psb_amx(ctxt,tgen)
call psb_amx(ctxt,tasb)
call psb_amx(ctxt,ttot)
if(iam == psb_root_) then
tmpfmt = a%get_fmt()
write(psb_out_unit,'("The matrix has been generated and assembled in ",a3," format.")')&
& tmpfmt
write(psb_out_unit,'("-allocation time : ",es12.5)') talc
write(psb_out_unit,'("-coeff. gen. time : ",es12.5)') tgen
write(psb_out_unit,'("-desc asbly time : ",es12.5)') tcdasb
write(psb_out_unit,'("- mat asbly time : ",es12.5)') tasb
write(psb_out_unit,'("-total time : ",es12.5)') ttot
end if
call psb_erractionrestore(err_act)
return
9999 call psb_error_handler(ctxt,err_act)
return
end subroutine psb_d_gen_pde3d
subroutine psb_spmv_overlap_kernel(ctxt)
use psb_base_mod
use psb_util_mod
implicit none
type(psb_ctxt_type), intent(in) :: ctxt
real(psb_dpk_) :: alpha, beta
type(psb_dspmat_type) :: a
type(psb_d_vect_type) :: x_isend, x_neighbor, x_persistent
type(psb_d_vect_type) :: y_ov_isend, y_ov_neighbor, y_ov_persistent
type(psb_d_vect_type) :: y_no_isend, y_no_neighbor, y_no_persistent
type(psb_desc_type) :: desc_a
character(len=64) :: env_buf
real(psb_dpk_), allocatable :: x_global(:), y_global(:)
integer(psb_ipk_) :: my_rank, np, info, err_act
integer :: env_len, env_status, ios
integer(psb_ipk_) :: n_global, idim
integer(psb_ipk_) :: i, times
real(psb_dpk_) :: t0, t1, dt
real(psb_dpk_) :: t_ov_isend, t_ov_neighbor, t_ov_persistent
real(psb_dpk_) :: t_no_isend, t_no_neighbor, t_no_persistent
real(psb_dpk_) :: err_isend, err_neighbor, err_persistent, tol
real(psb_dpk_) :: avg_ov, avg_no, speedup, gain_pct
info = psb_success_
tol = 1.0d-10
times = 100
t_ov_isend = 0.0_psb_dpk_
t_ov_neighbor = 0.0_psb_dpk_
t_ov_persistent = 0.0_psb_dpk_
t_no_isend = 0.0_psb_dpk_
t_no_neighbor = 0.0_psb_dpk_
t_no_persistent = 0.0_psb_dpk_
idim = 10
call psb_erractionsave(err_act)
call get_environment_variable('IDIM', env_buf, length=env_len, status=env_status)
if ((env_status == 0) .and. (env_len > 0)) then
read(env_buf(1:env_len), *, iostat=ios) idim
if ((ios /= 0) .or. (idim < 2)) idim = 10
end if
call get_environment_variable('TIMES', env_buf, length=env_len, status=env_status)
if ((env_status == 0) .and. (env_len > 0)) then
read(env_buf(1:env_len), *, iostat=ios) times
if ((ios /= 0) .or. (times < 1)) times = 100
end if
n_global = idim * idim * idim
alpha = done
beta = dzero
call psb_info(ctxt, my_rank, np)
call psb_barrier(ctxt)
call psb_d_gen_pde3d(ctxt,idim,a,y_ov_isend,x_isend,desc_a,"CSR",info,partition=1)
if (info /= psb_success_) goto 9999
call psb_barrier(ctxt)
if (my_rank == psb_root_) then
allocate(x_global(n_global))
allocate(y_global(n_global))
do i = 1, n_global
x_global(i) = real(mod(i,17)+1, psb_dpk_) / real(17, psb_dpk_)
y_global(i) = real(mod(i,13), psb_dpk_) / real(29, psb_dpk_)
end do
end if
call psb_geall(x_neighbor, desc_a, info)
call psb_geall(x_persistent, desc_a, info)
call psb_geall(y_ov_neighbor, desc_a, info)
call psb_geall(y_ov_persistent, desc_a, info)
call psb_geall(y_no_isend, desc_a, info)
call psb_geall(y_no_neighbor, desc_a, info)
call psb_geall(y_no_persistent, desc_a, info)
if (info /= psb_success_) goto 9999
call psb_scatter(x_global, x_isend, desc_a, info, root=psb_root_)
call psb_scatter(x_global, x_neighbor, desc_a, info, root=psb_root_)
call psb_scatter(x_global, x_persistent, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_ov_isend, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_ov_neighbor, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_ov_persistent, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_no_isend, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_no_neighbor, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_no_persistent, desc_a, info, root=psb_root_)
if (info /= psb_success_) goto 9999
! Set communication schemes on the x vectors used by psb_spmm.
call psb_comm_set(psb_comm_isend_irecv_, x_isend%v%comm_handle, info)
if (info /= psb_success_) goto 9999
call psb_comm_set(psb_comm_ineighbor_alltoallv_, x_neighbor%v%comm_handle, info)
if (info /= psb_success_) goto 9999
call psb_comm_set(psb_comm_persistent_ineighbor_alltoallv_, x_persistent%v%comm_handle, info)
if (info /= psb_success_) goto 9999
! Warm-up all schemes once: overlap and non-overlap paths.
call psb_spmm(alpha, a, x_isend, beta, y_ov_isend, desc_a, info, doswap=.true.)
call psb_halo(x_isend, desc_a, info)
call psb_spmm(alpha, a, x_isend, beta, y_no_isend, desc_a, info, doswap=.false.)
call psb_spmm(alpha, a, x_neighbor, beta, y_ov_neighbor, desc_a, info, doswap=.true.)
call psb_halo(x_neighbor, desc_a, info)
call psb_spmm(alpha, a, x_neighbor, beta, y_no_neighbor, desc_a, info, doswap=.false.)
call psb_spmm(alpha, a, x_persistent, beta, y_ov_persistent, desc_a, info, doswap=.true.)
call psb_halo(x_persistent, desc_a, info)
call psb_spmm(alpha, a, x_persistent, beta, y_no_persistent, desc_a, info, doswap=.false.)
if (info /= psb_success_) goto 9999
! -----------------------------
! isend/irecv scheme
! -----------------------------
call psb_scatter(x_global, x_isend, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_ov_isend, desc_a, info, root=psb_root_)
call psb_barrier(ctxt)
t0 = psb_wtime()
do i = 1, times
call psb_spmm(alpha, a, x_isend, beta, y_ov_isend, desc_a, info, doswap=.true.)
end do
t1 = psb_wtime()
dt = t1 - t0
call psb_amx(ctxt, dt)
t_ov_isend = dt
call psb_scatter(x_global, x_isend, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_no_isend, desc_a, info, root=psb_root_)
call psb_barrier(ctxt)
t0 = psb_wtime()
do i = 1, times
call psb_halo(x_isend, desc_a, info)
call psb_spmm(alpha, a, x_isend, beta, y_no_isend, desc_a, info, doswap=.false.)
end do
t1 = psb_wtime()
dt = t1 - t0
call psb_amx(ctxt, dt)
t_no_isend = dt
! -----------------------------
! ineighbor_alltoallv scheme
! -----------------------------
call psb_scatter(x_global, x_neighbor, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_ov_neighbor, desc_a, info, root=psb_root_)
call psb_barrier(ctxt)
t0 = psb_wtime()
do i = 1, times
call psb_spmm(alpha, a, x_neighbor, beta, y_ov_neighbor, desc_a, info, doswap=.true.)
end do
t1 = psb_wtime()
dt = t1 - t0
call psb_amx(ctxt, dt)
t_ov_neighbor = dt
call psb_scatter(x_global, x_neighbor, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_no_neighbor, desc_a, info, root=psb_root_)
call psb_barrier(ctxt)
t0 = psb_wtime()
do i = 1, times
call psb_halo(x_neighbor, desc_a, info)
call psb_spmm(alpha, a, x_neighbor, beta, y_no_neighbor, desc_a, info, doswap=.false.)
end do
t1 = psb_wtime()
dt = t1 - t0
call psb_amx(ctxt, dt)
t_no_neighbor = dt
! ----------------------------------------
! persistent_ineighbor_alltoallv scheme
! ----------------------------------------
call psb_scatter(x_global, x_persistent, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_ov_persistent, desc_a, info, root=psb_root_)
call psb_barrier(ctxt)
t0 = psb_wtime()
do i = 1, times
call psb_spmm(alpha, a, x_persistent, beta, y_ov_persistent, desc_a, info, doswap=.true.)
end do
t1 = psb_wtime()
dt = t1 - t0
call psb_amx(ctxt, dt)
t_ov_persistent = dt
call psb_scatter(x_global, x_persistent, desc_a, info, root=psb_root_)
call psb_scatter(y_global, y_no_persistent, desc_a, info, root=psb_root_)
call psb_barrier(ctxt)
t0 = psb_wtime()
do i = 1, times
call psb_halo(x_persistent, desc_a, info)
call psb_spmm(alpha, a, x_persistent, beta, y_no_persistent, desc_a, info, doswap=.false.)
end do
t1 = psb_wtime()
dt = t1 - t0
call psb_amx(ctxt, dt)
t_no_persistent = dt
if (info /= psb_success_) goto 9999
err_isend = maxval(abs(y_ov_isend%get_vect() - y_no_isend%get_vect()))
err_neighbor = maxval(abs(y_ov_neighbor%get_vect() - y_no_neighbor%get_vect()))
err_persistent = maxval(abs(y_ov_persistent%get_vect() - y_no_persistent%get_vect()))
call psb_amx(ctxt, err_isend)
call psb_amx(ctxt, err_neighbor)
call psb_amx(ctxt, err_persistent)
if (my_rank == 0) then
write(psb_out_unit,'(/,"SpMV overlap benchmark")')
write(psb_out_unit,'(" idim : ",i0)') idim
write(psb_out_unit,'(" global unknowns : ",i0)') n_global
write(psb_out_unit,'(" repetitions : ",i0)') times
write(psb_out_unit,'(" timing metric : max over MPI ranks")')
write(psb_out_unit,'(" gain(%) = 100*(1 - overlap/no_overlap)")')
write(psb_out_unit,'(/,"Scheme: isend_irecv")')
avg_ov = t_ov_isend / real(times, psb_dpk_)
avg_no = t_no_isend / real(times, psb_dpk_)
speedup = t_no_isend / max(t_ov_isend, tiny(done))
gain_pct = 100.0_psb_dpk_ * (done - (t_ov_isend / max(t_no_isend, tiny(done))))
write(psb_out_unit,'(" total overlap : ",es12.5)') t_ov_isend
write(psb_out_unit,'(" total no_overlap : ",es12.5)') t_no_isend
write(psb_out_unit,'(" avg overlap : ",es12.5)') avg_ov
write(psb_out_unit,'(" avg no_overlap : ",es12.5)') avg_no
write(psb_out_unit,'(" speedup (no/ov) : ",f10.4)') speedup
write(psb_out_unit,'(" gain (%) : ",f10.4)') gain_pct
write(psb_out_unit,'(" overlap vs no_overlap err = ",es12.5)') err_isend
write(psb_out_unit,'(/,"Scheme: ineighbor_alltoallv")')
avg_ov = t_ov_neighbor / real(times, psb_dpk_)
avg_no = t_no_neighbor / real(times, psb_dpk_)
speedup = t_no_neighbor / max(t_ov_neighbor, tiny(done))
gain_pct = 100.0_psb_dpk_ * (done - (t_ov_neighbor / max(t_no_neighbor, tiny(done))))
write(psb_out_unit,'(" total overlap : ",es12.5)') t_ov_neighbor
write(psb_out_unit,'(" total no_overlap : ",es12.5)') t_no_neighbor
write(psb_out_unit,'(" avg overlap : ",es12.5)') avg_ov
write(psb_out_unit,'(" avg no_overlap : ",es12.5)') avg_no
write(psb_out_unit,'(" speedup (no/ov) : ",f10.4)') speedup
write(psb_out_unit,'(" gain (%) : ",f10.4)') gain_pct
write(psb_out_unit,'(" overlap vs no_overlap err = ",es12.5)') err_neighbor
write(psb_out_unit,'(/,"Scheme: persistent_ineighbor_alltoallv")')
avg_ov = t_ov_persistent / real(times, psb_dpk_)
avg_no = t_no_persistent / real(times, psb_dpk_)
speedup = t_no_persistent / max(t_ov_persistent, tiny(done))
gain_pct = 100.0_psb_dpk_ * (done - (t_ov_persistent / max(t_no_persistent, tiny(done))))
write(psb_out_unit,'(" total overlap : ",es12.5)') t_ov_persistent
write(psb_out_unit,'(" total no_overlap : ",es12.5)') t_no_persistent
write(psb_out_unit,'(" avg overlap : ",es12.5)') avg_ov
write(psb_out_unit,'(" avg no_overlap : ",es12.5)') avg_no
write(psb_out_unit,'(" speedup (no/ov) : ",f10.4)') speedup
write(psb_out_unit,'(" gain (%) : ",f10.4)') gain_pct
write(psb_out_unit,'(" overlap vs no_overlap err = ",es12.5)') err_persistent
if ((err_isend > tol) .or. (err_neighbor > tol) .or. (err_persistent > tol)) then
write(psb_out_unit,'(" WARNING: mismatch exceeds tolerance ",es12.5)') tol
end if
end if
call psb_gefree(x_isend, desc_a, info)
call psb_gefree(x_neighbor, desc_a, info)
call psb_gefree(x_persistent, desc_a, info)
call psb_gefree(y_ov_isend, desc_a, info)
call psb_gefree(y_ov_neighbor, desc_a, info)
call psb_gefree(y_ov_persistent, desc_a, info)
call psb_gefree(y_no_isend, desc_a, info)
call psb_gefree(y_no_neighbor, desc_a, info)
call psb_gefree(y_no_persistent, desc_a, info)
call psb_spfree(a, desc_a, info)
call psb_cdfree(desc_a, info)
if (my_rank == 0) then
deallocate(x_global)
deallocate(y_global)
end if
return
9999 call psb_error(ctxt)
call psb_error_handler(ctxt, err_act)
end subroutine psb_spmv_overlap_kernel
end module psb_spmv_overlap_test