update

Author: aroccon

README.md

@@ -59,12 +59,8 @@ If you use this code, please cite the following work:
 ## Reference performance
 
 Performance (NS only)
-* 128 x 128 x 128 | 2 x RTX5000@milton |   14 ms/timestep
-* 256 x 256 x 256 | 2 x RTX5000@milton |  129 ms/timestep
-* 128 x 128 x 128 | 4 x A100@Leonardo  |    7 ms/timestep
 * 256 x 256 x 256 | 4 x A100@Leonardo  |   44 ms/timestep
-* 512 x 512 x 512 | 4 x A100@Leonardo  |  470 ms/timestep
-* 512 x 512 x 512 | 8 x A100@Leonardo  |  200 ms/timestep
+* 512 x 512 x 512 | 4 x A100@Leonardo  |  270 ms/timestep
 * 1024 x 1024 x 1024 | 512 x A100@Leonardo | 35 ms/timestep
 * 2048 x 2048 x 2048 | 256 x A100@Leonardo | 740 ms/timestep
 * 128 x 128 x 128 | 4 x H100@MN5-ACC   |    7 ms/timestep
@@ -74,11 +70,6 @@ Performance (NS only)
 * 1024 x 1024 x 1024 | 512 x H100@MN5-ACC  | 32 ms/timestep
 * 2048 x 2048 x 2048 | 512 x H100@MN5-ACC  | 259 ms/timestep
 
-Max resolution tested (Poisson only):
-*  768 x  768 x  768 | 2 x RTX5000@milton - 16 GB VRAM
-* 2048 x 2048 x 2048 | 32 x A100@Leonardo - 64 GB VRAm (also tested on 256/512 GPUs)
-* 2048 x 2048 x 2048 | 32 x H100@MN5-ACC  - 64 GB VRAm (also tested on 256/512 GPUs)
-
 Phase-field introduces about 15% of overhead compared to NS only.
 
 ## Scaling

multi/main.f90

@@ -10,7 +10,6 @@ program main
 use param
 use mpivar
 use cudecompvar
-use nvtx
 
 
 implicit none
@@ -331,7 +330,6 @@ program main
     if (rank.eq.0) write(*,*) "Time step",t,"of",tfin
     call cpu_time(times)
 
-   call nvtxStartRange("Phase-field",1)
    !########################################################################################################################################
    ! START STEP 4: PHASE-FIELD SOLVER (EXPLICIT)
    !########################################################################################################################################
@@ -488,9 +486,6 @@ program main
    !########################################################################################################################################
    ! END STEP 4: PHASE-FIELD SOLVER 
    !########################################################################################################################################
-   call nvtxEndRange
-
-
 
 
 

multi/module.f90

@@ -58,75 +58,5 @@ module phase
 end module phase
 
 
-! added NVTX for profiing from maxcuda/NVTX_example
-module nvtx
-use iso_c_binding
-implicit none
-integer,private :: col(7) = [ int(Z'0000ff00'), int(Z'000000ff'), int(Z'00ffff00'), int(Z'00ff00ff'), int(Z'0000ffff'), int(Z'00ff0000'), int(Z'00ffffff')]
-character,private,target :: tempName(256)
 
-type, bind(C):: nvtxEventAttributes
-  integer(C_INT16_T):: version=1
-  integer(C_INT16_T):: size=48 !
-  integer(C_INT):: category=0
-  integer(C_INT):: colorType=1 ! NVTX_COLOR_ARGB = 1
-  integer(C_INT):: color
-  integer(C_INT):: payloadType=0 ! NVTX_PAYLOAD_UNKNOWN = 0
-  integer(C_INT):: reserved0
-  integer(C_INT64_T):: payload   ! union uint,int,double
-  integer(C_INT):: messageType=1  ! NVTX_MESSAGE_TYPE_ASCII     = 1 
-  type(C_PTR):: message  ! ascii char
-end type
-
-interface nvtxRangePush
-  ! push range with custom label and standard color
-  subroutine nvtxRangePushA(name) bind(C, name='nvtxRangePushA')
-  use iso_c_binding
-  character(kind=C_CHAR) :: name(256)
-  end subroutine
-
-  ! push range with custom label and custom color
-  subroutine nvtxRangePushEx(event) bind(C, name='nvtxRangePushEx')
-  use iso_c_binding
-  import:: nvtxEventAttributes
-  type(nvtxEventAttributes):: event
-  end subroutine
-end interface
-
-interface nvtxRangePop
-  subroutine nvtxRangePop() bind(C, name='nvtxRangePop')
-  end subroutine
-end interface
-
-contains
-
-subroutine nvtxStartRange(name,id)
-  character(kind=c_char,len=*) :: name
-  integer, optional:: id
-  type(nvtxEventAttributes):: event
-  character(kind=c_char,len=256) :: trimmed_name
-  integer:: i
-
-  trimmed_name=trim(name)//c_null_char
-
-  ! move scalar trimmed_name into character array tempName
-  do i=1,LEN(trim(name)) + 1
-     tempName(i) = trimmed_name(i:i)
-  enddo
-
-
-  if ( .not. present(id)) then
-    call nvtxRangePush(tempName)
-  else
-    event%color=col(mod(id,7)+1)
-    event%message=c_loc(tempName)
-    call nvtxRangePushEx(event)
-  end if
-end subroutine
-
-subroutine nvtxEndRange
-  call nvtxRangePop
-end subroutine
-
-end module nvtx