[tput] empty commit with the current V100/T4 status for the paper

This is the status of the current upstream/master

NB! The EvtsPerSec[MatrixElems] in CUDA are varying wildly.
In double I see this go from 5.9E8 to 6.7E8 today, yesterday 7.2E8 with the same code.
The EvtsPerSec[MatrixElems] however is much more stable around 1.37E9 in all those cases.
I will therefore keep the numbers in the paper from previous measurements and vCHEP
for EvtsPerSec[MatrixElems] (namely 7.25E8 /double and 1.59E9 /float)
and add any recent EvtsPerSec[MECalcOnly] as 1.37E9 /double and 3.28E9 /float

On itscrd70.cern.ch [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: NVIDIA Tesla V100S-PCIE-32GB]:
=========================================================================
Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.0.221]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
EvtsPerSec[MatrixElems] (3) = ( 6.743408e+08                 )  sec^-1
EvtsPerSec[MECalcOnly] (3a) = ( 1.367534e+09                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     0.743706 sec
     2,588,597,917      cycles                    #    2.648 GHz
     3,526,137,396      instructions              #    1.36  insn per cycle
       1.048524808 seconds time elapsed
==PROF== Profiling "sigmaKin": launch__registers_per_thread 120
==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.0.221]
FP precision                = FLOAT (NaN/abnormal=2, zero=0)
EvtsPerSec[MatrixElems] (3) = ( 1.516160e+09                 )  sec^-1
EvtsPerSec[MECalcOnly] (3a) = ( 3.280236e+09                 )  sec^-1
MeanMatrixElemValue         = ( 1.371686e-02 +- 3.270219e-06 )  GeV^0
TOTAL       :     0.690254 sec
     2,368,009,568      cycles                    #    2.649 GHz
     3,367,981,007      instructions              #    1.42  insn per cycle
       0.978942760 seconds time elapsed
==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
=========================================================================
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 1.301297e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     7.192022 sec
    19,241,362,073      cycles                    #    2.673 GHz
    48,583,081,581      instructions              #    2.52  insn per cycle
       7.203110555 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:  614) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 2.532683e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     4.831134 sec
    12,924,690,078      cycles                    #    2.671 GHz
    29,940,147,028      instructions              #    2.32  insn per cycle
       4.842102414 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4: 3274) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 4.593254e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     3.662869 sec
     9,247,838,169      cycles                    #    2.520 GHz
    16,560,392,033      instructions              #    1.79  insn per cycle
       3.673581024 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2746) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 4.892414e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     3.603951 sec
     9,129,054,132      cycles                    #    2.528 GHz
    16,497,282,072      instructions              #    1.81  insn per cycle
       3.614598246 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2572) (512y:   95) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 3.738379e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     4.020666 sec
     8,891,618,256      cycles                    #    2.208 GHz
    13,361,398,930      instructions              #    1.50  insn per cycle
       4.035327755 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1127) (512y:  205) (512z: 2045)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=6, zero=0)
Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 1.211023e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
TOTAL       :     7.119093 sec
    19,060,830,463      cycles                    #    2.675 GHz
    47,728,069,981      instructions              #    2.50  insn per cycle
       7.129763876 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:  578) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=6, zero=0)
Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 4.507245e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
TOTAL       :     3.351029 sec
     8,971,452,327      cycles                    #    2.672 GHz
    19,719,600,560      instructions              #    2.20  insn per cycle
       3.362076705 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4: 3719) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=5, zero=0)
Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 8.160704e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
TOTAL       :     2.695110 sec
     6,906,959,097      cycles                    #    2.556 GHz
    12,504,366,929      instructions              #    1.81  insn per cycle
       2.706037106 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3077) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=5, zero=0)
Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 8.869343e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
TOTAL       :     2.621328 sec
     6,734,710,370      cycles                    #    2.562 GHz
    12,522,685,250      instructions              #    1.86  insn per cycle
       2.631936677 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2917) (512y:   81) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=5, zero=0)
Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 7.434054e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
TOTAL       :     2.750946 sec
     6,437,749,555      cycles                    #    2.334 GHz
    10,930,642,414      instructions              #    1.70  insn per cycle
       2.761224316 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1559) (512y:  179) (512z: 2157)
=========================================================================

On lxplus770.cern.ch [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: NVIDIA Tesla T4]:
=========================================================================
Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.0.221]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
EvtsPerSec[MatrixElems] (3) = ( 3.891814e+07                 )  sec^-1
EvtsPerSec[MECalcOnly] (3a) = ( 4.009506e+07                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     1.144038 sec
       672,779,391      cycles:u                  #    0.552 GHz
     1,362,529,910      instructions:u            #    2.03  insn per cycle
       1.285736848 seconds time elapsed
==PROF== Profiling "sigmaKin": launch__registers_per_thread 120
==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.0.221]
FP precision                = FLOAT (NaN/abnormal=2, zero=0)
EvtsPerSec[MatrixElems] (3) = ( 6.377864e+08                 )  sec^-1
EvtsPerSec[MECalcOnly] (3a) = ( 8.217691e+08                 )  sec^-1
MeanMatrixElemValue         = ( 1.371686e-02 +- 3.270219e-06 )  GeV^0
TOTAL       :     0.825550 sec
       386,914,589      cycles:u                  #    0.437 GHz
       828,824,446      instructions:u            #    2.14  insn per cycle
       0.963103649 seconds time elapsed
==PROF== Profiling "sigmaKin": launch__registers_per_thread 64
==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
=========================================================================
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 1.275975e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     7.515231 sec
    19,209,549,764      cycles:u                  #    2.561 GHz
    48,627,460,774      instructions:u            #    2.53  insn per cycle
       7.553956776 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:  614) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 2.542904e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     4.980303 sec
    12,899,873,466      cycles:u                  #    2.584 GHz
    29,995,111,387      instructions:u            #    2.33  insn per cycle
       5.033503232 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4: 3274) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 4.508129e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     3.899591 sec
     9,295,935,243      cycles:u                  #    2.391 GHz
    16,619,353,171      instructions:u            #    1.79  insn per cycle
       3.969700585 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2746) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 4.775861e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     3.870736 sec
     9,103,619,309      cycles:u                  #    2.376 GHz
    16,556,440,384      instructions:u            #    1.82  insn per cycle
       3.911828159 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2572) (512y:   95) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 3.381439e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
TOTAL       :     4.330337 sec
     9,151,794,863      cycles:u                  #    2.106 GHz
    13,418,541,159      instructions:u            #    1.47  insn per cycle
       4.463700520 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1127) (512y:  205) (512z: 2045)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=6, zero=0)
Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 1.169022e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
TOTAL       :     7.431409 sec
    19,242,935,493      cycles:u                  #    2.591 GHz
    47,777,821,032      instructions:u            #    2.48  insn per cycle
       7.485035275 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:  578) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=6, zero=0)
Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 4.425426e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
TOTAL       :     3.526400 sec
     8,942,606,781      cycles:u                  #    2.588 GHz
    19,782,130,775      instructions:u            #    2.21  insn per cycle
       3.631748518 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4: 3719) (avx2:    0) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=5, zero=0)
Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 8.068817e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
TOTAL       :     2.782846 sec
     6,900,906,289      cycles:u                  #    2.473 GHz
    12,568,910,963      instructions:u            #    1.82  insn per cycle
       2.833799464 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3077) (512y:    0) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=5, zero=0)
Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 8.514326e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
TOTAL       :     2.790172 sec
     6,805,090,790      cycles:u                  #    2.475 GHz
    12,587,787,277      instructions:u            #    1.85  insn per cycle
       2.943286999 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2917) (512y:   81) (512z:    0)
-------------------------------------------------------------------------
Process                     = EPOCH1_EEMUMU_CPP [gcc (GCC) 9.2.0]
FP precision                = FLOAT (NaN/abnormal=5, zero=0)
Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
OMP threads / `nproc --all` = 1 / 4
EvtsPerSec[MECalcOnly] (3a) = ( 6.798061e+06                 )  sec^-1
MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
TOTAL       :     2.938570 sec
     6,641,119,578      cycles:u                  #    2.249 GHz
    10,993,670,890      instructions:u            #    1.66  insn per cycle
       2.979389891 seconds time elapsed
=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1559) (512y:  179) (512z: 2157)
=========================================================================

Author: valassi