[ 4 / 4 ] Application profile is long enough (1091.23 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 0.19 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 91.91% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 0.24 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.

[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.

[ 0.24 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time)

To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (90.03%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] CPU activity is good

CPU cores are active 93.92% of time

[ 4 / 4 ] Threads activity is good

On average, more than 93.91% of observed threads are actually active

[ 2 / 4 ] Affinity stability is lower than 90% (65.93%)

Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (55.65%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (88.94%)

If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.09%) lower than cumulative innermost loop coverage (88.94%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 3 / 3 ] Less than 10% (0.04%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 2 / 2 ] Less than 10% (0.09%) is spend in BLAS2 operations

BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions)

[ 4 / 4 ] Application profile is long enough (621.67 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 0.19 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 91.98% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 0.24 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.

[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.

[ 0.24 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time)

To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (85.80%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 3 / 4 ] CPU activity is below 90% (87.12%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Threads activity is good

On average, more than 174.12% of observed threads are actually active

[ 3 / 4 ] Affinity stability is lower than 90% (71.32%)

Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (53.08%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (84.70%)

If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.10%) lower than cumulative innermost loop coverage (84.70%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 3 / 3 ] Less than 10% (0.04%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 2 / 2 ] Less than 10% (0.10%) is spend in BLAS2 operations

BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions)

[ 4 / 4 ] Application profile is long enough (374.46 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 0.19 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 92.25% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 0.23 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.

[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.

[ 0.23 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time)

To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (80.48%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 3 / 4 ] CPU activity is below 90% (77.90%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Threads activity is good

On average, more than 311.26% of observed threads are actually active

[ 3 / 4 ] Affinity stability is lower than 90% (86.98%)

Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (49.74%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (79.40%)

If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.07%) lower than cumulative innermost loop coverage (79.40%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 3 / 3 ] Less than 10% (0.04%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 2 / 2 ] Less than 10% (0.11%) is spend in BLAS2 operations

BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions)

[ 4 / 4 ] Application profile is long enough (253.64 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 0.18 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 92.64% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 0.22 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.

[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.

[ 0.22 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time)

To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (71.52%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 2 / 4 ] CPU activity is below 90% (67.01%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Threads activity is good

On average, more than 534.99% of observed threads are actually active

[ 4 / 4 ] Affinity is good (91.11%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (44.12%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (70.55%)

If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.98%) lower than cumulative innermost loop coverage (70.55%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 3 / 3 ] Less than 10% (0.06%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 2 / 2 ] Less than 10% (0.10%) is spend in BLAS2 operations

BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions)

[ 4 / 4 ] Application profile is long enough (194.44 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 0.16 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 93.35% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 0.20 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.

[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.

[ 0.20 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time)

To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (60.05%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 2 / 4 ] CPU activity is below 90% (57.00%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Threads activity is good

On average, more than 909.14% of observed threads are actually active

[ 4 / 4 ] Affinity is good (92.87%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (37.14%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (59.25%)

If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.80%) lower than cumulative innermost loop coverage (59.25%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 3 / 3 ] Less than 10% (0.09%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 2 / 2 ] Less than 10% (0.11%) is spend in BLAS2 operations

BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions)

[ 4 / 4 ] Application profile is long enough (170.90 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 0.12 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 94.99% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 0.15 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.

[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.

[ 0.15 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time)

To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (51.48%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 2 / 4 ] CPU activity is below 90% (50.85%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Threads activity is good

On average, more than 1621.51% of observed threads are actually active

[ 4 / 4 ] Affinity is good (91.86%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (31.85%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (50.83%)

If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.66%) lower than cumulative innermost loop coverage (50.83%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 3 / 3 ] Less than 10% (0.07%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 2 / 2 ] Less than 10% (0.12%) is spend in BLAS2 operations

BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions)