Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 1 1391349 32961525
executing #MPI = 3 and #OMP = 1
=================================================
MUMPS compiled with option -Dmetis
MUMPS compiled with option -Dpord
MUMPS compiled with option -Dptscotch
MUMPS compiled with option -Dscotch
This MUMPS version includes code for mixed precision BLR including memory accessors during solve
=================================================
L D L^T Solver for general symmetric matrices
Type of parallelism: Working host
****** ANALYSIS STEP ********
Processing a graph of size: 1391349
Average density of rows/columns = 45
Ordering based on METIS
ELAPSED TIME SPENT IN METIS reordering = 8.4085
SYMBOLIC based on column counts
ELAPSED TIME IN symbolic factorization = 0.5070
A root of estimated size 15552 has been selected for Scalapack.
Leaving analysis phase with ...
INFOG(1) = 0
INFOG(2) = 0
-- (20) Number of entries in factors (estim.) = 2796208095
-- (3) Real space for factors (estimated) = 2956130745
-- (4) Integer space for factors (estimated) = 19017628
-- (5) Maximum frontal size (estimated) = 25827
-- (6) Number of nodes in the tree = 32938
-- (32) Type of analysis effectively used = 1
-- (7) Ordering option effectively used = 5
ICNTL (6) Maximum transversal option = 0
ICNTL (7) Pivot order option = 7
ICNTL(12) Ordering symmetric indef. matrices = 1
ICNTL(13) Parallelism/splitting of root node = 0
ICNTL(14) Percentage of memory relaxation = 30
ICNTL(15) Analysis by block effectively used = 0
ICNTL(18) Distributed input matrix (on if >0) = 0
ICNTL(32) Forward elimination during facto. = 0
ICNTL(35) BLR activation = 0
ICNTL(48) Tree based multithreading (effective)= 0
ICNTL(58) Symbolic factorization option = 2
Number of level 2 nodes = 10
Number of split nodes = 0
RINFOG(1) Operations during elimination (estim)= 3.095D+13
MEMORY ESTIMATIONS ...
Estimations with standard Full-Rank (FR) factorization:
Maximum estim. space in Mbytes, IC facto. (INFOG(16)): 17272
Total space in MBytes, IC factorization (INFOG(17)): 51176
Maximum estim. space in Mbytes, OOC facto. (INFOG(26)): 10481
Total space in MBytes, OOC factorization (INFOG(27)): 29506
Elapsed time in analysis driver= 9.6263
Analysis time by clock_gettime(): 9.627 s
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 7 1391349 32961525
executing #MPI = 3 and #OMP = 1
Elapsed time in save structure driver= 0.0004
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
On return from DMUMPS, INFOG(1)= -71
On return from DMUMPS, INFOG(2)= 0
PRE FACTO START LPROF----------------------
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 2 1391349 32961525
executing #MPI = 3 and #OMP = 1
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
****** FACTORIZATION STEP ********
GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
Number of working processes = 3
ICNTL(22) Out-of-core option = 0
ICNTL(35) BLR activation (eff. choice) = 0
ICNTL(37) BLR CB compression (eff. choice) = 0
ICNTL(40) Mixed/adaptive precision BLR = 0
ICNTL(49) Compact workarray S (end facto.) = 0
ICNTL(56) Effective value during facto. = 0
ICNTL(14) Memory relaxation = 30
INFOG(3) Real space for factors (estimated)= 2956130745
INFOG(4) Integer space for factors (estim.)= 19017628
Maximum frontal size (estimated) = 25827
Number of nodes in the tree = 32938
ICNTL(23) Memory allowed (value on host) = 0
Sum over all procs = 0
Memory provided by user, sum of LWK_USER = 0
Effective threshold for pivoting, CNTL(1) = 0.1000D-01
* [MAQAO] Info: STARTING COUNTERS (kairoscomp22)
[0m
Statistics on the scaling phase
Elapsed time for scaling = 0.5493
Max difference from 1 after scaling the entries for ONE-NORM (option 7/8) = 0.28D+00
Average Effective size of S (based on INFO(39))= 2080946247
Elapsed time to reformat/distribute matrix = 0.6069
Allocated buffers
------------------
Size of reception buffer in bytes ...... = 39603204
Size of async. emission buffer (bytes).. = 158808844
Small emission buffer (bytes) .......... = 556
** Memory allocated, max in Mbytes (INFOG(18)): 17279
** Memory allocated, total in Mbytes (INFOG(19)): 51313
** Memory effectively used, max in Mbytes (INFOG(21)): 14172
** Memory effectively used, total in Mbytes (INFOG(22)): 37769
Elapsed time to process root node = 11.4072
Elapsed time for factorization = 206.2703
Leaving factorization with ...
RINFOG (2) Operations in node assembly = 9.039D+09
------ (3) Operations in node elimination = 3.095D+13
ICNTL (8) Scaling effectively used = 7
INFOG (9) Real space for factors = 2956137604
INFOG (10) Integer space for factors = 18945437
INFOG (11) Maximum front size = 25827
INFOG (29) Number of entries in factors = 2796208095
INFOG (12) Number of negative pivots = 0
INFOG (13) Number of delayed pivots = 0
Number of 2x2 pivots in type 1 nodes = 0
Number of 2x2 pivots in type 2 nodes = 0
RINFOG(19) Smallest pivot WITH perturbed pivots = 1.403D-03
RINFOG(20) Smallest pivot WITHOUT perturbed pivots = 1.403D-03
RINFOG(21) Largest pivot in absolute value = 1.000D+00
INFOG (24) Effective value of ICNTL(12) = 1
INFOG (14) Number of memory compress = 1
Elapsed time in factorization driver = 207.4359
Factorization time by clock_gettime(): 207.4348 s
Entering DMUMPS 5.9.0 from C interface with JOB = -2
executing #MPI = 3 and #OMP = 1
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 1 1391349 32961525
executing #MPI = 3 and #OMP = 2
=================================================
MUMPS compiled with option -Dmetis
MUMPS compiled with option -Dpord
MUMPS compiled with option -Dptscotch
MUMPS compiled with option -Dscotch
This MUMPS version includes code for mixed precision BLR including memory accessors during solve
=================================================
L D L^T Solver for general symmetric matrices
Type of parallelism: Working host
****** ANALYSIS STEP ********
Processing a graph of size: 1391349
Average density of rows/columns = 45
Ordering based on METIS
ELAPSED TIME SPENT IN METIS reordering = 8.3875
SYMBOLIC based on column counts
ELAPSED TIME IN symbolic factorization = 0.5074
A root of estimated size 15552 has been selected for Scalapack.
Leaving analysis phase with ...
INFOG(1) = 0
INFOG(2) = 0
-- (20) Number of entries in factors (estim.) = 2796208095
-- (3) Real space for factors (estimated) = 2956130745
-- (4) Integer space for factors (estimated) = 19017628
-- (5) Maximum frontal size (estimated) = 25827
-- (6) Number of nodes in the tree = 32938
-- (32) Type of analysis effectively used = 1
-- (7) Ordering option effectively used = 5
ICNTL (6) Maximum transversal option = 0
ICNTL (7) Pivot order option = 7
ICNTL(12) Ordering symmetric indef. matrices = 1
ICNTL(13) Parallelism/splitting of root node = 0
ICNTL(14) Percentage of memory relaxation = 30
ICNTL(15) Analysis by block effectively used = 0
ICNTL(18) Distributed input matrix (on if >0) = 0
ICNTL(32) Forward elimination during facto. = 0
ICNTL(35) BLR activation = 0
ICNTL(48) Tree based multithreading (effective)= 1
ICNTL(58) Symbolic factorization option = 2
Number of level 2 nodes = 10
Number of split nodes = 0
RINFOG(1) Operations during elimination (estim)= 3.095D+13
MEMORY ESTIMATIONS ...
Estimations with standard Full-Rank (FR) factorization:
Maximum estim. space in Mbytes, IC facto. (INFOG(16)): 18329
Total space in MBytes, IC factorization (INFOG(17)): 53537
Maximum estim. space in Mbytes, OOC facto. (INFOG(26)): 11859
Total space in MBytes, OOC factorization (INFOG(27)): 35267
Elapsed time in analysis driver= 9.6013
Analysis time by clock_gettime(): 9.602 s
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 7 1391349 32961525
executing #MPI = 3 and #OMP = 2
Elapsed time in save structure driver= 0.0003
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
On return from DMUMPS, INFOG(1)= -71
On return from DMUMPS, INFOG(2)= 0
PRE FACTO START LPROF----------------------
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 2 1391349 32961525
executing #MPI = 3 and #OMP = 2
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
****** FACTORIZATION STEP ********
GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
Number of working processes = 3
ICNTL(22) Out-of-core option = 0
ICNTL(35) BLR activation (eff. choice) = 0
ICNTL(37) BLR CB compression (eff. choice) = 0
ICNTL(40) Mixed/adaptive precision BLR = 0
ICNTL(49) Compact workarray S (end facto.) = 0
ICNTL(56) Effective value during facto. = 0
ICNTL(14) Memory relaxation = 30
INFOG(3) Real space for factors (estimated)= 2956130745
INFOG(4) Integer space for factors (estim.)= 19017628
Maximum frontal size (estimated) = 25827
Number of nodes in the tree = 32938
ICNTL(23) Memory allowed (value on host) = 0
Sum over all procs = 0
Memory provided by user, sum of LWK_USER = 0
Effective threshold for pivoting, CNTL(1) = 0.1000D-01
* [MAQAO] Info: STARTING COUNTERS (kairoscomp22)
[0m
Statistics on the scaling phase
Elapsed time for scaling = 0.2867
Max difference from 1 after scaling the entries for ONE-NORM (option 7/8) = 0.28D+00
Average Effective size of S (based on INFO(39))= 1422946184
Elapsed time to reformat/distribute matrix = 0.5301
Allocated buffers
------------------
Size of reception buffer in bytes ...... = 39603204
Size of async. emission buffer (bytes).. = 158808844
Small emission buffer (bytes) .......... = 556
** Memory allocated, max in Mbytes (INFOG(18)): 18329
** Memory allocated, total in Mbytes (INFOG(19)): 53537
** Memory effectively used, max in Mbytes (INFOG(21)): 16054
** Memory effectively used, total in Mbytes (INFOG(22)): 45236
Flops under L0 layer (avg/max across MPI) = 2.232D+12 2.875D+12
Elapsed time under L0 (avg/max across MPI) = 19.4147 23.3532
Elapsed time to process root node = 11.3805
Elapsed time for factorization = 182.6109
Leaving factorization with ...
RINFOG (2) Operations in node assembly = 9.039D+09
------ (3) Operations in node elimination = 3.095D+13
ICNTL (8) Scaling effectively used = 7
INFOG (9) Real space for factors = 2956137604
INFOG (10) Integer space for factors = 18945437
INFOG (11) Maximum front size = 25827
INFOG (29) Number of entries in factors = 2796208095
INFOG (12) Number of negative pivots = 0
INFOG (13) Number of delayed pivots = 0
Number of 2x2 pivots in type 1 nodes = 0
Number of 2x2 pivots in type 2 nodes = 0
RINFOG(19) Smallest pivot WITH perturbed pivots = 1.403D-03
RINFOG(20) Smallest pivot WITHOUT perturbed pivots = 1.403D-03
RINFOG(21) Largest pivot in absolute value = 1.000D+00
INFOG (24) Effective value of ICNTL(12) = 1
INFOG (14) Number of memory compress = 2
Elapsed time in factorization driver = 183.4343
Factorization time by clock_gettime(): 183.4334 s
Entering DMUMPS 5.9.0 from C interface with JOB = -2
executing #MPI = 3 and #OMP = 2
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 1 1391349 32961525
executing #MPI = 3 and #OMP = 4
=================================================
MUMPS compiled with option -Dmetis
MUMPS compiled with option -Dpord
MUMPS compiled with option -Dptscotch
MUMPS compiled with option -Dscotch
This MUMPS version includes code for mixed precision BLR including memory accessors during solve
=================================================
L D L^T Solver for general symmetric matrices
Type of parallelism: Working host
****** ANALYSIS STEP ********
Processing a graph of size: 1391349
Average density of rows/columns = 45
Ordering based on METIS
ELAPSED TIME SPENT IN METIS reordering = 8.4002
SYMBOLIC based on column counts
ELAPSED TIME IN symbolic factorization = 0.5070
A root of estimated size 15552 has been selected for Scalapack.
Leaving analysis phase with ...
INFOG(1) = 0
INFOG(2) = 0
-- (20) Number of entries in factors (estim.) = 2796208095
-- (3) Real space for factors (estimated) = 2956130745
-- (4) Integer space for factors (estimated) = 19017628
-- (5) Maximum frontal size (estimated) = 25827
-- (6) Number of nodes in the tree = 32938
-- (32) Type of analysis effectively used = 1
-- (7) Ordering option effectively used = 5
ICNTL (6) Maximum transversal option = 0
ICNTL (7) Pivot order option = 7
ICNTL(12) Ordering symmetric indef. matrices = 1
ICNTL(13) Parallelism/splitting of root node = 0
ICNTL(14) Percentage of memory relaxation = 30
ICNTL(15) Analysis by block effectively used = 0
ICNTL(18) Distributed input matrix (on if >0) = 0
ICNTL(32) Forward elimination during facto. = 0
ICNTL(35) BLR activation = 0
ICNTL(48) Tree based multithreading (effective)= 1
ICNTL(58) Symbolic factorization option = 2
Number of level 2 nodes = 10
Number of split nodes = 0
RINFOG(1) Operations during elimination (estim)= 3.095D+13
MEMORY ESTIMATIONS ...
Estimations with standard Full-Rank (FR) factorization:
Maximum estim. space in Mbytes, IC facto. (INFOG(16)): 18430
Total space in MBytes, IC factorization (INFOG(17)): 53797
Maximum estim. space in Mbytes, OOC facto. (INFOG(26)): 12463
Total space in MBytes, OOC factorization (INFOG(27)): 35803
Elapsed time in analysis driver= 9.6108
Analysis time by clock_gettime(): 9.611 s
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 7 1391349 32961525
executing #MPI = 3 and #OMP = 4
Elapsed time in save structure driver= 0.0004
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
On return from DMUMPS, INFOG(1)= -71
On return from DMUMPS, INFOG(2)= 0
PRE FACTO START LPROF----------------------
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 2 1391349 32961525
executing #MPI = 3 and #OMP = 4
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
****** FACTORIZATION STEP ********
GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
Number of working processes = 3
ICNTL(22) Out-of-core option = 0
ICNTL(35) BLR activation (eff. choice) = 0
ICNTL(37) BLR CB compression (eff. choice) = 0
ICNTL(40) Mixed/adaptive precision BLR = 0
ICNTL(49) Compact workarray S (end facto.) = 0
ICNTL(56) Effective value during facto. = 0
ICNTL(14) Memory relaxation = 30
INFOG(3) Real space for factors (estimated)= 2956130745
INFOG(4) Integer space for factors (estim.)= 19017628
Maximum frontal size (estimated) = 25827
Number of nodes in the tree = 32938
ICNTL(23) Memory allowed (value on host) = 0
Sum over all procs = 0
Memory provided by user, sum of LWK_USER = 0
Effective threshold for pivoting, CNTL(1) = 0.1000D-01
* [MAQAO] Info: STARTING COUNTERS (kairoscomp22)
[0m
Statistics on the scaling phase
Elapsed time for scaling = 0.1494
Max difference from 1 after scaling the entries for ONE-NORM (option 7/8) = 0.28D+00
Average Effective size of S (based on INFO(39))= 1490470837
Elapsed time to reformat/distribute matrix = 0.5211
Allocated buffers
------------------
Size of reception buffer in bytes ...... = 39603204
Size of async. emission buffer (bytes).. = 158808844
Small emission buffer (bytes) .......... = 556
** Memory allocated, max in Mbytes (INFOG(18)): 18430
** Memory allocated, total in Mbytes (INFOG(19)): 53797
** Memory effectively used, max in Mbytes (INFOG(21)): 15940
** Memory effectively used, total in Mbytes (INFOG(22)): 43917
Flops under L0 layer (avg/max across MPI) = 1.651D+12 2.056D+12
Elapsed time under L0 (avg/max across MPI) = 8.0766 9.7905
Elapsed time to process root node = 11.4091
Elapsed time for factorization = 176.4186
Leaving factorization with ...
RINFOG (2) Operations in node assembly = 9.039D+09
------ (3) Operations in node elimination = 3.095D+13
ICNTL (8) Scaling effectively used = 7
INFOG (9) Real space for factors = 2956137604
INFOG (10) Integer space for factors = 18945437
INFOG (11) Maximum front size = 25827
INFOG (29) Number of entries in factors = 2796208095
INFOG (12) Number of negative pivots = 0
INFOG (13) Number of delayed pivots = 0
Number of 2x2 pivots in type 1 nodes = 0
Number of 2x2 pivots in type 2 nodes = 0
RINFOG(19) Smallest pivot WITH perturbed pivots = 1.403D-03
RINFOG(20) Smallest pivot WITHOUT perturbed pivots = 1.403D-03
RINFOG(21) Largest pivot in absolute value = 1.000D+00
INFOG (24) Effective value of ICNTL(12) = 1
INFOG (14) Number of memory compress = 2
Elapsed time in factorization driver = 177.0969
Factorization time by clock_gettime(): 177.0961 s
Entering DMUMPS 5.9.0 from C interface with JOB = -2
executing #MPI = 3 and #OMP = 4
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 1 1391349 32961525
executing #MPI = 3 and #OMP = 8
=================================================
MUMPS compiled with option -Dmetis
MUMPS compiled with option -Dpord
MUMPS compiled with option -Dptscotch
MUMPS compiled with option -Dscotch
This MUMPS version includes code for mixed precision BLR including memory accessors during solve
=================================================
L D L^T Solver for general symmetric matrices
Type of parallelism: Working host
****** ANALYSIS STEP ********
Processing a graph of size: 1391349
Average density of rows/columns = 45
Ordering based on METIS
ELAPSED TIME SPENT IN METIS reordering = 8.4112
SYMBOLIC based on column counts
ELAPSED TIME IN symbolic factorization = 0.5080
A root of estimated size 15552 has been selected for Scalapack.
Leaving analysis phase with ...
INFOG(1) = 0
INFOG(2) = 0
-- (20) Number of entries in factors (estim.) = 2796208095
-- (3) Real space for factors (estimated) = 2956130745
-- (4) Integer space for factors (estimated) = 19017628
-- (5) Maximum frontal size (estimated) = 25827
-- (6) Number of nodes in the tree = 32938
-- (32) Type of analysis effectively used = 1
-- (7) Ordering option effectively used = 5
ICNTL (6) Maximum transversal option = 0
ICNTL (7) Pivot order option = 7
ICNTL(12) Ordering symmetric indef. matrices = 1
ICNTL(13) Parallelism/splitting of root node = 0
ICNTL(14) Percentage of memory relaxation = 30
ICNTL(15) Analysis by block effectively used = 0
ICNTL(18) Distributed input matrix (on if >0) = 0
ICNTL(32) Forward elimination during facto. = 0
ICNTL(35) BLR activation = 0
ICNTL(48) Tree based multithreading (effective)= 1
ICNTL(58) Symbolic factorization option = 2
Number of level 2 nodes = 10
Number of split nodes = 0
RINFOG(1) Operations during elimination (estim)= 3.095D+13
MEMORY ESTIMATIONS ...
Estimations with standard Full-Rank (FR) factorization:
Maximum estim. space in Mbytes, IC facto. (INFOG(16)): 18445
Total space in MBytes, IC factorization (INFOG(17)): 54565
Maximum estim. space in Mbytes, OOC facto. (INFOG(26)): 12236
Total space in MBytes, OOC factorization (INFOG(27)): 35157
Elapsed time in analysis driver= 9.6259
Analysis time by clock_gettime(): 9.626 s
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 7 1391349 32961525
executing #MPI = 3 and #OMP = 8
Elapsed time in save structure driver= 0.0004
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
On return from DMUMPS, INFOG(1)= -71
On return from DMUMPS, INFOG(2)= 0
PRE FACTO START LPROF----------------------
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 2 1391349 32961525
executing #MPI = 3 and #OMP = 8
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
****** FACTORIZATION STEP ********
GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
Number of working processes = 3
ICNTL(22) Out-of-core option = 0
ICNTL(35) BLR activation (eff. choice) = 0
ICNTL(37) BLR CB compression (eff. choice) = 0
ICNTL(40) Mixed/adaptive precision BLR = 0
ICNTL(49) Compact workarray S (end facto.) = 0
ICNTL(56) Effective value during facto. = 0
ICNTL(14) Memory relaxation = 30
INFOG(3) Real space for factors (estimated)= 2956130745
INFOG(4) Integer space for factors (estim.)= 19017628
Maximum frontal size (estimated) = 25827
Number of nodes in the tree = 32938
ICNTL(23) Memory allowed (value on host) = 0
Sum over all procs = 0
Memory provided by user, sum of LWK_USER = 0
Effective threshold for pivoting, CNTL(1) = 0.1000D-01
* [MAQAO] Info: STARTING COUNTERS (kairoscomp22)
[0m
Statistics on the scaling phase
Elapsed time for scaling = 0.0816
Max difference from 1 after scaling the entries for ONE-NORM (option 7/8) = 0.28D+00
Average Effective size of S (based on INFO(39))= 1674700192
Elapsed time to reformat/distribute matrix = 0.5211
Allocated buffers
------------------
Size of reception buffer in bytes ...... = 39603204
Size of async. emission buffer (bytes).. = 158808844
Small emission buffer (bytes) .......... = 556
** Memory allocated, max in Mbytes (INFOG(18)): 18445
** Memory allocated, total in Mbytes (INFOG(19)): 54564
** Memory effectively used, max in Mbytes (INFOG(21)): 15863
** Memory effectively used, total in Mbytes (INFOG(22)): 43271
Flops under L0 layer (avg/max across MPI) = 6.818D+11 7.783D+11
Elapsed time under L0 (avg/max across MPI) = 2.1174 2.2966
Elapsed time to process root node = 11.4064
Elapsed time for factorization = 176.4569
Leaving factorization with ...
RINFOG (2) Operations in node assembly = 9.039D+09
------ (3) Operations in node elimination = 3.095D+13
ICNTL (8) Scaling effectively used = 7
INFOG (9) Real space for factors = 2956137604
INFOG (10) Integer space for factors = 18945437
INFOG (11) Maximum front size = 25827
INFOG (29) Number of entries in factors = 2796208095
INFOG (12) Number of negative pivots = 0
INFOG (13) Number of delayed pivots = 0
Number of 2x2 pivots in type 1 nodes = 0
Number of 2x2 pivots in type 2 nodes = 0
RINFOG(19) Smallest pivot WITH perturbed pivots = 1.403D-03
RINFOG(20) Smallest pivot WITHOUT perturbed pivots = 1.403D-03
RINFOG(21) Largest pivot in absolute value = 1.000D+00
INFOG (24) Effective value of ICNTL(12) = 1
INFOG (14) Number of memory compress = 2
Elapsed time in factorization driver = 177.0668
Factorization time by clock_gettime(): 177.0660 s
Entering DMUMPS 5.9.0 from C interface with JOB = -2
executing #MPI = 3 and #OMP = 8
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 1 1391349 32961525
executing #MPI = 3 and #OMP = 16
=================================================
MUMPS compiled with option -Dmetis
MUMPS compiled with option -Dpord
MUMPS compiled with option -Dptscotch
MUMPS compiled with option -Dscotch
This MUMPS version includes code for mixed precision BLR including memory accessors during solve
=================================================
L D L^T Solver for general symmetric matrices
Type of parallelism: Working host
****** ANALYSIS STEP ********
Processing a graph of size: 1391349
Average density of rows/columns = 45
Ordering based on METIS
ELAPSED TIME SPENT IN METIS reordering = 8.3890
SYMBOLIC based on column counts
ELAPSED TIME IN symbolic factorization = 0.5070
A root of estimated size 15552 has been selected for Scalapack.
Leaving analysis phase with ...
INFOG(1) = 0
INFOG(2) = 0
-- (20) Number of entries in factors (estim.) = 2796208095
-- (3) Real space for factors (estimated) = 2956130745
-- (4) Integer space for factors (estimated) = 19017628
-- (5) Maximum frontal size (estimated) = 25827
-- (6) Number of nodes in the tree = 32938
-- (32) Type of analysis effectively used = 1
-- (7) Ordering option effectively used = 5
ICNTL (6) Maximum transversal option = 0
ICNTL (7) Pivot order option = 7
ICNTL(12) Ordering symmetric indef. matrices = 1
ICNTL(13) Parallelism/splitting of root node = 0
ICNTL(14) Percentage of memory relaxation = 30
ICNTL(15) Analysis by block effectively used = 0
ICNTL(18) Distributed input matrix (on if >0) = 0
ICNTL(32) Forward elimination during facto. = 0
ICNTL(35) BLR activation = 0
ICNTL(48) Tree based multithreading (effective)= 1
ICNTL(58) Symbolic factorization option = 2
Number of level 2 nodes = 10
Number of split nodes = 0
RINFOG(1) Operations during elimination (estim)= 3.095D+13
MEMORY ESTIMATIONS ...
Estimations with standard Full-Rank (FR) factorization:
Maximum estim. space in Mbytes, IC facto. (INFOG(16)): 18388
Total space in MBytes, IC factorization (INFOG(17)): 54309
Maximum estim. space in Mbytes, OOC facto. (INFOG(26)): 11931
Total space in MBytes, OOC factorization (INFOG(27)): 34512
Elapsed time in analysis driver= 9.6045
Analysis time by clock_gettime(): 9.605 s
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 7 1391349 32961525
executing #MPI = 3 and #OMP = 16
Elapsed time in save structure driver= 0.0004
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
On return from DMUMPS, INFOG(1)= -71
On return from DMUMPS, INFOG(2)= 0
PRE FACTO START LPROF----------------------
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 2 1391349 32961525
executing #MPI = 3 and #OMP = 16
****** FACTORIZATION STEP ********
GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
Number of working processes = 3
ICNTL(22) Out-of-core option = 0
ICNTL(35) BLR activation (eff. choice) = 0
ICNTL(37) BLR CB compression (eff. choice) = 0
ICNTL(40) Mixed/adaptive precision BLR = 0
ICNTL(49) Compact workarray S (end facto.) = 0
ICNTL(56) Effective value during facto. = 0
ICNTL(14) Memory relaxation = 30
INFOG(3) Real space for factors (estimated)= 2956130745
INFOG(4) Integer space for factors (estim.)= 19017628
Maximum frontal size (estimated) = 25827
Number of nodes in the tree = 32938
ICNTL(23) Memory allowed (value on host) = 0
Sum over all procs = 0
Memory provided by user, sum of LWK_USER = 0
Effective threshold for pivoting, CNTL(1) = 0.1000D-01
* [MAQAO] Info: STARTING COUNTERS (kairoscomp22)
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
[0m
Statistics on the scaling phase
Elapsed time for scaling = 0.0623
Max difference from 1 after scaling the entries for ONE-NORM (option 7/8) = 0.28D+00
Average Effective size of S (based on INFO(39))= 1744635224
Elapsed time to reformat/distribute matrix = 0.5226
Allocated buffers
------------------
Size of reception buffer in bytes ...... = 39603204
Size of async. emission buffer (bytes).. = 158808844
Small emission buffer (bytes) .......... = 556
** Memory allocated, max in Mbytes (INFOG(18)): 18388
** Memory allocated, total in Mbytes (INFOG(19)): 54309
** Memory effectively used, max in Mbytes (INFOG(21)): 15707
** Memory effectively used, total in Mbytes (INFOG(22)): 42628
Flops under L0 layer (avg/max across MPI) = 4.144D+11 5.012D+11
Elapsed time under L0 (avg/max across MPI) = 0.8574 0.9653
Elapsed time to process root node = 11.4020
Elapsed time for factorization = 180.8395
Leaving factorization with ...
RINFOG (2) Operations in node assembly = 9.039D+09
------ (3) Operations in node elimination = 3.095D+13
ICNTL (8) Scaling effectively used = 7
INFOG (9) Real space for factors = 2956137604
INFOG (10) Integer space for factors = 18945437
INFOG (11) Maximum front size = 25827
INFOG (29) Number of entries in factors = 2796208095
INFOG (12) Number of negative pivots = 0
INFOG (13) Number of delayed pivots = 0
Number of 2x2 pivots in type 1 nodes = 0
Number of 2x2 pivots in type 2 nodes = 0
RINFOG(19) Smallest pivot WITH perturbed pivots = 1.403D-03
RINFOG(20) Smallest pivot WITHOUT perturbed pivots = 1.403D-03
RINFOG(21) Largest pivot in absolute value = 1.000D+00
INFOG (24) Effective value of ICNTL(12) = 1
INFOG (14) Number of memory compress = 2
Elapsed time in factorization driver = 181.4316
Factorization time by clock_gettime(): 181.4307 s
Entering DMUMPS 5.9.0 from C interface with JOB = -2
executing #MPI = 3 and #OMP = 16
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 1 1391349 32961525
executing #MPI = 3 and #OMP = 32
=================================================
MUMPS compiled with option -Dmetis
MUMPS compiled with option -Dpord
MUMPS compiled with option -Dptscotch
MUMPS compiled with option -Dscotch
This MUMPS version includes code for mixed precision BLR including memory accessors during solve
=================================================
L D L^T Solver for general symmetric matrices
Type of parallelism: Working host
****** ANALYSIS STEP ********
Processing a graph of size: 1391349
Average density of rows/columns = 45
Ordering based on METIS
ELAPSED TIME SPENT IN METIS reordering = 8.3953
SYMBOLIC based on column counts
ELAPSED TIME IN symbolic factorization = 0.5075
A root of estimated size 15552 has been selected for Scalapack.
Leaving analysis phase with ...
INFOG(1) = 0
INFOG(2) = 0
-- (20) Number of entries in factors (estim.) = 2796208095
-- (3) Real space for factors (estimated) = 2956130745
-- (4) Integer space for factors (estimated) = 19017628
-- (5) Maximum frontal size (estimated) = 25827
-- (6) Number of nodes in the tree = 32938
-- (32) Type of analysis effectively used = 1
-- (7) Ordering option effectively used = 5
ICNTL (6) Maximum transversal option = 0
ICNTL (7) Pivot order option = 7
ICNTL(12) Ordering symmetric indef. matrices = 1
ICNTL(13) Parallelism/splitting of root node = 0
ICNTL(14) Percentage of memory relaxation = 30
ICNTL(15) Analysis by block effectively used = 0
ICNTL(18) Distributed input matrix (on if >0) = 0
ICNTL(32) Forward elimination during facto. = 0
ICNTL(35) BLR activation = 0
ICNTL(48) Tree based multithreading (effective)= 1
ICNTL(58) Symbolic factorization option = 2
Number of level 2 nodes = 10
Number of split nodes = 0
RINFOG(1) Operations during elimination (estim)= 3.095D+13
MEMORY ESTIMATIONS ...
Estimations with standard Full-Rank (FR) factorization:
Maximum estim. space in Mbytes, IC facto. (INFOG(16)): 18131
Total space in MBytes, IC factorization (INFOG(17)): 53828
Maximum estim. space in Mbytes, OOC facto. (INFOG(26)): 11662
Total space in MBytes, OOC factorization (INFOG(27)): 33569
Elapsed time in analysis driver= 9.6122
Analysis time by clock_gettime(): 9.612 s
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 7 1391349 32961525
executing #MPI = 3 and #OMP = 32
Elapsed time in save structure driver= 0.0004
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
On return from DMUMPS, INFOG(1)= -71
On return from DMUMPS, INFOG(2)= 0
PRE FACTO START LPROF----------------------
Entering DMUMPS 5.9.0 from C interface with JOB, N, NNZ = 2 1391349 32961525
executing #MPI = 3 and #OMP = 32
****** FACTORIZATION STEP ********
* [MAQAO] Info: STARTING COUNTERS (kairoscomp22)
GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
Number of working processes = 3
[0m ICNTL(22) Out-of-core option = 0
ICNTL(35) BLR activation (eff. choice) = 0
ICNTL(37) BLR CB compression (eff. choice) = 0
ICNTL(40) Mixed/adaptive precision BLR = 0
ICNTL(49) Compact workarray S (end facto.) = 0
ICNTL(56) Effective value during facto. = 0
ICNTL(14) Memory relaxation = 30
INFOG(3) Real space for factors (estimated)= 2956130745
INFOG(4) Integer space for factors (estim.)= 19017628
Maximum frontal size (estimated) = 25827
Number of nodes in the tree = 32938
ICNTL(23) Memory allowed (value on host) = 0
Sum over all procs = 0
Memory provided by user, sum of LWK_USER = 0
Effective threshold for pivoting, CNTL(1) = 0.1000D-01
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
** ERROR RETURN ** FROM DMUMPS INFO(1)= -71
** INFO(2)= 0
PRE FACTO START LPROF----------------------
Statistics on the scaling phase
Elapsed time for scaling = 0.0573
Max difference from 1 after scaling the entries for ONE-NORM (option 7/8) = 0.28D+00
Average Effective size of S (based on INFO(39))= 1828144725
Elapsed time to reformat/distribute matrix = 0.5161
Allocated buffers
------------------
Size of reception buffer in bytes ...... = 39603204
Size of async. emission buffer (bytes).. = 158808844
Small emission buffer (bytes) .......... = 556
** Memory allocated, max in Mbytes (INFOG(18)): 18131
** Memory allocated, total in Mbytes (INFOG(19)): 53828
** Memory effectively used, max in Mbytes (INFOG(21)): 15484
** Memory effectively used, total in Mbytes (INFOG(22)): 41684
Flops under L0 layer (avg/max across MPI) = 2.018D+11 2.452D+11
Elapsed time under L0 (avg/max across MPI) = 0.3980 0.4997
Elapsed time to process root node = 11.3795
Elapsed time for factorization = 183.6526
Leaving factorization with ...
RINFOG (2) Operations in node assembly = 9.039D+09
------ (3) Operations in node elimination = 3.095D+13
ICNTL (8) Scaling effectively used = 7
INFOG (9) Real space for factors = 2956137604
INFOG (10) Integer space for factors = 18945437
INFOG (11) Maximum front size = 25827
INFOG (29) Number of entries in factors = 2796208095
INFOG (12) Number of negative pivots = 0
INFOG (13) Number of delayed pivots = 0
Number of 2x2 pivots in type 1 nodes = 0
Number of 2x2 pivots in type 2 nodes = 0
RINFOG(19) Smallest pivot WITH perturbed pivots = 1.403D-03
RINFOG(20) Smallest pivot WITHOUT perturbed pivots = 1.403D-03
RINFOG(21) Largest pivot in absolute value = 1.000D+00
INFOG (24) Effective value of ICNTL(12) = 1
INFOG (14) Number of memory compress = 2
Elapsed time in factorization driver = 184.2333
Factorization time by clock_gettime(): 184.2324 s
Entering DMUMPS 5.9.0 from C interface with JOB = -2
executing #MPI = 3 and #OMP = 32
