OS: Linux Ubuntu 24.04.4 LTS
Executable: yafu-linux-avx512ifma v3.1.6 from the release page
CADO-NFS: commit 492429b4 (pre-flask)
Hello! I am using the factor function in YAFU to factorize numbers and cado when GNFS is needed. Most of the time it works fine, but sometimes it tries to find too many relations (>150M for a C120-ish) by scheduling more sieving work.
Today, I found another interesting behavior during the LA stage:
nfs: calling convert_poly to create nfs.fb from c*.poly
nfs: appending CADO relations into nfs.dat
nfs: now have 16668530 relations
nfs: commencing msieve filtering
5944070977399055440935383920361613677788954767864157757067793236516673309897160022711000348266084552715481677184489441579
commencing relation filtering
estimated available RAM is 31869.0 MB
commencing duplicate removal, pass 1
read 10M relations
found 4113179 hash collisions in 16781543 relations
added 1150 free relations
commencing duplicate removal, pass 2
found 4683419 duplicates and 12099274 unique relations
memory use: 90.6 MB
reading ideals above 720000
commencing singleton removal, initial pass
memory use: 376.5 MB
reading all ideals from disk
memory use: 327.1 MB
keeping 13604979 ideals with weight <= 200, target excess is 116147
commencing in-memory singleton removal
begin with 12099274 relations and 13604979 unique ideals
reduce to 3646982 relations and 3434978 ideals in 21 passes
max relations containing the same ideal: 79
removing 444594 relations and 405957 ideals in 38637 cliques
checking relations array at location 1
commencing in-memory singleton removal
begin with 3202388 relations and 3434978 unique ideals
reduce to 3155184 relations and 2981012 ideals in 10 passes
max relations containing the same ideal: 69
removing 325645 relations and 287008 ideals in 38637 cliques
checking relations array at location 1
commencing in-memory singleton removal
begin with 2829539 relations and 2981012 unique ideals
reduce to 2799542 relations and 2663608 ideals in 10 passes
max relations containing the same ideal: 65
relations with 0 large ideals: 798
relations with 1 large ideals: 9474
relations with 2 large ideals: 81505
relations with 3 large ideals: 335898
relations with 4 large ideals: 734929
relations with 5 large ideals: 862941
relations with 6 large ideals: 545805
relations with 7+ large ideals: 228192
commencing 2-way merge
reduce to 1514252 relation sets and 1378318 unique ideals
ignored 1 oversize relation sets
commencing full merge
memory use: 142.5 MB
found 658534 cycles, need 650518
weight of 650518 cycles is about 58600149 (90.08/cycle)
distribution of cycle lengths:
1 relations: 61998
2 relations: 55865
3 relations: 56626
4 relations: 53470
5 relations: 50496
6 relations: 44683
7 relations: 41884
8 relations: 36933
9 relations: 32229
10+ relations: 216334
heaviest cycle: 28 relations
commencing cycle optimization
start with 5180332 relations
pruned 137686 relations
memory use: 161.3 MB
distribution of cycle lengths:
1 relations: 61998
2 relations: 57084
3 relations: 58688
4 relations: 54948
5 relations: 51676
6 relations: 45703
7 relations: 42747
8 relations: 37327
9 relations: 32493
10+ relations: 207854
heaviest cycle: 28 relations
RelProcTime: 118
Elapsed time is now 2957.6750 seconds.
nfs: commencing msieve linear algebra
commencing linear algebra
read 650518 cycles
cycles contain 2633817 unique relations
read 2633817 relations
using 20 quadratic characters above 4294917295
building initial matrix
memory use: 330.8 MB
read 650518 cycles
matrix is 650336 x 650518 (241.3 MB) with weight 74761232 (114.93/col)
sparse part has weight 55458192 (85.25/col)
filtering completed in 2 passes
matrix is 649345 x 649527 (241.2 MB) with weight 74706006 (115.02/col)
sparse part has weight 55436030 (85.35/col)
matrix starts at (0, 0)
matrix is 649345 x 649527 (241.2 MB) with weight 74706006 (115.02/col)
sparse part has weight 55436030 (85.35/col)
saving the first 48 matrix rows for later
matrix includes 64 packed rows
matrix is 649297 x 649527 (232.0 MB) with weight 62033477 (95.51/col)
sparse part has weight 54317310 (83.63/col)
using block size 8192 and superblock size 2359296 for processor cache size 24576 kB
commencing Lanczos iteration (15 threads)
memory use: 212.4 MB
VBITS = 64
linear algebra at 0.5%, ETA 0h 3m649527 dimensions (0.5%, ETA 0h 3m)
^Cnear algebra completed 5691791 of 649527 dimensions (876.3%, ETA 357h22m)
Received signal 2... please wait
If you quit again, YAFU will exit immediately but you will LOSE ALL UNSAVED PROGRESS.
setting flag
^C
The LA stage runs indefinitely. In both of these scenarios running python3 cado-nfs.py {number} works perfectly fine.
OS: Linux Ubuntu 24.04.4 LTS
Executable: yafu-linux-avx512ifma v3.1.6 from the release page
CADO-NFS: commit 492429b4 (pre-flask)
Hello! I am using the
factorfunction in YAFU to factorize numbers and cado when GNFS is needed. Most of the time it works fine, but sometimes it tries to find too many relations (>150M for a C120-ish) by scheduling more sieving work.Today, I found another interesting behavior during the LA stage:
The LA stage runs indefinitely. In both of these scenarios running
python3 cado-nfs.py {number}works perfectly fine.