iOpenShell » Q-Chem specific questions » What's "Collapsing Subspace" in freq calc?

What's "Collapsing Subspace" in freq calc?

Page: 1

Author Post
Member
Registered: Mar 2012
Posts: 10
When doing CIS freq calculations, sometimes I meet something like this,


Calculating MO derivatives via CPSCF
Max. subspace vectors = 32410
Dim. of subspace vectors = 32410
Room left in mega-array = 1250000000

Q-Chem warning in module /home/scratch/svnadmin/28838_zgan/qchem/libcis/ov_dim.C, line 124:

Static memory may be insufficient to store the subspace vectors

1 3 87 0.165436 0.021410
2 3 87 0.042001 0.006653
3 3 87 0.014686 0.003535
4 3 87 0.109605 0.039371
5 6 84 0.002168 0.000650
6 40 50 0.000446 0.000132 Collapsing Subspace
7 47 43 0.000292 0.000090
8 47 43 0.000289 0.000089
9 47 43 0.000288 0.000089
10 47 43 0.000287 0.000088
11 47 43 0.000286 0.000088
12 47 43 0.000285 0.000088
13 47 43 0.000284 0.000087
14 47 43 0.000283 0.000087
15 47 43 0.000283 0.000087
16 47 43 0.000282 0.000087 Collapsing Subspace
17 47 43 0.000281 0.000087
18 47 43 0.000281 0.000086
19 47 43 0.000280 0.000086
20 47 43 0.000280 0.000086
21 47 43 0.000279 0.000086
22 47 43 0.000278 0.000086
23 47 43 0.000278 0.000085
24 47 43 0.000277 0.000085
25 47 43 0.000276 0.000085
26 47 43 0.000276 0.000085 Collapsing Subspace
27 47 43 0.000275 0.000085
28 44 46 0.000287 0.000084
29 44 46 0.000289 0.000084
30 44 46 0.000289 0.000084
31 42 48 0.000315 0.000084
32 42 48 0.000315 0.000084
33 42 48 0.000315 0.000084
34 42 48 0.000314 0.000084
35 42 48 0.000314 0.000083 Collapsing Subspace
36 10 80 0.002332 0.000162
37 0 90 0.845037 0.062012
38 0 90 0.834545 0.061242
39 0 90 0.824771 0.060514
40 0 90 0.812677 0.059639 Collapsing Subspace
41 0 90 0.749758 0.054689
42 0 90 1.536799 0.136364
43 0 90 0.769916 0.064630
44 0 90 0.765306 0.064041
45 0 90 0.761067 0.063235 Collapsing Subspace
46 0 90 1.034639 0.089347
47 0 90 0.523993 0.040062
48 0 90 0.512340 0.038656
49 0 90 0.493512 0.037075
50 0 90 0.484629 0.036440 Collapsing Subspace
51 0 90 0.510355 0.038160
52 0 90 1.193711 0.074070
53 0 90 1.066863 0.060538
54 0 90 0.915212 0.050088
55 0 90 0.819905 0.043357 Collapsing Subspace
56 0 90 0.842482 0.044935
57 0 90 0.551278 0.027048

The rightmost column (I'm not sure what it means) seems need to be small to finish this part. When I meet "collapsing subspace", usually it never gets small again so never ends. What's the solution here, more cycles allowed or more memories needed?
The input for this one is,

$molecule
0 2
C 1.1700849797 0.7065679400 -0.0514524548
C 0.7688597009 1.9958073312 -0.2170014587
C -0.6500001338 1.9495329612 -0.2011005570
C -1.0425649019 0.6528402134 -0.0375884513
O 0.0929821972 -0.0963604332 0.0519255396
H 1.3855194975 2.8589651607 -0.3526710028
O -1.4534621702 3.0582726405 -0.3344977783
C -2.3210665967 -0.0164791028 0.0598228726
C -2.3794499014 -1.4089212499 0.2796739518
C -3.5325286482 0.6984441495 -0.0613281486
C -3.6079419497 -2.0589414256 0.3741818946
H -1.4674775835 -1.9662389696 0.3776038382
C -4.7552901811 0.0392442971 0.0352492327
H -3.5236446959 1.7594562504 -0.2315033240
C -4.8017403359 -1.3414512174 0.2529743047
H -3.6333696858 -3.1209718147 0.5430485226
H -5.6677929127 0.6005890226 -0.0599084789
H -5.7482442529 -1.8464772404 0.3268254725
C 2.5017871545 0.0756494080 0.0046249490
C 2.6746229601 -1.2435248706 -0.4068768245
C 3.6032839661 0.7961345984 0.4592005064
C 3.9324424164 -1.8291116364 -0.3678633096
H 1.8286243242 -1.8048436340 -0.7574309518
C 4.8617048114 0.2114200388 0.4878919205
H 3.4770013756 1.8065433274 0.8050027543
C 5.0294700764 -1.1031429644 0.0753697569
H 4.0543553985 -2.8487710558 -0.6877526415
H 5.7044930895 0.7784274518 0.8414851057
H 6.0036643009 -1.5580168129 0.1028924307
$end

$rem
MEM_TOTAL 50000
MEM_STATIC 10000
JOBTYPE freq
SYMMETRY FALSE
EXCHANGE HF
CORRELATION none (built-in)
BASIS 6-31g**
cis_state_deriv 1
cis_n_roots 3
max_scf_cycles=100
max_cis_cycles=100
$end


It's still running, not sure whether gonna last forever, but I did meet some similar cases which failed, not at hand right now.
Member
Registered: Mar 2012
Posts: 10
It fails eventually.

58 0 90 0.576923 0.029608
59 0 90 0.584615 0.029666
60 0 90 0.577705 0.029016 Collapsing Subspace
61 0 90 0.580634 0.029405
62 0 90 0.558600 0.029560
63 0 90 0.578858 0.031469
64 0 90 0.591862 0.032650
65 0 90 0.589937 0.033361 Collapsing Subspace
66 0 90 0.592978 0.033578
67 0 90 0.590056 0.032496
68 0 90 0.566242 0.032361
69 0 90 0.556536 0.031600
70 0 90 0.554643 0.031659 Collapsing Subspace
71 0 90 0.558516 0.031935
72 0 90 0.532145 0.028931
73 0 90 0.528083 0.028244
74 0 90 0.537520 0.028878
75 0 90 0.535222 0.028474 Collapsing Subspace
76 0 90 0.537759 0.028604
77 0 90 0.760986 0.042268
78 0 90 0.693759 0.033097
79 0 90 0.638814 0.029651
80 0 90 0.615748 0.030026 Collapsing Subspace
81 0 90 0.617944 0.030072
82 0 90 3.201570 0.266779
83 0 90 1.839134 0.145889
84 0 90 1.493322 0.114051
85 0 90 1.238527 0.090184 Collapsing Subspace
86 0 90 2.854950 0.218294
87 0 90 0.975538 0.065074
88 0 90 0.618021 0.029933
89 0 90 0.606317 0.030622
90 0 90 0.584955 0.029860 Collapsing Subspace
91 0 90 0.589429 0.029970
92 0 90 0.952149 0.044177
93 0 90 0.613033 0.031974
94 0 90 0.576539 0.031049
95 0 90 0.559526 0.029995 Collapsing Subspace
96 0 90 1.104816 0.052265
97 0 90 0.890916 0.038747
98 0 90 1.081061 0.048923
99 0 90 1.479872 0.075239
100 0 90 3.476016 0.198605 Collapsing Subspace

Q-Chem fatal error occurred in module iterzp.F, line 161, PE 0:

Out of Iterations- IterZ

p0_47945: p4_error: interrupt SIGx: 6
Member
Registered: Mar 2012
Posts: 10
By setting cpscf_nseg to more than 1 I solved the problem but get a new one.
   1         1          44      0.004115    0.000433
2 4 41 0.000939 0.000097
3 18 27 0.000186 0.000027
4 33 12 0.000039 0.000006
5 41 4 0.000005 0.000002
6 45 0 0.000000 0.000000 Roots Converged
CPCIS time: CPU 2335.74 s wall 2338.07 s

Q-Chem fatal error occurred in module /home/scratch/svnadmin/28838_zgan/qchem/libmdc/newfileman.C, line 289:

Error reading in TMP file 27/0 (130448)

p0_56926: p4_error: interrupt SIGx: 6

Is this a memory problem? I tried to chenge the memory settings to
MEM_TOTAL       20000
MEM_STATIC 5000

or the default, but it doesn't help. Any suggestion? Thanks in advance.
Member
Registered: Oct 2016
Posts: 1
tcya, did you ever figure out what collapsing subspace means? I am encountering the same issue.

Page: 1

iOpenShell » Q-Chem specific questions » What's "Collapsing Subspace" in freq calc?