Latest news archive:  Sparse Solver

Strand7 Sparse Solver Release Beta 1

May 23, 2002

We are very pleased to announce this new release which offers significant increases in solution speed for all the Strand7 solvers.

During the past 18 months we have focused a great deal of effort into the development of this completely new solver for Strand7. As with the rest of the Strand7 development, the new solver has been developed entirely in-house from scratch and is fully integrated into the Strand7 environment.

The new solver is a Direct Sparse Solver. This means that it provides the same reliable result quality as the existing solver, but in a much shorter time. In other words, the new solver is not iterative so solver convergence is not an issue.

The following list presents a selection of comparative solver run-times for medium to large Strand7 models. All models were run on an PC with an AMD Athlon processor, running at 1.4GHz. The system had 1.5GB of installed RAM, but the important RAM measure is the amount actually used by the solver (as reported in each case below).

If you are interested in obtaining a copy of this new solver, please contact us to check your eligibility for a free update: support@strand7.com



Model Elements Solver Load
Cases
/
Modes
Equations R2.1.0 Skyline Solver New Sparse Solver
RAM
(MB)
Disk
(MB)
Time
(hh:mm:ss)
RAM
(MB)
Initial
Non-
Zeros
(MB)
Final
Non-
Zeros
(MB)
Time
(hh:mm:ss)
91 483 Nodes
1 102 Beams
90 656 Plates
Linear
Static
5 541 242 819 8 436 11:08:35 80 116 783 00:05:44
364 521 Nodes
1 102 Beams
358 786 Plates
Linear
Static
5 2 165 646 NA NA NA 404 470 4 191 00:53:51
807 357 Nodes
1 102 Beams
815 904 Plates
Linear
Static
5 4 808 838 NA NA NA 915 1 042 10 800 04:09:57
807 357 Nodes
1 102 Beams
815 904 Plates
Natural
Frequency
24
modes
4 808 838 NA NA NA 830 1 042 10 800 29:34:25
10 179 Nodes
8 625 Beams
12 781 Plates
Linear
Transient
10
time
steps
60 990 15 300 00:07:40 145 15 145 00:01:30
14 808 Nodes
5 630 Beams
14 529 Plates
Linear
Static
1 86 082 47 736 00:18:26 152 19 152 00:00:50
11 794 Nodes
423 Beams
11 201 Plates
384 Bricks
Non-linear
Transient
100
time
steps
68 896 41 637 48:02:12 77 15 77 00:58:36
29 662 Nodes
3 676 Beams
32 579 Plates
Linear
Static
1 177 186 149 4 151 02:24:02 411 41 709 00:10:21
29 662 Nodes
3 676 Beams
32 579 Plates
Linear
Buckling
20
modes
177 186 149 4 151 13:26:59 411 41 750 00:26:29
19 923 Nodes
4 169 Beams
6 806 Plates
58 808 Bricks
Linear
Static
1 73 887 428 1 045 00:57:01 134 13 133 00:00:44
35 138 Nodes
142 642 Tet4
Natural
Frequency
20
modes
102 764 165 1 003 01:13:30 141 15 141 00:03:59
42 945 Nodes
42 880 Hex8
Linear
Static
2 141 027 179 3 457 06:56:11 177 39 664 00:09:43
83 030 Nodes
18 090 Hex20
Linear
Static
2 244 446 NA NA NA 558 142 2551 01:17:09
91 737 Nodes
393 284 Tet4
Linear
Static
1 257 067 272 7 351 11:24:22 204 43 757 00:13:46
91 737 Nodes
393 284 Tet4
Natural
Frequency
20
modes
275 067 NA NA NA 204 43 759 00:36:34
352 082 Nodes
218 298 Tet10
Linear
Static
1 1 064 373 NA NA NA 1 066* 332 7 265 07:54:32
60 986 Nodes
3 052 Beams
58 467 Plates
Linear
Static
1 348 197 139 6 049 04:50:00 73 76 514 00:02:40
177 565 Nodes
3 052 Beams
58 467 Plates
Linear
Static
1 1 047 671 NA NA NA 431 395 2 515 00:30:45
* These runs will be significantly faster with more RAM.