FS 680 3: Chemical kinetics problems RCHEM radiation chemistry study -- 3rd output time step
[Download ]
[Visualizations ]
[Matrix Statistics ]
[Set Information ]

Mixed kinetics diffusion problem from radiation chemistry. 17 chemical species and one space dimension with 40 mesh points. The diffusion terms are large compared with the kinetics terms. However, 440 columns are identically those of the unit matrix which possibly provides a good test for methods which cannot exploit such a feature properly (for example, methods which assume symmetry).

Help: My browser can't read the compressed data files. What now?

Click on image to get an enlarged version.
Click on label to get an explanation.

Click on label to get an explanation.

Size
Type
680 x 680, 2646 entries
real unsymmetric
Nonzeros
total diagonal below diagonal above diagonal A-A' 2471 680 1247 544 2578

Column Data
Row Data
Average nonzeros per column : 3.6 Standard deviation : 3.9

index nonzeros longest 2 13 shortest 80 1

Average nonzeros per row : 3.6 Standard deviation : 2.2

index nonzeros longest 162 8 shortest 40 1

Bandwidths
Profile Storage
lower 561 upper 281 average |i-j| 1.1e+02 std.dev. 1.4e+02

min max ave. std.dev. lower bandwidth 0 560 2e+02 1.9e+02 upper bandwidth 0 280 37 73

Nonsymmetric skyline storage 158241

Heaviest diagonals
offset from main 0 1 -1 -120 -40 -160 -80 120 -200 -400
nonzeros 680 232 229 157 156 117 117 117 79 78
accumulated percent 27.52 36.91 46.18 52.53 58.84 63.58 68.31 73.05 76.24 79.40

Top 10 out of 20 nonvoid diagonals.
Conditioning
Frobenius norm 1.2e+14 condition number (est.) 4.2e+06 2-norm (est.) 7.2e+13 diagonal dominance no

Set Information
Set FACSIMILE
Source:
Alan Curtis, Computer Science and Systems Division, AERE Harwell.
Discipline: Chemical kinetics
Accession: August 1983

The Matrix Market is a service of the
Mathematical and Computational Sciences Division /
Information Technology Laboratory /
National Institute of Standards and Technology .
[ Home ]
[ Search ]
[ Browse ]
[ Resources ]

Last change in this page: Wed Sep 22 13:33:38 US/Eastern 2004
[Comments:
matrixmarket @ nist.gov
]