source: trunk/ga_docs/requirements/thoughts.txt @ 7935

Last change on this file since 7935 was 3060, checked in by duncan, 18 years ago

moving text to the correct document

File size: 2.2 KB
Line 
1QA: Have SMEC (and the likes) try the software for one of their projects
2and ask them to comment on:
3
4  quality
5  performance (size and speed)
6  ease of installation
7  documentation
8 
9
10  NAME:  How about naming the whole thing  GANU or ANUGA  ?
11 
12 
13PROJECTS: 
14 
15One way of implementing parallelism would be through domain decomposition (DD).
16Perhaps we can work with someone with experience in DD of PDEs and once that
17works, parallelise the code by having several sub-domains on each processor.
18
19
20Uncertainty modelling and MC simluations.
21
22Running a large simulation (Stress test): {Peter Row?) DONE (tsunamies)
23
24Momentum sink study. Scale, extent,
25 map various copmpositions of structures to friction terms     
26 
27 
28Stephen: Look at domain decomposition algorithms with a view of
29parallelising pyvolution
30
31
32Stephen: Timestepping to be controlled by error estimator.
33
34
35
36DESIGN PRINCIPLES (17/3/5) -  as agreed to by Duncan and Ole
37
38Objects shared across project should not precompute
39structures upon initialisation unless they are used by all clients.
40Rather precompute on demand or split using inheritance.
41Pro: Good separation of concepts
42Con: It may be harder to find the right file
43
44Also, modules shared across project should not generally depend on other modules.
45
46
47
48Idea (ON 28/4/5):
49Allow set_quantity to take a .pts file as argument. That should use least_squares to populate the quantity.
50Furthermore use caching.py to assert whether this should be computed or read from cache.
51
52
53Use logger module as suggested by steve
54
55See also HUSK.txt in pyvolution
56
57
58Have the option of changing boundaries and forcing functions on the fly depnding on e.g. time or value of conserved quantities.
59
60
61PROJECT IDEAS (NOV 2005)
62
63GBR Effect on tsunami (Stephen may get student)
64Convection (Grad proposal 2005/2006, Ole)
65Momentum sink (Grad proposal 2005/2006, Duncan)
66Convergence study (Ole)
67VRML export (David Beard, Malcolm Nicoll)
68Proper installer
69Overcome 2GB NetCDF limit
70IO (data_manager) needs some TLC
71
72See also work program ideas 2005-2006
73
74How to get the same amount of smoothly from a large mesh to a small
75mesh. How does the alpha value get scaled?
76
Note: See TracBrowser for help on using the repository browser.