Yeay, the summer semester is over! That means I can get some work done. Currently my push is towards the User Interface, little messages are great for me, but make demonstrations no good.
The User interface I'm developing is in GTK. I expect to have it done by the end of the semester.
After that, I'm going to develop some 'between task' copy semantics for GObjects. My current thinking strongly suggests that automatic load balancing will prove more useful than work on a barrier algorithm. In other words, figure out which models need more time to run and move them around instead of worrying over how to efficiantly backtrack when the load balance is off.
I expect this to take most of Fall Semester, but it promises to be fun!
Hey All,
It's finals week here at UCO. The project went through a successfull proof of concept, but we're now working on a mechanism by which to copy GObjects between MPI Tasks. I've been working on that for a couple of weeks, and am making good headway. The GObject code is found in ./gobject_move_test
The meeting with Dr. Turner went well today. He approved of our work to date, and suggested we go with an interpreter to toss the models around with, rather than a complex NFS layout.
This alters the time table, but I am strongly inclined to bow to his wisdom in this matter.
Ray and Mike are making progress. Mostly their project is still being designed, but it requires a lot of thought. They are trying to design an associative database for use with a NLP expert system/Inductive learning system.
Dr. Turner also suggested that EVERYTHING we read needs to go in our annotated bibliography.
I spent all yesterday reviewing CORBA. It offers facilities many facilities which would be usefull for the modles implementation, but it 'externalises' objects by just shipping off their data.
In short, it would be usefull only in the case of the RM/DE -- MT communication. For that, we have already settled on SOAP.
The functionality we need is often handled by using a dedicated file server whereby all the nodes do disk IO accross the network. My recomendation is we impliment such a solution as well. Principally for ease of development. New setups will be more difficult, but only slightly so.
Group Meeting
Date 02/05/2006
Time 3:00 PM
Plase Library
Purpose Work Validation
Duration 1-2 hrs
Our meeting with Dr. Turner went well today. Everyone seems in high spirits and both projects are making progress. Points of interest are as follows:
We are rescheduling our presentation for conferences next year. A student who is graduating in december voiced concern, but Dr. Turner assured her that even having a paper in submission was excellent for an application. Another student was concerned that we might be procrastinating. This consern was assuaged by noting the fact that this was not an excuse to work less, and should a project be ready, it should then apply for earlier conferences.
The rational behind this change in deadline is the conjunction of three environmental factors. Conference submission dates for this year are very close; it is better to present a solid project rather than one that barely works; the University won't have us budgeted till after July.
Dr. Turner asserts that the only database engines avialable for use on school machines will be mySQL and Oracle. These are already installed and mySQL implements ANSI standard SQL, well, at least it understands it.
Mike and Ray's project has a working user interface up, and their design work on the backend is progressing well.
Distsim's progress is most accurately reported in this task list.
The Wednesday 1 Jan 2006 meeting has been rescheduled to Sunday 5 Jan 2006 at 15:00. Location TBA
We had a meeting with Dr. Turner today. First real one of the semester. He's still trying to find conferences for us to present at.
Status reports need to be delivered weekly and we are responcible for keeping an archive of them.
M and W 10:00 AM - 1:00 PM (Class)
M 1:00 PM - 2:00 (class)
M 2:00 PM - 3:00 (Group Study)
W 1:00 PM - 3:00 (Group Study)
T and Th 3:00 PM - 6:00 PM (Class)
T and Th 6:00AM - 12:00 PM (Work)
Friday 6:00AM - 4:00 PM (Work)
MWF
09:00 - 09:50 Matrix Algebra
10:00 - 10:50 File Structures
13:00 - 15:00 Dist Sim Meeting
TR
09:30 - 10:45 Number Theory
The group will be meeting Monday's and Wednesday's at the UCO library from 13:00 till 14:00 for the duration of the semester.
Contact Charles or Farhanna for location the day of.