[CSC 435] Email 3 (sorry)

[Andrew J. Pounds]

Kevin -- in the handout I gave you from the CSC 335 textbook  the ILS 
and DLS algorithm that those authors used employed the augmented 
matrix.  Your code needs to match what I gave you in class for the project:

http://theochem.mercer.edu/csc435/assignments/proj2_2016.pdf

Having an A matrix and B and C vectors will make much easier to call the 
ATLAS library function as well.

As far as what has to be returned, I know we went over this in class 
more than once, but for the sake of clarity (and for the benfit of 
anyone that might have missed it) here it is again:

in addition to the number of threads and the dimension of the system:


*dot* is a double precision function that returns a double (scalar) that 
is the dot product of the two vectors.

*vvm* is a VOID function that takes the pointers vec1, vec2, and rmat 
and builds the rmat matrix

*mvv* is a VOID function that  takes the pointers to a matrix, a vector, 
and a resultant vector (which is computed in the function)

*mmm* is VOID function -- you have my example -- it takes three pointers 
corresponding to the three matrices - it computes rmat

*dls* is a VOID function that takes the A matrix pointer, the b vector, 
and the pointer of the solution vector, which it computes

*ils* is a VOID function that takes the A matrix pointer, the b vector, 
and the pointer of the solution vector, which it computes



On 04/07/2016 02:33 PM, Kevin L. Long wrote:
>
>     Okay I know I just sent 2 other emails, but I really need to know 
> the answer to this. It seems our arguments for ils and dls are:
>
>
>  (int numthreads,int N, double *matrix, double *vec, double*rvec)
>
>
> I thought our matrix was augmented with the last column being the 
> solution vector,and thus we would not need a vector vec. Was I 
> mistaken? Also is rvec for returning a vector, I only ask because 
> technically we were never told whether ils and dls returned anything 
> or not. I know it should be obvious but I don't want to make any 
> assumptions and get lost.
>
>
> Sorry to bother you again
>
> -Kevin
>


-- 
Andrew J. Pounds, Ph.D.  (pounds_aj at mercer.edu)
Professor of Chemistry and Computer Science
Mercer University,  Macon, GA 31207   (478) 301-5627
http://faculty.mercer.edu/pounds_aj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://theochem.mercer.edu/pipermail/csc435/attachments/20160407/b4b22ba5/attachment.html>