User talk:Pzgesvd

bidiagonal reduction code function matching

DGEQRT --- LQR1

DTSQRT --- LQR2

DLARTB --- LUP1

DSSRFT --- LUP2

DGEQRT --- RQR1

DTSQRT --- RQR2

DLARTB --- RUP1

DSSRFT --- RUP2

delete lines from files using sed

sed -ie '1,11d' dgetrf.c

original files are backed up in the .ce files

refer to [1]

low cost stack for function call?

can the stacking operation of the dag chasing be reduced so the overhead is minimized?

clapack3.1.1 blas testing handtune

in the testing routines, eps is wrongly calculated as 1e-19, which should be 1e-7. In each files (dblat3.c, for example), a new piece of eps code is inserted.

For double precision (complex16), problem exists for not accurate element result.

a recall of all the work that I've done

(1) Complex version SVD for ScaLAPACK (done)

(2) RFP for LAPACK (paper to read)

(3) Performance comparison between PLAPACK & SCALAPACK

(4) Re-write PDGETRF, optimization with look-ahead and other threading method for better performance on multi-core cluster (potential topic direction)

(5) CELL learning (done)

(6) Variations of LU, Chol and QR for LAPACK (report)

(7) CLAPACK conversion (report)

(8) ScaLAPACK auto-tuning (potential topic direction)

(9) DAG automatic generation for LAPACK

(10) Using random butterfly transformation to remove pivoting

(11) ILP64 support for LAPACK, ScaLAPACK

(12) tiled LU without pivoting(code, paper)

(13) PLASMA testing

What?

Pardon my asking, but what is all this you're doing with this page? — User:Sean Fennel@ 00:58, 10 September 2008 (EDT)

I'm using it as a diary for my work. Isn't this a place for this kind of work?

Not really — unless your work happens to be documenting every wiki on the Web. — User:Sean Fennel@ 02:11, 25 December 2008 (EST)