Thursday, December 26, 2013

Lecture on convective heat transfer Instructuer : Prof. C. J. ChenNumber of lectures : 9Duration : 1hr+ approx.Source : youtube ...
Lecture on TurbulenceInstructuer : Prof. C. J. ChenNumber of lectures : 19Duration : 1hr+ approx.Source : youtube ...
Lecture on Turbulent heat transfer Instructuer : Prof. C. J. Chen Number of lectures : 7 Duration : 1hr approx. Source : youtube ...

Friday, December 13, 2013

Some awesome projects to look out for... MUSCLE 2 MUSCLE 2 - The Multiscale Coupling Library and Environment is a portable framework to do multiscale modeling and simulation on distributed computing resources. The generic coupling mechanism of MUSCLE is suitable for many types of multiscale applications, notably for multiscale models as defined by the  MAPPER project or complex automata as defined in the  COAST project....
Some awesome projects to look out for... OpenPALM In order to efficiently represent complex systems, numerical modelling has to rely on many physical models at a time: an ocean model coupled with an atmospheric model is at the basis of climate modelling; a combustion model coupled with a radiation model allows the computation of a combustion chamber temperature. The continuity of the solution is granted only if these models can constantly...

Tuesday, November 26, 2013

Short LaTeX Tutorial Here is a short introduction to LaTeX that I taught to Surrey PhDs.... Download PDF [636KB]  ...

Monday, November 18, 2013

quickly some ... Quick Reference Cards VI Editor LaTeX C Programming FORTRAN 77 [ Quick Ref  / Quick Tutorial ] FORTRAN 90 BLAS LAPACK MPI C/C++ [ MPI FORTRAN ] OpenMP [ C / FORTRAN ] CUDA [ C / FORTRAN / Quick Tutorial ] OpenCL 2.0 Let me know to add more ......

Wednesday, November 13, 2013

CUDA-accelerated Sparse matrix assembly and solution using CUSP 1. If you use finite element methods for your numerical PDEs, chances are good that at some point in time you need to generate one or more large matrices.2. Of course, chances are also good that you don't. It seems especially prevalent today, in the midst of an explosion of development in computing with massively parallel processors, that "global" variants of finite element methods that require large sparse matrices are shunned in favor of discontinuous, "local" formulations...
Subscribe to RSS Feed Follow me on Twitter!