Rough-n-ready Roofline: Titan V editionIn this post we discuss rules of thumb for performance limiters when using shared memory in a CUDA compute kernel running on a Titan V -...
libParanumal: Galerkin-Boltzmann 3D flow simulationlibParanumal simulation for 3D flow over a finite fence modeled with the Galerkin-Boltzmann flow equations of gas dynamics (Toelke et al...
libParanumal: Galerkin-Boltzmann flow simulationSimulation on 100K quartic triangle elements with discontinuous Galerkin discretization in space and adaptive Runge-Kutta time stepper....
Undergraduate Summer Researchers Join the Paranumal TeamFour VT undergraduates have joined the paranumal team as summer research assistants. From left to right: Nick Polidoro, Dallas Viar,...
Jesse Chan Talk on Entropy Stable Schemes @VTJesse Chan gave a colloquium talk in the Math Department @VT on novel entropy stable flux differencing discontinuous Galerkin...
Finite Element Stiffness Matrix Action: to precompute or not to precomputeQ: does it make sense to partially assembled elemental stiffness matrices for affine tetrahedral finite elements when running on a Volta...
Finite Element Stiffness Matrix Action: to BLAS or not to BLAS, that is the question.BLAS (Basic Linear Algebra Subprograms) is a specification for performing multiple common, basic linear algebra routines. BLAS functions...
Finite Element Stiffness Matrix Action: monolithic kernel optimization on Titan VIn this post we demonstrate how to optimize the performance of a specific finite element operation expressed as a GPU kernel. One might...
Basic GPU optimization strategiesWhen I started writing GPU code, I often heard that using shared memory is the only way to get good performance out of my code. As I kept...
Titan-V @ V-Tech: initial benchmarking resultsA new Titan V arrived at Virginia Tech today. Installation went relatively smoothly thanks to the patience of Bill Reilly. The Titan V...