Show simple item record

dc.contributor.authorOhannessian, H. Gorune
dc.contributor.authorTurkiyyah, George
dc.contributor.authorAhmadia, Aron
dc.contributor.authorKetcheson, David I.
dc.date.accessioned2019-04-28T13:14:19Z
dc.date.available2019-04-28T13:14:19Z
dc.date.issued2018-05-21
dc.identifier.urihttp://hdl.handle.net/10754/632531
dc.description.abstractWe present cudaclaw, a CUDA-based high performance data-parallel framework for the solution of multidimensional hyperbolic partial differential equation (PDE) systems, equations describing wave motion. cudaclaw allows computational scientists to solve such systems on GPUs without being burdened by the need to write CUDA code, worry about thread and block details, data layout, and data movement between the different levels of the memory hierarchy. The user defines the set of PDEs to be solved via a CUDA- independent serial Riemann solver and the framework takes care of orchestrating the computations and data transfers to maximize arithmetic throughput. cudaclaw treats the different spatial dimensions separately to allow suitable block sizes and dimensions to be used in the different directions, and includes a number of optimizations to minimize access to global memory.
dc.publisherarXiv
dc.relation.urlhttp://arxiv.org/abs/1805.08846v1
dc.relation.urlhttp://arxiv.org/pdf/1805.08846v1
dc.rightsArchived with thanks to arXiv
dc.titleCUDACLAW: A high-performance programmable GPU framework for the solution of hyperbolic PDEs
dc.typePreprint
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentKAUST Supercomputing Laboratory (KSL)
dc.contributor.departmentApplied Mathematics and Computational Science Program
dc.eprint.versionPre-print
dc.contributor.institutionAmerican University of Beirut (AUB)
dc.identifier.arxivid1805.08846
kaust.personAhmadia, Aron
kaust.personKetcheson, David I.
dc.versionv1
refterms.dateFOA2019-04-29T06:27:47Z


Files in this item

Thumbnail
Name:
1805.08846v1.pdf
Size:
1.513Mb
Format:
PDF
Description:
Preprint

This item appears in the following Collection(s)

Show simple item record