Xeon Phi Acceleration of Domain Decomposition Iterations via Heterogeneous Active Messages

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27120%2F18%3A10239485" target="_blank" >RIV/61989100:27120/18:10239485 - isvavai.cz</a>
Alternative codes found
RIV/61989100:27240/18:10239485 RIV/61989100:27730/18:10239485 RIV/61989100:27740/18:10239485
Result on the web
<a href="https://aip.scitation.org/doi/abs/10.1063/1.5043963" target="_blank" >https://aip.scitation.org/doi/abs/10.1063/1.5043963</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1063/1.5043963" target="_blank" >10.1063/1.5043963</a>

Result language
angličtina
Original language name
Xeon Phi Acceleration of Domain Decomposition Iterations via Heterogeneous Active Messages
Original language description
We present an acceleration strategy for a domain-decomposition iterative solver based on local Schur complements with respect to the skeleton of the computational domain. In finite element tearing and interconnecting (FETI) this results in an iterative application of a large number of dense matrices. For the offload of such kernels to the Intel (R) Xeon Phi (TM) coprocessors we use the Heterogeneous Active Messages (HAM) library. A simple load balancing strategy is presented to efficiently utilize both the host CPU and the available coprocessors during individual iterations.
Czech name
—
Czech description
—

Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)