Locally recursive non-locally asynchronous algorithms for stencil computation

DOI10.1134/S1995080218040108MaRDI QIDQ1662781zbMATH OpenOpenAlexWikidataFDO

Authors V. D Levchenko, A. Y. Perepelkina

Publication date 20 August 2018

Published in Lobachevskii Journal of Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1134/s1995080218040108

zbMATH Keywords

cache blockingcache-obliviousdependence graphlocally recursive non-locally asynchronous algorihtmsmemory hierarchyparallel efficiencypolyhedral optimizationroofline modelstencil computationtemporal tiling

Mathematics Subject Classification ID

Parallel numerical computation (65Y05)

Recommendations

• Algorithm 942
• Optimization and Performance Modeling of Stencil Computations on Modern Microprocessors
• Multicore-optimized wavefront diamond blocking for optimizing stencil updates
• scientific article; zbMATH DE number 1747026
• Parallelizable adjoint stencil computations using transposed forward-mode algorithmic differentiation

Cites work

• scientific article; zbMATH DE number 766286 (Why is no real title available?)
• Multicore-optimized wavefront diamond blocking for optimizing stencil updates
• Some efficient solutions to the affine scheduling problem. I: One- dimensional time
• The parallel execution of DO loops

Cited in

(2)

• Multicore-optimized wavefront diamond blocking for optimizing stencil updates
• Heterogeneous LBM Simulation Code with LRnLA Algorithms

This page was built for publication: Locally recursive non-locally asynchronous algorithms for stencil computation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1662781)