Multilevel Parallelization: Grid Methods for Solving Direct and Inverse Problems

Sofya Titarenko, Igor Kulikov, Igor Chernykh, Maxim A. Shishlenin, Olga Krivorot'ko, Dmitry A. Voronov, Mark Hildyard

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Citations (Scopus)


In this paper we present grid methods which we have developed for solving direct and inverse problems, and their realization with different levels of optimization. We have focused on solving systems of hyperbolic equations using finite difference and finite volume numerical methods on multicore architectures. Several levels of parallelism have been applied: geometric decomposition of the calculative domain, workload distribution over threads within OpenMP directives, and vectorization. The run-time efficiency of these methods has been investigated. These developments have been tested using the astrophysics code AstroPhi on a hybrid cluster Polytechnic RSC PetaStream (consisting of Intel Xeon Phi accelerators) and a geophysics (seismic wave) code on an Intel Core i7-3930K multicore processor. We present the results of the calculations and study MPI run-time energy efficiency.
Original languageEnglish
Title of host publicationSupercomputing
Subtitle of host publicationSecond Russian Supercomputing Days, RuSCDays 2016, Moscow, Russia, September 26–27, 2016, Revised Selected Papers
EditorsVladimir Voevodin, Sergey Sobolev
Place of PublicationCham
PublisherSpringer International Publishing AG
Number of pages14
ISBN (Electronic)9783319556697
ISBN (Print)9783319556680
Publication statusPublished - 12 Mar 2017
Externally publishedYes
Event2nd Russian Supercomputing Days 2016 - Moscow, Russian Federation
Duration: 26 Sep 201627 Sep 2016
Conference number: 2

Publication series

NameCommunications in Computer and Information Science
PublisherSpringer International Publishing AG
ISSN (Print)1865-0929


Conference2nd Russian Supercomputing Days 2016
Abbreviated titleRuSCDays 2016
Country/TerritoryRussian Federation


Dive into the research topics of 'Multilevel Parallelization: Grid Methods for Solving Direct and Inverse Problems'. Together they form a unique fingerprint.

Cite this