guest ::
| Home > Publications database > Towards Automated Load Balancing via Spectrum Slicing for FEAST-like solvers > print |
| Home > Publications database > Towards Automated Load Balancing via Spectrum Slicing for FEAST-like solvers > print |
| 001 | 825383 | ||
| 005 | 20221109161714.0 | ||
| 037 | _ | _ | |a FZJ-2016-07846 |
| 041 | _ | _ | |a English |
| 100 | 1 | _ | |a Di Napoli, Edoardo |0 P:(DE-Juel1)144723 |b 0 |u fzj |
| 111 | 2 | _ | |a Joint Laboratory for Extreme Scale Computing |g JLESC |c Kobe |d 2016-11-30 - 2016-12-02 |w Japan |
| 245 | _ | _ | |a Towards Automated Load Balancing via Spectrum Slicing for FEAST-like solvers |
| 260 | _ | _ | |c 2016 |
| 336 | 7 | _ | |a Conference Paper |0 33 |2 EndNote |
| 336 | 7 | _ | |a Other |2 DataCite |
| 336 | 7 | _ | |a INPROCEEDINGS |2 BibTeX |
| 336 | 7 | _ | |a conferenceObject |2 DRIVER |
| 336 | 7 | _ | |a LECTURE_SPEECH |2 ORCID |
| 336 | 7 | _ | |a Conference Presentation |b conf |m conf |0 PUB:(DE-HGF)6 |s 1482342962_11191 |2 PUB:(DE-HGF) |x After Call |
| 520 | _ | _ | |a Subspace iteration algorithms accelerated by rational filtering, such as FEAST, have recently re-emerged as a research topic in solving for interior eigenvalue problems. FEAST-like solvers are Rayleigh-Ritz solvers with rational filter functions, and as a result require re-orthogonalization on long vectors only in rare cases. Application of the filter functions, the computationally most expensive part, offers three levels of parallelism: 1) multiple spectral slices, 2) multiple linear system solves per slice, and 3) multiple right-hand sides per system solves. While the second and third level of parallelism are currently exploited, the first level is often difficult to efficiently realize.An efficient algorithmic procedure to load-balance multiple independent spectral slices is not yet available. Currently, existing solvers must rely on the user's prior knowledge. An automatic procedure to split a user specific interval into multiple load-balanced slices would greatly improve the state of the art. We outline how, both the algorithmic selection of filter functions and the spectral slices, can be at the center of load-balancing issues. Additionally, we present the tools and heuristics developed in an effort to tackle the problems. |
| 536 | _ | _ | |a 511 - Computational Science and Mathematical Methods (POF3-511) |0 G:(DE-HGF)POF3-511 |c POF3-511 |f POF III |x 0 |
| 536 | _ | _ | |a Simulation and Data Laboratory Quantum Materials (SDLQM) (SDLQM) |0 G:(DE-Juel1)SDLQM |c SDLQM |f Simulation and Data Laboratory Quantum Materials (SDLQM) |x 2 |
| 700 | 1 | _ | |a Winkelmann, Jan |0 P:(DE-Juel1)167415 |b 1 |e Corresponding author |u fzj |
| 909 | C | O | |o oai:juser.fz-juelich.de:825383 |p VDB |
| 910 | 1 | _ | |a Forschungszentrum Jülich |0 I:(DE-588b)5008462-8 |k FZJ |b 0 |6 P:(DE-Juel1)144723 |
| 910 | 1 | _ | |a Forschungszentrum Jülich |0 I:(DE-588b)5008462-8 |k FZJ |b 1 |6 P:(DE-Juel1)167415 |
| 913 | 1 | _ | |a DE-HGF |b Key Technologies |1 G:(DE-HGF)POF3-510 |0 G:(DE-HGF)POF3-511 |2 G:(DE-HGF)POF3-500 |v Computational Science and Mathematical Methods |x 0 |4 G:(DE-HGF)POF |3 G:(DE-HGF)POF3 |l Supercomputing & Big Data |
| 914 | 1 | _ | |y 2016 |
| 915 | _ | _ | |a No Authors Fulltext |0 StatID:(DE-HGF)0550 |2 StatID |
| 920 | 1 | _ | |0 I:(DE-Juel1)JSC-20090406 |k JSC |l Jülich Supercomputing Center |x 0 |
| 920 | 1 | _ | |0 I:(DE-82)080012_20140620 |k JARA-HPC |l JARA - HPC |x 1 |
| 980 | _ | _ | |a conf |
| 980 | _ | _ | |a VDB |
| 980 | _ | _ | |a UNRESTRICTED |
| 980 | _ | _ | |a I:(DE-Juel1)JSC-20090406 |
| 980 | _ | _ | |a I:(DE-82)080012_20140620 |
| Library | Collection | CLSMajor | CLSMinor | Language | Author |
|---|