TY  - CONF
AU  - Di Napoli, Edoardo
AU  - Winkelmann, Jan
TI  - Towards Automated Load Balancing via Spectrum Slicing for FEAST-like solvers
M1  - FZJ-2016-07846
PY  - 2016
AB  - 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.
T2  - Joint Laboratory for Extreme Scale Computing
CY  - 30 Nov 2016 - 2 Dec 2016, Kobe (Japan)
Y2  - 30 Nov 2016 - 2 Dec 2016
M2  - Kobe, Japan
LB  - PUB:(DE-HGF)6
UR  - https://juser.fz-juelich.de/record/825383
ER  -