===firstname:
Masha
===firstname3:
Yiming
===affil6:

===lastname3:
Bu
===email:
masha.sosonkina@acm.org
===keyword_other2:

===lastname6:

===affil5:

===lastname4:

===lastname7:

===affil7:

===postal:
Old Dominion University 
1300 ECSB
4700 Elkhorn Ave
Norfolk VA 23529
===ABSTRACT:
In this talk, we consider a recently proposed variable-block variant VBARMS of the ARMS preconditioner for solving general nonsymmetric linear systems. This preconditioner can detect automatically exact or approximate dense structures in a matrix to improve reliability and throughput during factorization. We show how offloading parts of VBARMS to Xeon Phi
accelerators may speed up the preconditioner construction by revealing a high-degree of the parallelism. We compare the resulting performance and convergence with a parallel VBARMS and distributed Schur-Complement preconditioner on large turbulent-flow test cases.
===affil3:
Institute of Mathematics and Computing Science - University of Groningen
===title:
Multilevel Variable-Block based preconditioning offloaded to Xeon Phi accelerators
===affil2:
Institute of Mathematics and Computing Science - University of Groningen
===lastname2:
Carpentieri
===firstname4:

===keyword1:
Iterative Linear Algebraic Data Mining Techniques
===workshop:
no
===lastname:
Sosonkina
===firstname5:

===keyword2:
Iterative solvers/linear algebra on high concurrency node architectures
===otherauths:

===affil4:

===competition:
no
===firstname7:

===firstname6:

===keyword_other1:

===lastname5:

===affilother:

===firstname2:
Bruno