QPACE 2 and Domain Decomposition on the Intel Xeon Phi

نویسندگان

  • Paul Arts
  • Jacques Bloch
  • Peter Georg
  • Benjamin Glaessle
  • Simon Heybrock
  • Yu Komatsubara
  • Robert Lohmayer
  • Simon Mages
  • Bernhard Mendl
  • Nils Meyer
  • Alessio Parcianello
  • Dirk Pleiter
  • Florian Rappl
  • Mauro Rossi
  • Stefan Solbrig
  • Giampietro Tecchiolli
  • Tilo Wettig
  • Gianpaolo Zanier
چکیده

Paul Artsa, Jacques Blochb, Peter Georgb, Benjamin Glässleb, Simon Heybrockb, Yu Komatsubarac, Robert Lohmayerb, Simon Magesb, Bernhard Mendlb, Nils Meyerb, Alessio Parcianelloa, Dirk Pleiterb,d , Florian Rapplb, Mauro Rossia, Stefan Solbrigb, Giampietro Tecchiollia, Tilo Wettig†b, Gianpaolo Zaniera aEurotech HPC, Via F. Solari 3/A, 33020 Amaro, Italy bDepartment of Physics, University of Regensburg, 93040 Regensburg, Germany cAdvanet Inc., 616-4 Tanaka, Kita-ku, Okayama 700-0951, Japan dJSC, Jülich Research Centre, 52425 Jülich, Germany E-mail: [email protected]

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

DD-αAMG on QPACE 3

We describe our experience porting the Regensburg implementation of the DD-αAMG solver from QPACE 2 to QPACE 3. We first review how the code was ported from the first generation Intel Xeon Phi processor (Knights Corner) to its successor (Knights Landing). We then describe the modifications in the communication library necessitated by the switch from InfiniBand to Omni-Path. Finally, we present ...

متن کامل

DD-$\alpha$AMG on QPACE 3

We describe our experience porting the Regensburg implementation of the DD-αAMG solver from QPACE 2 to QPACE 3. We first review how the code was ported from the first generation Intel Xeon Phi processor (Knights Corner) to its successor (Knights Landing). We then describe the modifications in the communication library necessitated by the switch from InfiniBand to Omni-Path. Finally, we present ...

متن کامل

Lattice QCD with Domain Decomposition on Intel

The gap between the cost of moving data and the cost of computing continues to grow, making it ever harder to design iterative solvers on extreme-scale architectures. This problem can be alleviated by alternative algorithms that reduce the amount of data movement. We investigate this in the context of Lattice Quantum Chromodynamics and implement such an alternative solver algorithm, based on do...

متن کامل

A Technology of 3D Elastic Wave Propagation Simulation Using Hybrid Supercomputers

We present a technology of 3D seismic field simulation for high-performance computing systems with GPUs or Intel Xeon Phi coprocessors. This technology covers adaptation of a mathematical modeling method and development of a parallel algorithm. We describe the parallel realization designed for simulation based on using staggeredgrids and 3D domain decomposition method. We study the parallel alg...

متن کامل

Scalability Improvement of the Projected Conjugate Gradient Method used in FETI Domain Decomposition Algorithms

This report summarizes the results of the scalability improvements of the algorithms used in Total FETI (TFETI). A performance evaluation of two new techniques is presented in this report: (1) a novel pipelined implementation of CG method in PETSc and (2) a MAGMA LU solver running on following many-cores accelerators: GPU Nvidia Tesla K20m and Intel MIC Xeon Phi 5110P.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • CoRR

دوره abs/1502.04025  شماره 

صفحات  -

تاریخ انتشار 2015