DD-AMG on QPACE 3
arXiv:1710.07041 · doi:10.1051/epjconf/201817502007
Abstract
We describe our experience porting the Regensburg implementation of the DD-AMG 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 the performance of the code on a single processor as well as the scaling on many nodes, where in both cases the speedup factor is close to the theoretical expectations.
12 pages, 6 figures, Proceedings of Lattice 2017
Cited by in corpus (11)
- Light-cone distribution amplitudes of pseudoscalar mesons from lattice QCD
- Masses and decay constants of the and mesons from lattice QCD
- Scale setting and the light baryon spectrum in QCD with Wilson fermions
- Octet baryon isovector charges from lattice QCD
- Leading order mesonic and baryonic SU(3) low energy constants from lattice QCD
- Mellin moments of spin dependent and independent PDFs of the pion and rho meson
- Lattice results for the longitudinal spin structure and color forces on quarks in a nucleon
- Double parton distributions in the nucleon from lattice QCD
- Baryonic and mesonic 3-point functions with open spin indices
- Object-oriented implementation of algebraic multi-grid solver for lattice QCD on SIMD architectures and GPU clusters
- Loss-function learning for digital tissue deconvolution