Ashish Jha

MILC staggered conjugate gradient performance on Intel KNL

We review our work done to optimize the staggered conjugate gradient (CG) algorithm in the MILC code for use with the Intel Knights Landing (KNL) architecture. KNL is the second gener- ation Intel Xeon Phi processor. It is capable of massive thread parallelism, data parallelism, and high on-board memory bandwidth and is being adopted in supercomputing centers for scientific research. The CG solver consumes the majority of time in production running, so we have spent most of our effort on it. We compare performance of an MPI+OpenMP baseline version of the MILC code with a version incorporating the QPhiX staggered CG solver, for both one-node and multi-node runs.

Numerical Weather Prediction Optimization

This chapter describes the efforts to improve performance of the widely used WSM6 microphysics package used in numerical weather prediction via a variety of techniques including threading, vectorization, array alignment, improving data locality, and optional use of compile-time constants for loop and array index bounds. Code examples are included to illustrate each technique along with its performance benefits on Intel Xeon processors and Intel Xeon Phi coprocessors. Use of tools such as the Intel Thread Inspector to speed the performance tuning process are also described.