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 domain decomposition, on Intel R © Xeon Phi co-processor (KNC) clusters. We demonstrate close-to-linear on-chip scaling to all 60 cores of the KNC. With a mix of singleand half-precision the domain-decomposition method sustains 400-500 Gflop/s per chip. Compared to an optimized KNC implementation of a standard solver [1], our full multi-node domain-decomposition solver strong-scales to more nodes and reduces the time-to-solution by a factor of 5. Keywords—Domain decomposition, Intel R © Xeon Phi coprocessor, Lattice QCD Categories and subject descriptors: D.3.4 [Programming Languages]: Processors – Optimization; G.1.3 [Numerical Analysis]: Numerical Linear Algebra – Sparse, structured, and very large systems (direct and iterative methods); G.4 [Mathematical Software]: Algorithm design and analysis, Efficiency, Parallel and vector implementations; J.2 [Physical Sciences and Engineering]: Physics General Terms: Algorithms, Performance