Finite‐Temperature Lanczos for Anisotropic Spin Systems Using Triple‐Hybrid High‐Performance Computing

ABSTRACT

Lanczos concepts are often realized on supercomputers. In view of modern architectures of high‐performance computing triple‐hybrid schemes employing MPI, CPU‐openMP as well as GPU‐openMP should be utilized. The present article sketches how such a scheme could be utilized in order to meet the demands of the finite‐temperature Lanczos method for anisotropic spin systems.