The full-potential linearized augmented-plane wave (EP-LAPW) method is well
known to enable most accurate calculations of the electronic structure and
magnetic properties of crystals and surfaces. The implementation of atomic
forces has greatly increased its applicability, but it is still generally
believed that FP-LAPW calculations require substantial higher computational
effort compared to the pseudopotential plane wave (PPW) based methods. In
the present paper we analyze the FP-LAPW method from a computational point
of view. Starting from an existing implementation (WIEN95 code), we identif
ied the time consuming parts and show how some of them can be formulated mo
re efficiently.
In this context also the hardware architecture plays a crucial role. The re
maining computational effort is mainly determined by the setup and diagonal
ization of the Hamiltonian matrix. For the latter. two different iterative
schemes are compared. The speed-up gained by these optimizations is compare
d to the runtime of the "original" Version of the code, and the PPW approac
h. We expect that the strategies described here, can also be used to speed
up other computer codes, where similar tasks must be performed. (C) 2000 El
sevier Science B.V. All rights reserved.