Any precision tests of the elementary particles are carried out on the
assumption that the fine-structure constant alpha is sufficiently known.
Currently, there are two methods that provide equally accurate values
of the alpha. One is the electron g-2 measurement together with its
theoretical prediction from the QED theory. The other is the quotient
of the Planck constant and the atomic mass (h/M) determined by using
an atomic interferometer. I will report recent progress in both
determination of the alpha including our own work on the five-loop
QED calculation of the electron g-2.