On new definitions of SI base units. Why is the "atomic" kilogram preferable

We discuss the role of fundamental constants and measurement data for the Planck, Avogadro and Boltzmann constants and the elementary electric charge in connection with the planned transition to new definitions of four base SI units (the kilogram, mole, ampere and kelvin) in terms of fixed values of these constants. It is proposed to choose a new definition of any base SI unit in terms of a particular fundamental physical constant using a number of criteria, or principles, such as succession relative to the current SI, a sufficient stability of the new unit standards, and concordance between physical dimensions of the unit and the corresponding fundamental constant. It is argued that a redefinition of the kilogram and mole by fixing the values of the atomic mass unit and the Avogadro constant satisfies all these criteria and bears some more advantages against the version with fixed Planck constant: a well founded approach to definition of the ampere and the opportunity to preserve the current relationship between definitions of the mole and the kilogram. It is also argued that the kelvin can be redefined independently of the other three units.