Dust is present in almost every astrophysical environment, ranging from circumstellar shells and disks to spiral, elliptical, starburst, and active galaxies, and to pre-galactic objects such as QSO absorption-line and damped Ly systems. The abundance and composition of the dust in galaxies affect the galaxies' spectral appearance, and influence the determination of their underlying physical properties, such as their star formation rate, metallicity, and attenuation properties. Understanding the properties of interstellar dust particles is therefore essential for the interpretation of galactic spectra.
In this manuscript I will briefly review what constitutes an interstellar dust model, list the observational constraints on such models, and briefly describe viable interstellar dust models that satisfy these constraints in the local interstellar medium (ISM). Special emphasis will be placed on the interstellar abundance constraints, which until recently, have not been explicitly included in dust models.
Interstellar dust exhibits spatial and temporal variations, and I will briefly review the ingredients in constructing models for the evolution of dust, stressing the current uncertainties in the yield of dust from supernovae and AGB stars. Finally, I will describe the effect of dust evolution on the spectral energy distribution of galaxies, and, using a very simple criterion, present a simple estimate of the redshift when galaxies first become opaque.
More detailed information on observational aspects of interstellar dust and the physics of dust can be found in the recent review article by Draine , in the workshop on "Solid Interstellar matter: The ISO Revolution"  and the conference on "Astrophysics of Dust" , in the books by Whittet  and Krügel , and the recent issue of The Astrophysical Journal Supplement Series (volume 154) dedicated to the first observations with the Spitzer Space Telescope.