How does dark matter suppress baryonic osculations?
The dark matter density in the early universe has a significant effect on the peaks in the CMB power spectrum, which strongly affects structure formation in the early universe, e.g., SDSS [see; https://arxiv.org/abs/1509.06529, SDSS-III Baryon Oscillation Spectroscopic Survey Data Release 12: galaxy target selection and large scale structure catalogues. You need a pretty healthy dose of DM to reconcile the observed structure with that expected based solely on baryonic matter and radiation pressure in the early universe [DM is not affected by radiation pressure]. For discussion, see http://background.uchicago.edu/~whu/intermediate/driving2.html. It is difficult to appreciate the strength of the case for DM without familiarity with CMB and SDSS studies.
I don't think it suppresses them. It just doesn't participate in them (as dark matter doesn't experience pressure to any significant degree). In practical terms this means that dark matter contributes to all of the odd peaks in the CMB power spectrum, while only the normal matter contributes to the even peaks.
But it does, doesn't it? - it suppresses the even peaks as it acts gravitationally against the bouncing baryonic matter so that it doesn't expand as much as a result.
This animation from Max Tegmark's site:
shows the effect each cosmological parameter should have on peak distribution.
@wolram Sean Carroll briefly explains how this works in this video:
starting at around 15:40 (or, -9.45).
As Wayne Hu explains in http://background.uchicago.edu/~whu/intermediate/driving2.html, a dark matter excess in the early universe suppresses peaks in the CMB power spectrum.
Separate names with a comma.