Cannibals are dark matter particles with a scattering process that allows three particles to annihilate to two. This exothermic process keeps the gas of the remaining particles warm long after they become non-relativistic. A cannibalizing dark sector which is decoupled from the Standard Model naturally arises from a pure-glue confining hidden sector. It has an effective field theory description with a single massive interacting real scalar field, the lightest glueball. Since warm dark matter strongly suppresses growth of structure cannibals cannot be all of the dark matter. Thus we propose a scenario where most dark matter is non-interacting and cold but about 1 percent is cannibalistic. We review the cannibals' unusual scaling of the temperature and energy and number densities with redshift and generalize the equations for the growth of matter density perturbations to the case of cannibals. We solve the equations numerically to predict the scaling of the Hubble parameter and the characteristic shape of the linear matter power spectrum as a function of model parameters. Our results may have implications for the $\sigma_8$ and $H_0$ problems.

We consider a recently proposed model in which dark matter interacts with a thermal background of dark radiation. Dark radiation consists of relativistic degrees of freedom which allow larger values of the expansion rate of the universe today to be consistent with CMB data ($H_0$-problem). Scattering between dark matter and radiation suppresses the matter power spectrum at small scales and can explain the apparent discrepancies between $\Lambda$CDM predictions of the matter power spectrum and direct measurements of Large Scale Structure LSS ($\sigma_8$-problem). We go beyond previous work in two ways: 1. we enlarge the parameter space of our previous model and allow for an arbitrary fraction of the dark matter to be interacting and 2. we update the data sets used in our fits, most importantly we include LSS data with full $k$-dependence to explore the sensitivity of current data to the shape of the matter power spectrum. We find that LSS data prefer models with overall suppressed matter clustering due to dark matter - dark radiation interactions over $\Lambda$CDM at 3-4 $\sigma$. However recent weak lensing measurements of the power spectrum are not yet precise enough to clearly distinguish two limits of the model with different predicted shapes for the linear matter power spectrum. In two Appendices we give a derivation of the coupled dark matter and dark radiation perturbation equations from the Boltzmann equation in order to clarify a confusion in the recent literature, and we derive analytic approximations to the solutions of the perturbation equations in the two physically interesting limits of all dark matter weakly interacting or a small fraction of dark matter strongly interacting.

We study a two-parameter extension of the cosmological standard model $\Lambda$CDM in which cold dark matter interacts with a new form of dark radiation. The two parameters correspond to the energy density in the dark radiation fluid $\Delta N_\mathrm{fluid}$ and the interaction strength between dark matter and dark radiation. The interactions give rise to a very weak "dark matter drag" which damps the growth of matter density perturbations throughout radiation domination, allowing to reconcile the tension between predictions of large scale structure from the CMB and direct measurements of $\sigma_8$. We perform a precision fit to Planck CMB data, BAO, large scale structure, and direct measurements of the expansion rate of the universe today. Our model lowers the $\chi$-squared relative to $\Lambda$CDM by about 12, corresponding to a preference for non-zero dark matter drag by more than $3 \sigma$. Particle physics models which naturally produce a dark matter drag of the required form include the recently proposed non-Abelian dark matter model in which the dark radiation corresponds to massless dark gluons.

We propose a new class of dark matter models with unusual phenomenology. What is ordinary about our models is that dark matter particles are WIMPs, they are weakly coupled to the Standard Model and have weak scale masses. What is unusual is that they come in multiplets of a new "dark" non-Abelian gauge group with milli-weak coupling. The massless dark gluons of this dark gauge group contribute to the energy density of the universe as a form of weakly self-interacting dark radiation. In this paper we explore the consequences of having i.) dark matter in multiplets ii.) self-interacting dark radiation and iii.) dark matter which is weakly coupled to dark radiation. We find that i.) dark matter cross sections are modified by multiplicity factors which have significant consequences for collider searches and indirect detection, ii.) dark gluons have thermal abundances which affect the CMB as dark radiation. Unlike additional massless neutrino species the dark gluons are interacting and have vanishing viscosity and iii.) the coupling of dark radiation to dark matter represents a new mechanism for damping the large scale structure power spectrum. A combination of additional radiation and slightly damped structure is interesting because it can remove tensions between global $\Lambda$CDM fits from the CMB and direct measurements of the Hubble expansion rate ($H_0$) and large scale structure ($\sigma_8$).