The emergence of the field of nanofluidics in the last decade has led to the development of important applications including water desalination, ultrafiltration and osmotic energy conversion. Most applications make use of carbon nanotubes, boron nitride nanotubes, graphene and graphene oxide. In particular, understanding water transport in carbon nanotubes is key for designing ultrafiltration devices and energy-efficient water filters. However, although theoretical studies based on molecular dynamics simulations have revealed many mechanistic features of water transport at the molecular level, further advances in this direction are limited by the fact that the lowest flow velocities accessible by simulations are orders of magnitude higher than those measured experimentally. Here, we extend molecular dynamics studies of water transport through carbon nanotubes to flow velocities comparable with experimental ones using massive crowd-sourced computing power. We observe previously undetected oscillations in the friction force between water and carbon nanotubes and show that these oscillations result from the coupling between confined water molecules and the longitudinal phonon modes of the nanotube. This coupling can enhance the diffusion of confined water by more than 300%. Our results may serve as a theoretical framework for the design of new devices for more efficient water filtration and osmotic energy conversion devices.