In a rotating frame, the artificial gravity has a peculiar form. At the center, this "gravity" is zero, while it's increasing in proportion to $R^2$, where $R$ is the distance to the center.
Now imagine we find ourselves on the inner side of a huge rotating cylinder, on the inner side of which are villages, an infrastructure, flora, fauna, etc. Imagine a person throws a ball in such a way that the velocity of the ball is exactly opposite to the rotation velocity of the cylinder (or, more accurate, opposite to the rotation velocity of the person's hand when he starts the throwing). The ball is then standing still in the rest frame for a person who's not in the rotating cylinder frame. So a person in the rotating frame (of course standing on the ground, the inner side of the cylinder) sees the ball moving in a circle, (with a radius $R$ being about the same as the radius $R$ of the person's head) just a little above the cylinder's inner side.
Another effect would show itself if you would let loose a ball from a tower so the ball goes in free fall. The ball doesn't end up at a place exactly under the point where you let it fall. Off course, it ends up at the bottom (the inner side of the cylinder) but farther away from the tower than when it started to fall. And how strange a game of baseball would look like! You can even wonder if they can find the equations governing gravity, but let's assume they know.
Now to come to the question: How would persons inside the cylinder think the mass outside the cylinder is distributed to cause the gravitational behavior (just like a person in an accelerated lift thinks the gravity he feels is caused by a mass outside the lift)?
Will it be a rotating ring of mass (which is consistent with the circular motion of the ball, described above, because in that case, the ball is standing still with respect to the rotating ring of mass and the gravitational on a mass inside a spherical shell of mass is zero)?
For the rotating bucket with water, you can say the rising water is caused by a massive enough rotating sphere around the bucket (so the rising is caused by frame dragging), as well that the rising is caused by the rotation of the bucket. Both cases are equivalent. This is not the same as the example above, but related though.