Answer from cs61c-aj (Kevin Li 15855812) for Question 2
The advantage of having fixed-szie chunks is that boundary between page numbers and offset are unknown to the compiler and the programmer, while segmentation which uses variable size chunks splits address into separate pieces that must be calculated using the segment number and offset and also a bound check must be included to know if the offset is within the segment. Bigger chunks allow high access time to go up because like cache bigger blocks make miss rate will be reduce because of spatial locality. Smaller chunks is good for temporal locality because a lot more chunks can be placed in the VM and also access time will go up.