Skiplists can be very effective in disk structures for few of reasons.
For one, a sorted sequence of keys can be written to file in a single pass, and it's easy to interleave with the data itself. For example, Lucene uses multi-level skip lists in compressed posting list files in order to quickly jump to a term and then jump to the lowest matching document ID for that term. Since these files are immutable, the tree is only built once and all the data can be stored in sorted order, which has the added benefit that these files can also be merged in a single pass.
Append-only is a very convenient corner case for skiplists to optimize. I call it "skiplogs" because it looks really different from the general case. An average node can be 3 bytes, for example.
For one, a sorted sequence of keys can be written to file in a single pass, and it's easy to interleave with the data itself. For example, Lucene uses multi-level skip lists in compressed posting list files in order to quickly jump to a term and then jump to the lowest matching document ID for that term. Since these files are immutable, the tree is only built once and all the data can be stored in sorted order, which has the added benefit that these files can also be merged in a single pass.