Looks very interesting, but seems to serve a pretty different use case: > This is an ordered data structure, and supports very high throughput iteration over l

Looks very interesting, but seems to serve a pretty different use case: > This is an ordered data structure, and supports very high throughput iteration over lexicographically sorted ranges of values. If you are looking for simple point operation performance, you may find a better option among one of the many concurrent hashmap implementations that are floating around. Pay for what you actually use :)