FJP-based query execution

[incubos]

We can speed up heavy queries by throwing cores at parallel ForkJoinPool-based filtering and opening sorting Iterator in IndexedDatabases or we can distribute segments between threads to help data locality.