The don't have index permutations spo, ops, pos, etc. like typically used in triple stores. The reason is quite obvious, the main goal was to have a very compressed binary format. You can see the subject driven index scheme here [1]. Benchmarking results and discussion shows that for patterns VAR URI VAR and VAR URI URI other triple stores resp. indexing/serialization formats perform better. Anyways, HDT discussion is out of scope for this mailing list, thus, I'd avoid to go into details.
To bring this thread to an end, I guess we finally answered your question? Or are the any open issues? Maybe you can sum up a bit, it's a pretty big thread right now. Lorenz [1] http://www.rdfhdt.org/hdt-internals/#triples On 17.12.2017 18:29, Laura Morales wrote: >> Regarding performance, from what I understood while reading the HDT >> paper, the worst pattern is VAR URI URI which makes sense as triples are >> ordered by subject in BitmapTriples. > The have other indexes in .hdt.index
